Encyclopedia of
Human
Behavior
Second Edition
Volume One A-D
Encyclopedia of
HUMAN BEHAVIOR
SECOND EDITION
Volume 1
A-D
EDITORIAL ADVISORY BOARD
David M Buss
University of Texas, Austin, Texas, USA
Richard B Buxton
University of California, San Diego, California, USA
Nicholas Christenfeld
University of California, San Diego, California, USA
Orrin Devinsky
New York University, New York, New York, USA
Albert M Galaburda
Harvard Medical School, Boston, Massachusetts, USA
William Hirstein
Elmhurst College, Elmhurst, Illinois, USA
William G Iacono
University of Minnesota, Minneapolis, Minnesota, USA
John T Jost
New York University, New York, New York, USA
Jerome Kagan
Harvard University, Boston, Massachusetts, USA
Leif E O Kennair
Norwegian University of Science and Technology,
Trondheim, Norivay
David E Presti
University of California, Berkeley, California, USA
Encyclopedia of
HUMAN BEHAVIOR
SECOND EDITION
EDITOR-IN-CHIEF
V. S. RAMACHANDRAN
University of California, San Diego, California
Volume 1
A-D
ELSEVIER
AMSTERDAM BOSTON HEIDELBERG LONDON NEW YORK
PARIS SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY
Academic Press is an imprint of Elsevier
OXFORD
TOKYO
ACADEMIC
PRESS
Academic Press is an imprint of Elsevier
32 Jamestown Road, London NW1 7BY, UK
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
Second edition 2012
Copyright © 2012 Elsevier, Inc. All rights reserved.
The following articles are US Government works in the public domain and are not subject to copyright:
Creativity
Glial Cells
Military Psychology
Neuroimaging of Dementia
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior written permission of the publisher.
Permissions may be sought from Elsevier's Science & Technology Rights department in Oxford, UK: phone (+44) (0) 1865 843830;
fax (+44) (0) 1865 853333; email: permissions@elsevier.com
Alternatively you can submit your request online by visiting the Elsevier website at http://elsevier.com/locate/permissions and selecting
Obtaining permission to use Elsevier material.
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Control Number: 2012930155
ISBN: 978-0-12-375000-6
For information on all Elsevier publications
visit our website at elsevierdirect.com
Printed and bound in the United States of America
11 12 13 14 10 9 8 7 6 5 4 3 2 1
Working together to grow
libraries in developing countries
www.elsevier.com | www.bookaid.org | www.sabre.org
Editorial : Mica Haley, Anna Gebicka, Kate Mittell
Production : Justin Taylor
ELSEVIER
BOOK AID
International
Sabre Foundation
ABOUT THE EDITOR-IN-CHIEF
V.S. Ramachandran, MD, PhD., is Director of the Center for Brain and Cognition and professor with the Psychology
Department and the Neurosciences Program at the University of California, San Diego, and Adjunct Professor of
Biology at the Salk Institute. Ramachandran trained as a physician and obtained an MD from Stanley Medical
College and subsequently a PhD from Trinity College at the University of Cambridge, where he was elected a Senior
Rouse Ball Scholar. Ramachandran's early research was on visual perception, but he is best known for his work in
neurology.
He has received many honors and awards, including a fellowship from All Souls College, Oxford, an honorary
doctorate from Connecticut College, a gold medal from the Australian National University, the Ariens Kappers
Medal from the Royal Netherlands Academy of Sciences for landmark contributions in neuroscience, and the
presidential lecture award from the American Academy of Neurology. He is also a fellow of the Neurosciences Institute in La Jolla and a
fellow of the Institute for Advanced Studies in Behavioral Sciences at Stanford. He was invited by the BBC to give the Reith Lectures on 'The
Emerging Mind' in 2003 and is the first physician/ experimental psychologist to be given this honor since the series was begun by Bertrand
Russell in 1 949 - these lectures were subsequently published as A Brief Tour of Human Consciousness : From Impostor Poodles to Purple Numbers.
In 1995, he gave the Decade of the Brain Lecture at the 25th annual (silver jubilee) meeting of the Society for Neuroscience and more
recently, the inaugural keynote lecture at the Decade of the Brain conference held by NIMH at the Library of Congress and a public lecture at
the Getty Museum in Los Angeles. He also gave the first Hans Lucas Teuber lecture at MIT, the D.O. Hebb lecture at McGill, the Rudel-Moses
lecture at Columbia, the Dorcas Cumming (inaugural keynote) lecture at Cold Spring Harbor, the Raymond Adams neurology grand
rounds at Massachusetts General Hospital, Harvard, and the Jonas Salk memorial lecture, Salk Institute.
Ramachandran is a trustee for the San Diego Museum of Art and has lectured widely on art, visual perception, and the brain.
Ramachandran has published over 120 papers in scientific journals (including three invited review articles in Scientific American), coauthor
(with Sandra Blakesleee) of Phantoms in the Brain that has been translated into eight languages and formed the basis for a two-part series on
Channel Four TV in the UK and a one-hour PBS special in the United States. His work is featured frequently in the major news media
including BBC and PBS. Newsweek magazine recently named him a member of The Century Club,' one of the "hundred most prominent
people to watch in the next century. "
V
This page intentionally left blank
ABOUT THE EDITORIAL ADVISORY BOARD
David M. Buss received his BA from the University of Texas at Austin and his PhD from the
University of California at Berkeley. He served in professorial positions at Harvard University, the
University of Michigan, and the University of Texas, where he is currently professor of psychology.
David Buss received the American Psychological Association (APA) Distinguished Scientific Award
for Early Career Contribution to Psychology, the APA G. Stanley Hall Award, and the APA Distin-
guished Scientist Lecturer Award. The University of Texas awarded Buss the President's Associates
Teaching Excellence Award. He served as President of the Human Behavior and Evolution Society
(HBES). He is currently Head of the Individual Differences and Evolutionary Psychology Area of the
Psychology Department at the University of Texas at Austin.
Buss's books include The Evolution of Desire: Strategies of Hitman Mating (Basic Books; translated
into 11 languages); Evolutionary Psychology: The Neiv Science of the Mind (Allyn & Bacon); The
Dangerous Passion: Why Jealousy is as Necessary as Love and Sex (Free Press; translated into 13 languages);
The Murderer Next Door: Why the Mind is Designed to Kill (Penguin; translated into 12 languages); The
Handbook of Evolutionary Psychology (Wiley), for which he served as editor; W?ty Women Have Sex (Holt;
co-authored with Cindy Meston; translated into 16 languages); and The Evolution of Personality and
Individual Differences (Oxford University Press). Buss has more than 250 scientific publications to his
credit.
Buss has extensive cross-cultural research collaborations and lectures widely within the United
States and abroad. His primary research interests include the psychology of sex differences, human
sexuality, mating strategies, conflict between the sexes, why people kill, warfare, terrorism, stalking, and the psychology of prestige, status,
and reputation.
Richard B. Buxton received his BS (1976) and PhD (1981) degrees in physics from the
Massachusetts Institute of Technology, followed by a postdoc in biomedical imaging (PET and
MRI) at the Massachusetts General Hospital. He has been on the faculty of the Department of
Radiology at the University of California, San Diego, since 1990. Buxton's primary research is in
functional magnefic resonance imaging (fMRI), focused on understanding the connections between
neural activity, blood flow, and energy metabolism in the human brain. His experimental work
combines arterial spin labeling (ASL) methods with blood oxygenation-level dependent (BOLD)
methods to estimate changes in brain oxygen metabolism in response to a stimulus or a drug. His
more theoretical work involves mathematical modeling of the imaging methods, the BOLD effect,
and oxygen transport from blood to tissue. Buxton is based at the Center for fMRI and served as the
center's founding Director (2000-2007). He has written a textbook on fMRI, now in its second
edition, published by Cambridge University Press (2009).
VII
viii About the Editorial Advisory Board
Nicholas Christenfeld received a bachelor's degree in psychology and social relations from Harvard,
and a doctorate in social psychology from Columbia, and joined the faculty at the University of
California - San Diego and has remained there for two decades. His research program comprises
multiple, changing, only partly overlapping areas. He has addressed such topics as why some people
say 'um' so often, whether babies look like their mothers or their fathers, how we choose which box
of cereal to buy, why a baseball season is ten times as long as a football season, which parts of the
month carry the greatest risk of untimely death, whether the support of a woman is better for one's
blood pressure than that of a man, what sort of music might be useful in stress reduction, whether
some initials extend and some shorten the lives of their bearers, if people who live in, or even just
visit, New York City are at risk of heart attacks, and whether story spoilers do spoil stories. Currently,
Christenfeld is also exploring the role of forgiveness in poststress cardiovascular recovery, the impact
of crying on hormones, and the nature of a sense of humor. An overarching theme, if one exists,
might be the empirical exploration of the everyday phenomena of the world.
Orrin Devinsky is professor of neurology, neurosurgery, and psychiatry at New York University
(NYU). He directs the NYU Epilepsy Center and St. Barnabas Institute of Neurology. His research on
epilepsy includes phenomic-genomic relations, sudden unexpected death in epilepsy, surgical
therapies, new medicines and devices, quality of life, cognition and behavior, and neuroinflamma-
tion. His research on behavioral neurology includes hyperfamiliarity, delusions, autonomic nervous
function, and anterior cingulate cortex functions. He founded the organization Finding A Cure for
Epilepsy and Seizures and cofounded epilepsy.com and the Epilepsy Therapy Project. He serves on
the boards of these organizations and has served on the board of the American Epilepsy Society and
Epilepsy Foundation. His other interests include the history of neuropsychiatry, evolution, anthro-
pology, and animal intelligence.
Albert M. Galaburda is a cognitive neurologist and neuroscientist, specializing in developmental
cognitive disorders in adults and children. He received his medical degree from Boston University in
1971, trained in neurology at Harvard Medical School from 1973 to 1976, and began his research on
»|f“ dyslexia in 1979, when with neuropathologist Thomas Kemper, he reported for the first time minor
cortical malformations in the brain of an adult dyslexic who had died in an accident. Following that
original report, he published several other cases with similar malformations and launched a research
program modeling these malformations in rodent models, which continues to this date.
He uses RNAi and transgenic technologies in his present research to induce cortical malformations
h in rat and mouse brains, which in turn are tested anatomically, molecularly, and behaviorally in
collaboration with Glenn Rosen, Joseph LoTurco, and Holly Fitch, respectively. This latest research
I has shown that manipulation of dyslexia candidate gene homolog's I rodent embryos results in
7 molecular interference with neuronal migration, cell autonomous and noncell autonomous effects
on cortical development that mimic the changes seen in dyslexic brains, and behavioral changes
affecting auditory perception. Ex vivo imaging of the brains of genetically manipulated animals
discloses changes in cortical circuits affecting corticocortical and corticothalamic relationships.
Galaburda is presently the Emily Fisher Landau Professor of Neurology at Harvard Medical
School, and Head of the Cognitive Neurology Unit at Beth Israel Deaconess Medical Center in Boston.
He has published more than 150 scholarly articles on various topics of cognitive anatomy, cognitive
neurology, and dyslexia, as well as several books on related topics. Galaburda has received the Pattison Prize in Neuroscience, the Neuronal
Plasticity Prize from the IPSEN Foundation of France, Scientist of the Year from the Association for Children with Learning Disabilities
(now, Learning Disorders Association), and has been elected to membership in the American Neurological Association. He has delivered
the Rita Rudel Memorial Lecture, the Norman Geschwind Memorial Lecture, the Sylvio Conte Decade of the Brain Symposium Lecture, and
the Curt von Euler Memorial Lecture, among others. His research is funded by the Eunice Kennedy Shrive National Institute of Child Health
& Human Development. His teaching is funded by the National Institute of Neurological Disorders and Stroke.
About the Editorial Advisory Board ix
William Hirstein is Professor and Chair of the Philosophy Department at Elmhurst College,
Elmhurst, IL, USA. He received his PhD from the University of California, Davis, in 1994. He is
the author of several books, including On the Churchlands (Wadsworth Publishing, 2004), Brain
Fiction: Self-Deception and the Riddle of Confabulation (MIT Press, 2005), and Mindmelding : Conscious-
ness, Neuroscience, and the Mind's Privacy (Oxford University Press, 2012). His other interests include
autism, sociopathy, brain laterality, sense of self, and the misidentification syndromes.
William G. Iacono is Regents Professor of Psychology, Psychiatry, Law, and Neuroscience; Distin-
guished McKnight University Professor; Adjunct Professor of Child Development; and Codirector of
the Minnesota Center for Twin & Family Research at the University of Minnesota.
He studies the development of common mental disorders, focusing on the genetic liability for
substance abuse and antisocial behavior. He is past president of the Society for Psychophysiological
Research (SPR) . He has also received the Distinguished Contributions to Psychophysiology Award
from SPR for his lifetime contributions to this field. In addition, he has been awarded the Distin-
guished Scientist Award from the Society for a Science of Clinical Psychology. His research is funded
by grants from the National Institute of Mental Health, National Institute on Drug Abuse, and
National Institute on Alcoholism and Alcohol Abuse. He currently holds a National Institute of
Health MERIT (Method to Extend Research in Time) Award for his research achievement. He has
published over 350 scientific articles.
John T. Jost is Professor of Psychology at New York University. His research, which addresses
stereotyping, prejudice, political ideology, and system justification theory, has appeared in top
scientific journals and received national and international media attention. He has published over
90 articles and chapters and coedited four books: The Psychology of Legitimacy (2001), Perspectivism in
Social Psychology (2004), Political Psychology: Key Readings (2004), and Social and Psychological Bases of
Ideology and System Justification (2011). Jost has received numerous honors, including the Gordon
Allport Intergroup Relations Prize (three times), Erik Erikson Award for Early Career Research
Achievement in Political Psychology, International Society for Self and Identity Early Career
Award, Society for Personality and Social Psychology Theoretical Innovation Prize, Society of
Experimental Social Psychology Career Trajectory Award, and the Morton Deutsch Award for
Distinguished Scholarly and Practical Contributions to Social Justice.
x About the Editorial Advisory Board
Jerome Kagan is Emeritus Professor of Psychology at Harvard University. His research over the past
57 years has centered on the cognitive and emotional development of infants and children. During
the last 30 years, he and his colleagues have concentrated on the temperamental biases called high
and low reactivity in infants. He is a fellow of the American Academy of Arts and Sciences and a
member of the Institute of Medicine in the National Academy of Sciences. He has received distin-
guished scientist awards from the American Psychological Association, the Society for Research in
Child Development, and the Child-Mind Institute. Kagan received his PhD from Yale University in
1954 and taught at the Ohio State University and Antioch College before joining the Harvard faculty
in 1964.
Leif Edward Ottesen Kennair, PhD, is Head of Department at the Department of Psychology,
Norwegian University of Science and Technology, Trondheim, Norway. He graduated as a clinical
psychologist from the University of Bergen, 1999, and is a specialist of clinical adult psychology
(Norwegian Psychology Association), and a qualified cognitive therapist and supervisor of cogni-
tive-behavioral therapy (Norwegian Cognitive Therapy Association). He has worked as a clinical
psychologist/ chief psychologist at the Nordfjord Psychiatric Centre and as a researcher at St. Olav's
Hospital, Trondheim. His research interests span from mainstream clinical psychology efficacy research
(generalized anxiety disorder, OCD) to evolutionary psychology (evolutionary psychopathology, the
evolutionary psychology of war, mate choice and sexual behavior). Kennair's PhD thesis was on the
topic of evolutionary clinical psychology. He has written an introduction to evolutionary psychology,
which has been published in Norwegian and Danish. He lectures on evolutionary psychology,
behavioral genetics, and personality psychology and is a much-used lecturer for training psychologists
and psychiatrists on topics such as cognitive-behavior therapy and treatment of anxiety and OCD.
David E. Presti is a neurobiologist and cognitive scientist at the University of California in Berkeley,
where he has taught in the Department of Molecular and Cell Biology for more than 20 years. For
more than a decade ( 1990-2000) he also worked in the treatment of addiction and of posttraumatic
stress disorder (PTSD) at the Department of Veterans Affairs Medical Center in San Francisco. His
areas of expertise include the chemistry of the human nervous system, the effects of drugs on the
brain and the mind, the treatment of addiction, and the scientific study of the mind and the
consciousness. He has doctorates in molecular biology and biophysics from the California Institute
of Technology and in clinical psychology from the University of Oregon. Since 2004, he has also
been teaching neuroscience to Tibetan monks in India in a program initiated by the Dalai Lama. His
primary research interest is the relation between mentality, consciousness, and brain physiology, the
so-called mind-body problem.
REVIEWERS
We would like to thank the following for their valuable assistance in peer-reviewing the articles within the Encyclopedia of Human Behavior:
Mike Aamodt
maamodt@radford.edu
Donelson Forsyth
dforsyth@richmond.edu
Josh Ackerman
joshack@mit.edu
Willem E. Frankenhuis
wfrankenhuis @ gmail . com
Patrick Ackles
packles@elmhurst.edu
Robert Freidin
freidin@princeton.edu
Robert Allan
allanr@lafayette.edu
Alan Fridlund
fridlund@psych.ucsb.edu
Lisa Aspinwall
lisa.aspinwall@psych.utah.edu
Luciano Gasser
luciano.gasser@phz.ch
Elizabeth Austin
elizabeth.austin@ed.ac.uk
Itimar Gati
msgati@mscc.huji.ac.il
Efrain Azmitia
efrain.azmitia@nyu.edu
David Geary
gearyd@missouri.edu
James Beggan
james.beggan@louisville.edu
William Gerin
bgerinx@gmail.com
Stephen Benning
stephen.d.benning@vanderbilt.edu
Jennifer Gillis
jmg0001@auburn.edu
Anna Berti
annamaria.berti@unito.it
David Cleaves
david.gleaves@canterbury.ac.nz
Prasad Bestha
durgabestha@creighton.edu
Dan Grodner
dgrodnel @swarthmore.edu
David Bjorklund
dbjorklu@fau.edu
Edward Hagen
ehagen@gmail.com
Herbert Bless
bless @uni-mannheim. de
Roger Hagen
roger.hagen@svt.ntnu.no
Jason Bocarro
j ason_bocarro @ncsu.edu
James Hampton
j.a.hampton@city.ac.uk
Robert Bomstein
bornstein@adelphi.edu
Curt Haugtvedt
haugtved @fisher.osu.edu
Jos Bosch
j.a.bosch@bham.ac.uk
Marie Helweg-Larsen
helwegm @ dickinson. edu
Bruce Bridgeman
bruceb@ucsc.edu
Ralph Hood
ralph-hood@utc.edu
Richard Brislin
rbrislin@hawaii.edu
John Hunsley
john.hunsley@uottawa.ca
Lllrike Buhlmann
ulrike.buhlmann@hu-berlin.de
William Ickes
ickes@uta.edu
William Bukowski
william. bukowski@concordia.ca
Nancy Jackson
nancy- j ackson @ uiowa .edu
Josh Burk
jabur2@wm.edu
Rebecca Jackson
bjackson@pgsp.edu
Michael Burton
mlburton@uci.edu
Kali Johnson
klj03@fsu.edu
Shawn Charlton
srcharlton@mac.com
Michael Kavsek
kavsek@uni-bonn.de
Nishimura Chiaki
Susan Kemper
skemper@ku.edu
Jason Clark
jasonanthonyclark@gmail.com
Monica Harris Kern
harris@pop.uky.edu
Taya Cohen
tcohen@cmu.edu
Ashley King
akingl@email.arizona.edu
Michael Cole
mcole@ucsd.edu
Gerjo Kok
g.kok@maastrichtuniversity.nl
Dennis R. Combs
dcombs@uttyler.edu
Nate Kornell
nate.kornell@williams.edu
Karen Cook
kcook@stanford.edu
Megan Kozak
mkozak@pace.edu
Joel Cooper
jcoops@princeton.edu
Howard Krishner
hkirshner@aol.com
Katja Riiter Corcoran
katja.corcoran@uni-koeln.de
Peter Kuppens
peter, kuppens @ psy .kuleuven .be
Alita Cousins
cousinsa@easternct.edu
David Kutzik
dkutzik@aol.com
Arthur Cropley
acropley@dodo.com.au
Peter Lamal
palamal@uncc.edu
Walter Crozier
r.crozier@uea.ac.uk
Richard Lapan
lapan@educ.umass.edu
Martin Daly
daly@mcmaster.ca
Daniel Lapsley
dlapslel@nd.edu
Maayan Davidov
mdavidov@mscc.huji.ac.il
Steve Lee
stevelee@psych.ucla.edu
Hasker Davis
hdavis@uccs.edu
Antoon Leenaars
draalee@sympatico.ca
Susanne Denham
sdenham@gmu.edu
Daniel S. Levine
levine@uta.edu
Brent Donnellan
donnel59@msu.edu
Debra Lieberman
debra@miami.edu
David Dozois
ddozois@uwo.ca
Scott Liening
sliening@mail. utexas . edu
Steve Duck
steve-duck@uiowa.edu
Scott Lilienfeld
slilien@emory.edu
Roger Dunn
rdunn@sunstroke.sdsu.edu
Wolfgang Linden
wlinden@psych.ubc.ca
Paul Eastwick
eastwick@tamu.edu
John Loehlin
loehlin@psy.utexas.edu
David W. Evans
dwevans @bucknell.edu
Brian Lowery
lowery_brian @ gsb . Stanford .edu
Gary W. Evans
gwe 1 @ Cornell . edu
Mark Lumley
mlumley@wayne.edu
Terry Falcomata
falcomata@austin.utexas.edu
Robert Maiden
fmaiden@alfred.edu
Maggie Faulkner
mfaulkner@towson.edu
Greg Maio
maio@cardiff.ac.uk
Anwei Feng
eds808@bangor.ac.uk
Robert Mair
rgm@unh.edu
Gereon Fink
g.r.fmk@fz-juelich.de
Gaurav Mathur
gaurav.mathur@gallaudet.edu
XI
xii Reviewers
Susan Matt
smatt@weber.edu
Nicolas Rothen
beat. meier@psy.unibe. ch
Michael D. Matthews
mike.matthews@usma.edu
Charles Sanislow
csanislow@wesleyan.edu
lim Mcgough
jmcgough@mednet.ucla.edu
David Sarwer
dsarwer@mail.med.upenn.edu
David Mcneill
dmcneill @ uchicago . edu
Geoffrey Schoenbaum
schoenbg@schoenbaumlab.org
Elizabeth Meadows
meadolea@cmich.edu
David Schroeder
dave@uark.edu
Jeff Miller
millerjeff@duq.edu
Donald Shankweiler
donald.shankweiler@uconn.edu
Paul Mills
pmills@ucsd.edu
David Share
dshare@constmct.haifa.ac.il
Ian Mitroff
imitroff@alliant.edu
Katrina L. Sifferd
sifferdk@elmhurst.edu
Evan Murray
edmurray@partners.org
Barry Smith
bdsmith@psyc.umd.edu
lordi Navarra
jordi.navarra@gmail.com
Peter Smith
p.smith@gold.ac.uk
Bernard Nijstad
b.a.nijstad@mg.nl
John Smythies
smythies@psy.ucsd.edu
Kathleen O'Connor
kmo8@cornell.edu
Hugo Spiers
h.spiers@ucl.ac.uk
Christin Ogle
christin.ogle@duke.edu
Jerry Suls
jerry-suls@uiowa.edu
Tom Ollendick
tho@vt.edu
Stephan Swinnen
stephan.swinnen@faber.kuleuven.be
Dick Olson
rolson@psych.colorado.edu
Ross Thompson
rathompson@ucdavis.edu
Robert O'Neill
roneill@ed.utah.edu
Geoffrey Thorpe
geoffrey_thorpe@umit.maine.edu
Joel Paris
joel.paris@mcgill.ca
Bmce Thyer
bthyer@fsu.edu
Gilbert Parra
gparra@memphis.edu
Elliot Tucker-Drob
tuckerdrob@psy.utexas.edu
Ty Partridge
tpartrid@wayne.edu
Brent Turvey
bturvey@corpus-deliai.com
Del Paulhus
dpaulhus@psych.ubc.ca
Michael Vanyukov
mmv@pitt.edu
Keith Payne
payne@unc.edu
Joel Voss
joelvoss@illinois.edu
Reinhard Pekrun
pekmn@lmu.de
Mark v an Vugt
m.van.vugt@psy.vu.nl
Carin Perilloux
perilloux@mail.utexas.edu
Rich Walker
walkerr@wssu.edu
John Christopher Perry
jchristopher.perry@mcgill.ca
Colleen Ward
colleen.ward@vuw.ac.nz
Chris Peterson
chrispet@umich.edu
Duane Watson
dgwatson@gmail.com
Elizabeth Pomery
elizabeth.pomery@yale.edu
William Watson
william_watson@urmc.rochester.edu
Radek Ptak
radek.ptak@hcuge.ch
John Welte
welte @ ria. buffalo .edu
Kisha Radliff
radliff. 2 @ osu. edu
Robb Wilier
willer@berkeley.edu
Gabriel Radvansky
gabriel.a.radvansky. 1 @nd.edu
Gwen Wittenbaum
gwittenb @ msu . edu
Ranga Ramanujam
ranga @ owen . vanderb ilt.edu
Steve Wonderlich
swonderlich@nrifargo.com
Sara Rix
srix@aarp.org
Jenny Yiend
jenny.yiend@kcl.ac.uk
Richard Roberts
rroberts@ets.org
Tom Zentall
zentall@email.uky.edu
Kelly Rohan
kelly . rohan @ uvm . edu
CONTRIBUTORS
P C G Adam
The University of New South Wales, Sydney, NSW,
Australia
M Allemand
University of Zurich, Zurich, Switzerland
M W Allen
California State Polytechnic University, Pomona, CA,
USA
S Allison
University of Richmond, Richmond, VA, USA
A al-Nowaihi
University of Leicester, Leicester, UK
S M Andersen
New York University, New York, NY, USA
K M Antshel
State University of New York - Upstate Medical
University, New York, NY, USA
L Asarian
University of Zurich, Zurich, Switzerland
J R Averill
University of Massachusetts, Amherst, MA, USA
E C Azmitia
New York University, New York, NY, USA
B Bailer
Temple University, Philadelphia, PA, USA
J M Bailey
Northwestern University, Evanston, IL, USA
D S Bailis
University of Manitoba, Winnipeg, MB, Canada
M D Baker
East Carolina University, Greenville, NC, USA
J D Baldwin
University of California at Santa Barbara, Santa Barbara,
CA, USA
J I Baldwin
University of California at Santa Barbara, Santa Barbara,
CA, USA
J Barden
Howard University, Washington, DC, USA
E E Bartolini
Wesleyan University, Middletown, CT, USA
P Bartolomeo
Hopital Salpetriere, Paris, France; Catholic University,
Milan, Italy
M R Basso
University of Tulsa, Tulsa, OK, USA
A Baum '
The University of Texas at Arlington, Arlington, TX, USA
1 B Becker
Zina Pitcher Place, Ann Arbor, MI, USA
1 Beggan
University of Louisville, Louisville, KY, USA
D C Beidel
University of Central Florida, Orlando, FL, USA
D R Beike
University of Arkansas, Fayetteville, AR, USA
S Bell
University of Saskatchewan, Saskatoon, SK, Canada
S Bensmaia
University of Chicago, Chicago, IL, USA
L Bensman
Hilbert College, Hamburg, NY, USA
G G Berntson
Ohio State University, Columbus, OH, USA
B Berry
University of Nebraska - Lincoln, Lincoln, NE, USA
1 Deceased.
Xiii
xiv Contributors
M E Berryhill
University of Nevada, Reno, NV, USA
A Berti
University of Torino, Torino, Italy
D P Bestha
Carolinas Healthcare System, Charlotte, NC, USA
A M Betz
Florida Institute of Technology, Melbourne, FL, USA
R S Bhatt
University of Kentucky, Lexington, KY, USA
T Bjorgvinsson
McLean Hospital, Belmont, MA, USA
C J Black
University of Arizona, Tucson, AZ, USA
S Blum
University of California, Irvine, CA, USA
L Bobadilla
Western Carolina University, Cullowhee, NC, USA
P C Borchgrevink
St. Olavs University Hospital, Trondheim, Norivay;
Norwegian University of Science and Technology,
Trondheim, Norway
E Borgida
University of Minnesota, Minneapolis, MN, USA
M J Boyle
La Salle University, Philadelphia, PA, USA
M M Bradley
University of Florida, Gainesville, FL, USA
B Bridgeman
University of California, Santa Cruz, CA, USA
S F Brosnan
Georgia State University, Atlanta, GA, USA
M Brow
University of California, Irvine, CA, USA
K R Brown
Houston OCD Program, Houston, TX, USA
R T Brown
Wayne State University, Detroit, MI, USA
D Bub
University of Victoria, Victoria, BC, Canada
C L Buchanan
The Children’s Hospital of Philadelphia, Philadelphia,
PA, USA
K A Bujarski
Dartmouth Medical College, Lebanon, NH, USA
P Bull
University of York, Heslington, York, UK
D C Burdick
The Richard Stockton College of New Jersey, Galloway,
NJ, USA
M M Burke
Vanderbilt University, Nashville, TN, USA
N L Burke
University of South Florida, Tampa, FL, USA
J Burrus
Center for Academic and Workplace Readiness and
Success, Princeton, NJ, USA
S A Burt
Michigan State University, East Lansing, MI, USA
B J Bushman
The Ohio State University, Columbus, OH, USA; VU
University, Amsterdam, The Netherlands
L M Bylsma
University of South Florida, Tampa, FL, USA
J T Cacioppo
University of Chicago, Chicago, IL, USA
D D Cahn
State University of New York at New Paltz, New Paltz,
NY, USA
E H Callahan
Institute for Child Development, SUNY at Binghamton,
Binghamton, NY, USA
M C Capozzoli
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
E Cardena
Lund University, Lund, Sweden
P J Carnevale
University of Southern California, Los Angeles, CA, USA
P Case
Saint Andrews Presbyterian College, Laurinburg, NC,
USA
S E Cassin
Sunnybrook Health Sciences Centre and University of
Toronto, Toronto, ON, Canada
L Castro
University of Iowa, Iowa City, I A, USA
E P Charles
Pennsylvania State University, Altoona, PA, USA
S R Charlton
University of Central Arkansas, Conway, AR, USA
Contributors xv
V Chauhan
New School for Social Research, New York, NY, USA
E Chemerinski
Mount Sinai School of Medicine, New York, NY, USA;
fames J. Peters Veterans Affairs Medical Center, Bronx,
NY, USA
C F Chick
Cornell University, Ithaca, NY, USA
J G Chipperfield
University of Manitoba, Winnipeg, MB, Canada
S D Christman
University of Toledo, Toledo, OH, USA
T Chung
Pittsburgh Adolescent Alcohol Research Center,
Pittsburgh, PA, USA
L A Clark
University of Notre Dame, Notre Dame, IN, USA
E G Clary
Kutztown University, Kutztown, PA, USA
A Collins
University of Nebraska - Lincoln, Lincoln, NE, USA
P J Colombo
Tulane University, New Orleans, LA, USA
D R Combs
University of Texas at Tyler, Tyler, TX, USA
D M Condon
Northwestern University, Evanston, IL, USA
M Conrad
University of Illinois, Chicago, IL, USA
K S Cook
Stanford University, Stanford, CA, USA
P R Corlett
Yale University, New Haven, CT, USA
H Coskun
Abant Izzet Baysal University, Golkoy Bolu, Turkey
K Cote
Brock University, St. Catharines, ON, Canada
N Cowan
University of Missouri, Columbia, MO, USA
J Craner
University of Maine, Orono, ME, USA
M E Crawley
Tulane University, New Orleans, LA, USA
T D Crespi
The University of Hartford, West Hartford, CT, USA
A Culebras
Upstate Medical University, Syracuse, NY, USA;
Community General Hospital, Syracuse, NY, USA
C L Culpepper
University of Tennessee, Knoxville, TN, USA
] A Cummings
Zina Pitcher Place, Ann Arbor, MI, USA
R C Curtis
Adelphi University and W.A. White Institute, New York,
NY, USA
V Curtis
London School of Hygiene and Tropical Medicine,
London, UK
B P Daly
Drexel University, Philadelphia, PA, USA
J M Daly
Temple University, Philadelphia, PA, USA
M P Daly
Temple University, Philadelphia, PA, USA
J E Davidson
Lewis & Clark College, Portland, OR, USA
H P Davis
University of Colorado, Colorado Springs, CO, USA
A T DeMarco
Temple University, Philadelphia, PA, USA
M D'Esposito
University of California at Berkeley, Berkeley, CA, USA
J B F de Wit
The University of New South Wales, Sydney, NSW,
Australia
L de-Wit
University of Leuven (KU. Leuven), Leuven, Belgium
S Dhami
University of Leicester, Leicester, UK
C Di Dio
University of Parma, Parma, Italy
K S Dobson
University of Calgary, Calgary, AB, Canada
J Domingues
University of Connecticut, Storrs, CT, USA
M B Donnellan
Michigan State University, East Lansing, MI, USA
R L Doty
University of Pennsylvania School of Medicine,
Philadelphia, PA, USA
xvi Contributors
A L Dougall
The University of Texas at Arlington, Arlington, TX, USA
V Drago
IRCCS Centro San Giovanni di Dio Fatebenefratelli,
Brescia, Italy
S Duck
University of Iowa, Iowa City, I A, USA
J Duntley
The Richard Stockton College of New Jersey, Pomona, NJ,
USA
A B Durawa
Warsaw School of Social Sciences and Humanities,
Wroclaw, Poland
B S East Jr.
Tulane University, New Orleans, LA, USA
R J Edelmann
University of Roehampton, London, UK
D M Eichen
Temple University, Philadelphia, PA, USA
B J Ellis
University of Arizona, Tucson, AZ, USA
E K Englander
Bridgewater State University, Bridgewater, MA, USA
J T Enns
University of British Columbia, Vancouver, BC, Canada
A M Epp
University of Calgary, Calgary, AB, Canada
A Erdemir
Vanderbilt University, Nashville, TN, USA
T Ertelt
University of North Dakota, Fargo, ND, USA
EMR Escobar
York University, Toronto, ON, Canada
J Ey e
Lehigh University, Bethlehem, PA, USA
H J Eysenck'
Institute of Psychiatry, London, UK
S B G Eysenck
Institute of Psychiatry, London, UK
M Fabbri-Destro
Universita di Parma and IIT Brain Center for Motor and
Social Cognition, Parma, Italy
^Deceased.
A Fang
Boston University, Boston, MA, USA
E Fantino
University of California, San Diego, La Jolla, CA, USA
S V Faraone
State University of New York - Upstate Medical
University, New York, NY, USA
B Fehr
University of Winnipeg, Winnipeg, MB, Canada
A Fenigstein
Kenyon College, Gambier, OH, USA
R D Fields
National Institutes of Health (NICHD), Bethesda, MD,
USA
A J Figueredo
University of Arizona, Tucson, AZ, USA
D Fiset
Universite du Quebec en Outaouais, Gatineau, QC,
Canada
P Fisher
University of Liverpool, Liverpool, UK
W A Fisher
University of Western Ontario, London, ON, Canada
E B Foa
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
K A Force
Ryerson University, Toronto, ON, Canada
E A Fors
St. Olavs University Hospital, Trondheim, Norway;
Norwegian University of Science and Technology,
Trondheim, Norivay
J L Freeman
Forensic Solutions LLC, Sitka, AK, USA
A Freitas-Magalhaes
University Fernando Pessoa Health Sciences School,
Porto, Portugal
H S Friedman
University of California, Riverside, CA, USA
G O Gabbard
Baylor College of Medicine, Houston, TX, USA
A Gallace
Universita di Milano-Bicocca, Milan, Italy
M W Gallagher
University of Kansas, Lawrence, KS, USA
Contributors xvii
V Gallese
University of Parma, Parma, Italy
E Gambrill
University of California, Berkeley, CA, USA
G Ganis
University of Plymouth, Plymouth, UK; Harvard Medical
School, Boston, MA, USA
C Garbacz
Flora, MS, USA
R A Garcia
University of Arizona, Tucson, AZ, USA
Y Gazes
Columbia University Medical Center, New York, NY,
USA
D C Geary
University of Missouri, Columbia, MO, USA
N Geary
Institute of Food, Nutrition and Health, Schwerzenbach,
Switzerland
K Gee
University of Sheffield, Sheffield, UK
M Gelfand
University of Maryland, College Park, MD, USA
D Gentner
Northwestern University, Evanston, IL, USA
A Gerbasi
California State University, Northridge, CA, USA
W Gerin
The Pennsylvania State University, Pennsylvania, PA,
USA
F X Gibbons
Dartmouth College, Hanover, NH, USA
R Gifford
University of Victoria, Victoria, BC, Canada
A L Gilchrist
University of Missouri, Columbia, MO, USA
) M Gillis
Auburn University, Auburn, AL, USA
G Gillund
College of Wooster, Wooster, OH, USA
E L Glisky
University of Arizona, Tucson, AZ, USA
VGloy
Institute of Food, Nutrition and Health, Schwerzenbach,
Switzerland
D A Goldman
Cornell University, Ithaca, NY, USA
E B Goldsmith
Florida State University, Tallahassee, FL, USA
D C Gooch
University of York, York, UK
M Goodman
University of California, Davis, CA, USA
R Goodvin
Western Washington University, Bellingham, WA, USA
SCC Graham
Friedrich-Alexander University, Erlangen, Germany
J E Grant
University of Minnesota Medical Center, Minneapolis,
MN, USA
S L Gray
University of Tennessee, Knoxville, TN, USA
B Greyson
University of Virginia, Charlottesville, VA, USA
M M Griffin
Vanderbilt University, Nashville, TN, USA
S Grossberg
Boston University, Boston, MA, USA
R Gupta
University of Iowa, Iowa City, IA, USA
C A Gutner
Boston University, Boston, MA, USA
R Hagen
Norwegian University of Science and Technology,
Trondheim, Norway
P W Halligan
Cardiff University, Cardiff, UK
P A Hancock
University of Central Florida, Orlando, FL, USA
F Happe
King's College, London, UK
E Harmon-Jones
Texas A&M University, College Station, TX, USA
C R Harris
University of California, San Diego, CA, USA
J H Harvey
The University of Iowa, Iowa City, IA, USA
E Hatfield
University of Hawaii at Manoa, Honolulu, HI, USA
xviii Contributors
B Hayslip Jr.
University of North Texas, Denton, TX, USA
J A Headley
The University of Akron, Akron, OH, USA
K M Heilman
University of Florida, Gainesville, FL, USA
K A Henderson
North Carolina State University, Raleigh, NC, USA
L M Henderson
University of York, York, UK
A R Hill
Tulane University, New Orleans, LA, USA
O Hjemdal
Norwegian University of Science and Technology,
Trondheim, Norway
R M Hodapp
Vanderbilt University, Nashville, TN, USA
S Hoeffler
Vanderbilt University, Nashville, TN, USA
C Hoelscher
University of Freiburg, Freiburg, Germany
S G Hofmann
Boston University, Boston, MA, USA
A E Holmes
Baylor University, Waco, TX, USA
R W Hood Jr.
University of Tennessee at Chattanooga, Chattanooga,
TN, USA
V Hoorens
Katholieke Universiteit Leuven, Leuven, Belgium
S Horning
University of Colorado, Colorado Springs, CO, USA
C Hunt
University of Minnesota, Minneapolis, MN, USA
E R Igou
University of Limerick, Co. Limerick, Ireland
J Z Ingber
Fordham University, New York, NY, USA
R Ingram
University of Kansas, Lawrence, KS, USA
W J Jacobs
University of Arizona, Tucson, AZ, USA
L E James
University of Colorado at Colorado Springs, Colorado
Springs, CO, USA
B L Janzen
University of Saskatchewan, Saskatoon, SK, Canada
J A Jimenez
University of California, San Diego, La Jolla, CA, USA
J Johnson
York University, Toronto, ON, Canada
P K Jonason
University of South Alabama, Mobile, AL, USA
D Jones
University of Utah, Salt Lake City, UT, USA
K Jones
University of Nevada, Reno, NV, USA
A C Jordan
University of Arizona, Tucson, AZ, USA
R A Josephs
The University of Texas at Austin, Austin, TX, USA
K W Jung
Asan Medical Center, Seoul, South Korea
J Kagan
Harvard University, Cambridge, MA, USA
S J Karau
Southern Illinois University, Carbondale, IL, USA
F R Kardes
University of Cincinnati, Cincinnati, OH, USA
J D Kassel
University of Illinois, Chicago, IL, USA
A Kayser
University of California at San Francisco, Emeryville, CA,
USA
AKeil
University of Florida, Gainesville, FL, USA
E F Kelly
University of Virginia, Charlottesville, VA, USA
E W Kelly
University of Virginia, Charlottesville, VA, USA
I W Kelly
University of Saskatchewan, Saskatoon, SK, Canada
P C Kendall
Temple University, Philadelphia, PA, USA
R V Kendrick
University of Tennessee, Knoxville, TN, USA
LEO Kennair
Department of Psychology, Norwegian University of
Science and Technology, Trondheim, Norway
Contributors xix
D T Kenrick
Arizona State University, Tempe, AZ, USA
J Kenworthy
The University of Texas at Arlington, Arlington, TX, USA
M L Kern
University of California, Riverside, CA, USA
) F Kihlstrom
University of California, Berkeley, CA, USA
D R Kille
University of Waterloo, Waterloo, ON, Canada
Y K Kim
University of Southern California, Los Angeles, CA, USA
Y Y Kim
University of Oklahoma, Norman, OK, USA
F A A Kingdom
McGill University, Montreal, QC, Canada
H S Kirshner
Vanderbilt University School of Medicine, Nashville, TN,
USA
S Kitayama
University of Michigan, Ann Arbor, MI, USA
A M Klahr
Michigan State University, East Lansing, MI, USA
S J Ko
San Diego State University, San Diego, CA, USA
T Kohut
University of Western Ontario, London, ON, Canada
A Kordik
Friedrich-Alexander University, Erlangen, Germany
J K Kreuz
The University of Akron, Akron, OH, USA
R F Krueger
University of Minnesota, Minneapolis, MN, USA
K Krysinska
University of Queensland, Brisbane, QLD, Australia
R Kurzban
University of Pennsylvania, Philadelphia, PA, USA
D Laible
Lehigh University, Bethlehem, PA, USA
Y Lampl
Wolfson Medical Center, Holon, Israel; Tel Aviv
University, Holon, Israel
D K Lapsley
University of Notre Dame, Notre Dame, IN, USA
H Lasnik
The University of Maryland, College Park, MD, USA
B Latzko
University of Leipzig, Leipzig, Germany
J H Laurence
Temple University, Philadelphia, PA, USA
H Lavine
State University of New York - Stony Brook, Stony Brook,
NY, USA
M J Lawler
University of South Dakota, Vermillion, SD, USA
E-K Lee
University of Illinois, Urbana-Champaign, IL, USA
B T Legesse
McLean Hospital, Belmont, MA, USA
A M Legg
University of California, Riverside, CA, USA
R Lehmann
University of Zurich, Zurich, Switzerland
A F Lemieux
Emory University, Atlanta, GA, USA; Purchase College -
State University of New York, Purchase, NY, USA; and
National Consortium for the Study of Terrorism and
Responses to Terrorism, University of Maryland, College
Park, MD, USA
J G Lenz
Florida State University, Tallahassee, EL, USA
S J Lepore
Temple University, Philadelphia, PA, USA
D Lester
The Richard Stockton College ofNeiv Jersey, Pomona, NJ,
USA
A LeUnes
Texas A&M University, College Station, TX, USA
D S Levine
The University of Texas at Arlington, Arlington, TX, USA
N P Li
Singapore Management University, Singapore
YJ Li
Arizona State University, Tempe, AZ, USA
S H Liening
The University of Texas at Austin, Austin, TX, USA
S O Lilienfeld
Emory University, Atlanta, GA, USA
xx Contributors
D A Lishner
University of Wisconsin - Oshkosh, Oshkosh, WI, USA
J A Litman
University of South Florida, St. Petersburg, FL, USA
J H Liu
Victoria University of Wellington, Wellington,
New Zealand
K Lorduy
The University of Texas at Arlington, Arlington, TX, USA
A Luszczynska
University of Colorado, Colorado Springs, CO, USA;
Warsaw School of Social Sciences and Humanities,
Wroclaw, Poland
S J Lynn
Binghamton University, Binghamton, NY, USA
C N Macrae
University of Aberdeen, Aberdeen, UK
V Madaan
University of Virginia Health System, Charlottesville, VA,
USA
D Maestripieri
The University of Chicago, Chicago, IL, USA
R G Mair
University of New Hampshire, Durham, NH, USA
T Malti
University of Toronto Mississauga, Mississauga, ON,
Canada
L Manfredi
University of Chicago, Chicago, IL, USA
J Marino
University of North Dakota, Fargo, ND, USA
P F Martelli
University of California, Berkeley, CA, USA
D M Marx
San Diego State University, San Diego, CA, USA
J C Masdeu
New York Medical College, Valhalla, New York, NY,
USA
R Mata
University of Basel, Basel, Switzerland
K J Mathewson
McMaster University, Hamilton, ON, Canada
G Matthews
University of Cincinnati, Cincinnati, OH, USA
M D Matthews
United States Military Academy, West Point, NY, USA
R E Mayer
University of California, Santa Barbara, CA, USA
K C McCulloch
Idaho State University, Pocatello, ID, USA
K L McDonald
University of Alabama, Tuscaloosa, AL, USA
M McGrath
Unity Health System, Rochester, NY, USA
R E McGrath
Fairleigh Dickinson University, Teaneck, NJ, USA
W A McKim
Memorial University of Newfoundland, St. John's, NL,
Canada
D W McNeil
West Virginia University, Morgantown, WV, USA
E A Meadows
Central Michigan University, Mount Pleasant, MI, USA
P H Mehta
The University of Oregon, Eugene, OR, USA
F Mesa
University of Central Florida, Orlando, FL, USA
R G Meyer
University of Louisville, Louisville, KY, USA
W-U Meyer
University of Bielefeld, Bielefeld, Germany
J G Miller
New School for Social Research, New York, NY, USA
P J Mills
University of California, San Diego, La Jolla, CA, USA
N Minniti
Temple University, Philadelphia, PA, USA
V Miskovic
McMaster University, Hamilton, ON, Canada
J E Mitchell
University of North Dakota, Fargo, ND, USA
CMok
Palo Alto University, Palo Alto, CA, USA
T A Morgan
University of Iowa, Iowa City, I A, USA
A Morin
Mount Royal University, Calgaiy, AB, Canada
Contributors xxi
E D Murray
McLean Hospital, Belmont, MA, USA
M P Mychailyszyn
Temple University, Philadelphia, PA, USA
S-J Myung
Asan Medical Center, Seoul, South Korea
C T Nagoshi
Arizona State University, Tempe, AZ, USA
K Naragon-Gainey
University of Iowa, Iowa City, I A, USA
G Narasimham
Vanderbilt University, Nashville, TN, USA
M Natishyn
The University of Texas at Arlington, Arlington, TX, USA
N Nayak
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
M Neppi-Modona
University of Torino, Torino, Italy
K Neville
Northwestern University, Chicago, IL, USA
A B Newberg
University of Pennsylvania, Philadelphia, PA, USA
M Niepel
University of Bielefeld, Bielefeld, Germany
M M Nieves
University of Central Florida, Orlando, FL, USA
M Nikolas
University of Washington School of Medicine/Seattle
Children's Hospital, Seattle, WA, USA
G J Norman
Ohio State University, Columbus, OH, USA
P L Nunez
Tulane University, New Orleans, LA, USA
S Oberle
University of Colorado at Colorado Springs, Colorado
Springs, CO, USA
E H O'Brien
University of Michigan, Ann Arbor, MI, USA
P J O'Brien
Alberta Health Services, Calgary, AB, Canada
B L Odlaug
University of Minnesota Medical Center, Minneapolis,
MN, USA
L A O'Leary
RHAM High School, Hebron, CT, USA
M A Olson
University of Tennessee, Knoxville, TN, USA
J Omarzu
Loras College, Dubuque, IA, USA
S H Osipow
The Ohio State University, Columbus, OH, USA
W F Overton
Temple University, Philadelphia, PA, USA
M E Owens
University of Central Florida, Orlando, FL, USA
D J Packer
Lehigh University, Bethlehem, PA, USA
K S Page
University of North Texas, Denton, TX, USA
V Panaite
University of South Florida, Tampa, FL, USA
S T Pardo
Cornell University, Ithaca, NY, USA
J Paris
McGill University, Montreal, QC, Canada;
Institute of Community and Family Psychiatry,
Montreal, QC, Canada
J Park
University of Michigan, Ann Arbor, MI, USA
A Parker
Gothenburg University, Gothenburg, Sweden
C Parkinson
Dartmouth College, Hanover, MA, USA
A E Parrish
Georgia State University, Atlanta, GA, USA
T Partridge
Wayne State University, Detroit, MI, USA
J Pascual-Leone
York University, Toronto, ON, Canada
M L Patterson
University of Missouri, St. Louis, MO, USA
J H Patton
Baylor University, Waco, TX, USA
P B Paulus
The University of Texas at Arlington, Arlington, TX, USA
A D Pellegrini
University of Minnesota /Twin Cities Campus,
Minneapolis, MN, USA
xxii Contributors
A N Perry
Zina Pitcher Place, Ann Arbor, MI, USA
R P Perry
University of Manitoba, Winnipeg, MB, Canada
D Peterson
University of Nevada, Reno, NV, USA
G W Peterson
Florida State University, Tallahassee, FL, USA
J D Pettigrew
University of Queensland, St Lucia, QLD, Australia
R E Petty
Ohio State University, Columbus, OH, USA
D Paez
Universidad del Pais Vasco, San Sebastian, Spain
M V Pezzo
University of South Florida, St. Petersburg, FL, USA
B O Ploog
College of Staten Island, City University of New York,
Staten Island, NY, USA
E A Pomery
Yale University, New Haven, CT, USA
M E Portman
Private Practice, Jerusalem, Israel
S S Posavac
Vanderbilt University, Nashville, TN, USA
F C Power
University of Notre Dame, Notre Dame, IN, USA
M B Powers
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
D E Presti
University of California, Berkeley, CA, USA
B El Price
McLean Hospital, Belmont, MA, USA
D Pusch
Sheldon Chumir Centre, Calgary, AB, Canada
J E Pyers
Wellesley College, Wellesley, MA, USA
S Quadflieg
Catholic University of Louvain, Louvain-La-Neuve,
Belgium
P C Quinn
University of Delaware, Newark, DE, USA
J S Rabin
Towson University, Towson, MD, USA
M Rabinowitz
Fordham University, New York, NY, USA
E Race
Boston University School of Medicine, Boston, MA, USA
F Rachal
Baylor College of Medicine, Houston, TX, USA
El Rachlin
Stony Brook University, Stony Brook, NY, USA
G A Radvansky
University of Notre Dame, Notre Dame, IN, USA
A J Raines
Columbia University, New York, NY, USA
J W Ransom
University of Arkansas, Fayetteville, AR, USA
R L Rapson
University of Hawaii at Manoa, Honolulu, HI, USA
N A Rector
Sunnybrook Health Sciences Centre and University of
Toronto, Toronto, ON, Canada
S Reicher
University of St. Andrews, Fife, UK
R Reisenzein
University of Greifswald, Greifswald, Germany
W Revel le
Northwestern University, Evanston, IL, USA
V F Reyna
Cornell University, Ithaca, NY, USA
R B Ricco
California State University, San Bernardino, CA, USA
B J Ries
Minnesota State University, Mankato, MN, USA
J J Rieser
Vanderbilt University, Nashville, TN, USA
R R Riffle
Wright State University, Dayton, OH, USA
J H Riskind
George Mason University, Fairfax, VA, USA
G Rizzolatti
Universita di Parma and IIT Brain Center for Motor and
Social Cognition, Parma, Italy
El S Roane
SUNY Upstate Medical University, Syracuse, NY, USA
K El Roberts
University of California, Berkeley, CA, USA
Contributors xxiii
R D Roberts
Center for Academic and Workplace Readiness and
Success, Philadelphia, PA, USA
R W Robins
University of California, Davis, CA, USA
J L Rodriguez
Central Michigan University, Mount Pleasant, MI, USA
R G Romanczyk
Institute for Child Development, SUNY at Binghamton,
Binghamton, NY, USA
R Rosenthal
University of California, Riverside, CA, USA
M Roth
Auburn University, Auburn, AL, USA
P Roussos
Mount Sinai School of Medicine, New York, NY, USA
A G Rosch
Friedrich-Alexander University, Erlangen, Germany
K H Rubin
University of Maryland, College Park, MD, USA
M A Runco
University of Georgia, Athens, GA, USA
S W Russell
Northwestern University, Chicago, IL, USA
D H Saklofske
University of Western Ontario, London, ON, Canada
C Salmon
University of Redlands, Redlands, CA, USA
C A Sanislow
Wesleyan University, Middletown, CT, USA
B A Sarb
Western Washington University, Bellingham, WA, USA
S A Saribay
Bogazici University, Bebek, Istanbul, Turkey
A Scarpa
Virginia Tech, Blacksburg, VA, USA
L M Schaefer
University of South Florida, Tampa, FL, USA
H E Schendan
Plymouth University, Plymouth, UK
G L Schlomer
University of Arizona, Tucson, AZ, USA
L A Schmidt
McMaster University, Hamilton, ON, Canada
T R Schneider
Wright State University, Dayton, OH, USA
A Schnider
Hopitaux Universitaires de Geneve, Geneva, Switzerland
S M Schueller
University of California, San Francisco, CA, USA
O C Schultheiss
Friedrich-Alexander University, Erlangen, Germany
J A Sefcek
Kent State University, Ashtabula, OH, USA
R A Sevcik
Georgia State University, Atlanta, GA, USA
M Sevlever
Auburn University, Auburn, AL, USA
D Shankweiler
Haskins Laboratories, New Haven, CT, USA; University
of Connecticut, Storrs, CT, USA
D Shannahoff-Khalsa
University of California San Diego, La Jolla, CA, USA
K J Sher
The University of Missouri and the Midwest Alcoholism
Research Center, Columbia, MO, USA
T F Shipley
Temple University, Philadelphia, PA, USA
L J Siever
Mount Sinai School of Medicine, New York, NY, USA;
James J. Peters Veterans Affairs Medical Center, Bronx,
NY, USA
K L Sifferd
Elmhurst College, Elmhurst, IL, USA
S T Sigmon
University of Maine, Orono, ME, USA
R C Silver
University of California, Irvine, CA, USA
J Simner
University of Edinburgh, Edinburgh, UK
M Simonescu
State University of New York - Upstate Medical
University, New York, NY, USA
J D Sinnott
Towson University, Towson, MD, USA
J J Skowronski
Northern Illinois University, DeKalb, IL, USA
J C Smith
Roosevelt University, Chicago, IL, USA
xxiv Contributors
L Smith
Northwestern University, Evanston, IL, USA
P K Smith
University of California, San Diego, CA, USA
S F Smith
Vanderbilt University, Nashville, TN, USA
M Smithson
Australian National University, Canberra, ACT,
Australia
P Snell
Bridgewater State University, Bridgewater, MA, USA
M J Snowling
University of York, York, UK
KVSobel
University of Central Arkansas, Conway, AR, USA
A Soldan
University of Massachusetts - Dartmouth, North
Dartmouth, MA, USA
R Sommer
University of California, Davis, CA, USA
S C South
Purdue University, West Lafayette, IN, USA
C Spence
University of Oxford, Oxford, UK
C Spencer
University of Sheffield, Sheffield, UK
M R Sperling
Jefferson Medical College, Philadelphia, PA, USA
H J Spiers
University College London, London, UK
B Spinath
Heidelberg University, Heidelberg, Germany
B Spring
Northwestern University, Chicago, IL, USA
R Srinivasan
University of California, Irvine, CA, USA
S Stack
Wayne State University, Detroit, MI, USA
M A Stanley
Veterans Affairs Health Services Research and
Development Center of Excellence, Baylor College of
Medicine, Houston, TX, USA
P N Stearns
George Mason University, Fairfax, VA, USA
G A Stefanatos
Temple University, Philadelphia, PA, USA
K Steffen
University of North Dakota, Fargo, ND, USA
R M Stelmack
University of Ottawa, Ottawa, ON, Canada
Y Stern
Columbia University Medical Center, New York, NY,
USA
R J Sternberg
Oklahoma State University, Stillwater, OK, USA
H L Sterns
The University of Akron, Akron, OH, USA
T L Stewart
University of Manitoba, Winnipeg, MB, Canada
P C Stey
University of Notre Dame, Notre Dame, IN, USA
T C Stiles
St. Olavs University Hospital, Trondheim, Norway;
Norwegian University of Science and Technology,
Trondheim, Norway
M K Stimmler
University of California, Berkeley, CA, USA
R Stinchfield
University of Minnesota Medical School, Minneapolis,
MN, USA
M L Stock
The George Washington University, Washington, DC,
USA
E L Stocks
University of Texas at Tyler, Tyler, TX, USA
S Stolarz-Fantino
University of California, San Diego, La Jolla, CA, USA
A Strasser
Technical University, Munich, Germany
M Stroebe
Utrecht University, Utrecht, The Netherlands
W Stroebe
Utrecht University, Utrecht, The Netherlands
G L Strosser
Southern Utah University, Cedar City, UT, USA
A A Stukas
La Trobe University, Bundoora, VIC, Australia
J N Swanson
The University of Texas at Arlington, Arlington, TX, USA
Contributors xxv
S M Swearer
University of Nebraska - Lincoln, Lincoln, NE, USA
K Sweeny
University of California, Riverside, CA, USA
S P Swinnen
Katholieke Universiteit Leuven, Leuven, Belgium
E B Talbot
University of South Dakota, Vermillion, SD, USA
A K Tamplin
University of Notre Dame, Notre Dame, IN, USA
R Tanoue
University of Nevada, Reno, NV, USA
J Taylor
Florida State University, Tallahassee, FL, USA
M Thayer
San Diego State University, San Diego, CA, USA
J K Thompson
University of South Florida, Tampa, FL, USA
L A Thompson
New Mexico State University, Las Cruces, NM, USA
R A Thompson
University of California, Davis, CA, USA
G L Thorpe
University of Maine, Orono, ME, USA
S J Tobin
University of Queensland, St Lucia, QLD, Australia
D Tranel
University of Ioiva, Iowa City, I A, USA
K Treadwell
University of Connecticut, Storrs, CT, USA
L A Trevino
The University of Texas at Arlington, Arlington, TX, USA
H C Triandis
University of Illinois, Champaign, IL, USA
B Tuller
National Science Foundation, Arlington, VA, USA
C L Turk
Washburn University, Topeka, KS, USA
B E Turvey
Forensic Solutions LLC, Sitka, AK, USA
M L Uhelski
The University of Texas at Arlington, Arlington, TX, USA
G E Vaillant
Massachusetts General Hospital, Boston, MA, USA
K A Valentine
Singapore Management University, Singapore
F van Dongen
University of Limerick, Co. Limerick, Ireland
WAP van Tilburg
University of Limerick, Co. Limerick, Ireland
A M Vargovich
West Virginia University, Morgantown, WV, USA
M Verfaellie
Boston University School of Medicine, Boston, MA, USA
A Verges
The University of Missouri and the Midwest Alcoholism
Research Center, Columbia, MO, USA
E A Virmani
Center for Child and Family Studies, Sausalito, CA, USA
K M von Ranson
University of Calgary, Calgary, AB, Canada
J Wagemans
University of Leuven (K.U. Leuven), Leuven, Belgium
J Waguspack
University of Texas at Tyler, Tyler, TX, USA
W R Walker
Winston-Salem State University, Winston-Salem, NC,
USA
G L Wallace
King's College, London, UK
P Wallace
The Johns Hopkins University, Baltimore, MD, USA
E A Wasserman
University of Iowa, Iowa City, IA, USA
S F Waters
University of California, Davis, CA, USA
D Watson
University of Notre Dame, Notre Dame, IN, USA
D G Watson
University of Illinois, Urbana-Champaign, IL, USA
M R Watson
University of British Columbia, Vancouver, BC, Canada
W H Watson
University of Rochester Medical Center, Rochester, NY,
USA
C A Weaver
Baylor University, Waco, TX, USA
C M Weaver
Palo Alto University, Palo Alto, CA, USA
xxvi Contributors
L Wei
University of London, London, UK
M M Weissman
Columbia University, New York, NY, USA; New York
State Psychiatric Institute, New York, NY, USA
D P Welsh
University of Tennessee, Knoxville, TN, USA
C Westenbroek
Zina Pitcher Place, Ann Arbor, Ml, USA
T Wheatley
Dartmouth College, Hanover, MA, USA
J L Whitaker
The Ohio State University, Columbus, OH, USA
T A Widiger
University of Kentucky, Lexington, KY, USA
0 Wilder
University of South Florida, St. Petersburg, FL, USA
1 Wilhelmsen
Haraldsplass Deaconal Hospital, Institute of Medicine,
University of Bergen, Bergen, Norway
A Wilke
Clarkson University, Potsdam, NY, USA
D R Wilson
Creighton University, Omaha, NE, USA
L Wilson
Virginia Tech, Blacksburg, VA, USA
J Wilt
Northwestern University, Evanston, IL, USA
R Winograd
The University of Missouri and the Midwest Alcoholism
Research Center, Columbia, MO, USA
K C Winters
University of Minnesota Medical School, Minneapolis,
MN, USA
J C Wise
Oglethorpe University, Atlanta, GA, USA
J V Wood
University of Waterloo, Waterloo, ON, Canada
F J Yammarino
State University of New York at Binghamton,
Binghamton, NY, USA
D-H Yang
Asan Medical Center, Seoul, South Korea
K L Yoon
University of Maine, Orono, ME, USA
F M Zaromb
Center for Validity Research, Princeton, NJ, USA
M J Zawadzki
The Pennsylvania State University, Pennsylvania, PA,
USA
E C Zoloth
Wesleyan University, Middletown, CT, USA
HOW TO USE THE ENCYCLOPEDIA
The Encyclopedia of Human Behavior is intended for
use by students, research professionals, and interested
others. Articles have been chosen to reflect major dis-
ciplines in the study of human behavior, common
topics of research by professionals in this domain,
and areas of public interest and concern. Each article
serves as a comprehensive overview of a given area,
providing both breadth of coverage for students and
depth of coverage for research professionals. We have
designed the encyclopedia with the following features
for maximum accessibility for all readers.
Articles in the encyclopedia are arranged alphabet-
ically by subject. Complete tables of contents appear
in all volumes. The index is located in Volume 3.
Because the reader's topic of interest may be listed
under a broader article title, we encourage use of the
index for access to a subject area, rather than use of
the table of contents alone.
Each article contains a glossary, cross-references,
and a further reading list. The outline allows a quick
scan of the major ideas discussed within each article.
The glossary contains terms that may be unfamiliar to
the reader, with each term defined in the context of its
use in that article. Thus, a term may appear in the
glossary of another article defined in a slightly differ-
ent manner, or with a subtle nuance specific to that
article. For clarity, we have allowed these differences
to remain so that terms are defined relative to the
context of each article.
Each article has been cross-referenced to other
related articles in the encyclopedia. Cross-references
will always appear at the end of an article. Where
multiple cross-references apply to an article, the cross-
references will be listed alphabetically. We encourage
readers to use the cross-references to locate other arti-
cles in the encyclopedia that will provide more detailed
information about a particular subject.
The further reading section lists recent secondary
sources to aid the reader in locating more detailed or
technical information. Review and research articles
that are considered to be of primary importance to
the understanding of a given subject area are also
listed. The further reading lists are not intended to
provide a full reference listing of all the material
covered in the context of a given article, but are
provided as guides.
XXVII
This page intentionally left blank
PREFACE
The social scientists have a long way to go to catch up, but they may be up to the most important scientific business of
all, if and when they finally get to the right questions. Our behavior toward each other is the strangest, most
unpredictable, and almost entirely unaccountable of all the phenomena with which we are obliged to live.
Lewis Thomas
Psychology, the study of the human mind, has made many rapid strides during the past four decades.
There is now, more than ever before, a real need for a standard reference work covering all aspects of human
behavior. The Encyclopedia of Human Behavior is the most up-to-date and comprehensive collection of reviews
currently available. The essays will be of interest not only to clinical and experimental psychologists but also to
students in fields such as psychiatry, neuroscience, philosophy, cognitive science, and medicine. Indeed, given
the enormous range of topics covered, no one interested in human nature can fail to find something of interest
in each volume. The format of the volumes lends itself just as readily to casual perusal as it does to serious
inquiry.
My colleagues and I are often asked questions such as: What is the superego? What is repression? How
reliable is eyewitness testimony? How much sleep do we need? What do we know of the psychology of laughter,
language, cruelty, or politics? Or of love, cunning, and deceit? We have all experienced the frustration of not
being able to answer such questions or find the answers quickly without recourse to extensive library research.
This encyclopedia should prove to be an invaluable resource in such situations. Also, students of psychology
and related health professions will find this collection of articles useful as a starting point when they embark on
new research projects dealing with specific aspects of human behavior.
The study of human behavior is an enterprise that covers an enormous variety of subjects, ranging from the
minutiae of neurophysiology to such familiar but poorly understood topics as Freudian psychology. Psychology
is a science that is still very much in its infancy even though it has had a very long history, almost as long as
that of physics and biology. Anyone interested in the history of ideas should be puzzled by the differences
between advances in biology and advances in psychology. The progress in biology has been characterized by
a number of landmark discoveries, each of which resulted in a breakthrough in understanding, for example,
the discoveries of cells, Mendel's laws of heredity, chromosomes, mutations, and most recently DNA and the
genetic code. Psychology, however, has until recently been characterized by an embarrassingly long sequence of
'theories,' each of which was really nothing more than a passing fad that rarely outlived the person who
proposed it. I have always found this contrast to be quite remarkable and can think of no simple explanation
for it other than the fact that human behavior is inherently more complex, quixotic, and difficult to fathom.
Fortunately, the picture has changed radically over the past four decades, particularly in psychiatry and
cognitive neuroscience. There are two reasons for this change: First, there has been a growing dissatisfaction
with metaphorical explanations (Peter Medawar calls explanations of this kind "analgesics," for "they dull the
ache of incomprehension without removing the cause'') and a healthy trend toward replacing them with more
mechanistic explanations. Second, progress has been aided by the advent of several new technological innova-
tions for studying the structure and function of the human brain. This encyclopedia covers as many of these
recent advances as possible within a three-volume set.
My own experience is mainly in neuropsychology, medicine, and visual science, and I am therefore indebted
to the editorial advisory board, as they selected the authors in other areas and saw each essay through the long
process of peer review, revisions, and copy-editing. Most of the entries are by acknowledged experts in the field.
XXIX
xxx Preface
Given the nature and scope of this enterprise, some degree of overlap among the essays was not only inevitable
but also desirable since our goal was to ensure that each article was a self-contained summary of one specific
aspect of human behavior. Given the space limitations, each author was encouraged to provide an overview
and convey the general flavor of an area of research rather than attempt an exhaustive review. The result is a very
stimulating and informative collection of essays.
I have no doubt that this work will prove useful to specialists. If it also succeeds in kindling a spark of interest
in some aspect of human behavior among undergraduate and graduate students, then our efforts will have been
amply rewarded, for no enterprise is more important to the future of our species than an understanding of
human nature in all its diverse manifestations.
V S Ramachandran
CONTENTS
About the Editor-in-Chief
V
About the Editorial Advisory Board
vii
Reviewers
xi
Contributors
xiii
How to Use the Encyclopedia
xxvii
Preface
XXTX
VOLUME 1
Academic Achievement
B Spinath
1
Addictions and Adolescence
K C Winters, T Chung, R Stinchfield, I D Kassel, and M Conrad
9
Adolescence
S L Gray, C L Culpepper, and D P Welsh
22
Adrenal Glands
L A Trevino, M L Uhelski, A L Dougall, and A Baum
30
Aggression
B J Bushman and E H O'Brien
37
Aging and Cognition
S Horning and H P Davis
44
Aging and the Brain
P 1 Colombo, M E Crawley, B S East Jr., and A R Hill
53
Agnosia (including Prosopagnosia and Anosognosia)
A Berti and M Neppi-Modona
60
Agoraphobia and Panic Disorder
G L Thorpe, S T Sigmon, and K L Yoon
68
Agraphia and Alexia
D Fiset and D Bub
76
Alcohol: Psychosocial Effects
K J Sher, R Winograd, and A Verges
84
Alexithymia
V Panaite and L M Bylsma
92
Altruism and Helping Behavior
A A Stukas and E G Clary
100
XXXI
xxxii Contents
Alzheimer's Disease
Y Gazes, A Soldan, and Y Stern
108
Amnesia
/ F Kihlstrom and E L Glisky
116
Amnesia and the Brain
E Race and M Verfaellie
125
Analogical Reasoning
D Centner and L Smith
130
Anger
1 R Averill
137
Animal Cognition
L Castro and E A Wasserman
145
Antisocial and Narcissistic Personality Disorder
L Bobadilla and j Taylor
154
Anxiety and Fear
D W McNeil, A M Vargovich, B 1 Ries, and C L Turk
161
Anxiety Disorders
1 L Rodriguez and E A Meadows
169
Aphasia
H S Kirshner
177
Appetite
N Geary
187
Asperger's Syndrome and Nonverbal Learning Disorder
1 M Gillis, M Sevlever, and M Roth
199
Associative Learning
E Fantino and S Stolarz-Fantino
204
Attention
C Spence
211
Attention Deficit Hyperactivity Disorder
M Simonescu, K M Antshel, and S V Faraone
218
Attitude Change
R E Petty
224
Attitude Formation
M A Olson and R V Kendrick
230
Attribution
S / Tobin
236
Autism and Pervasive Developmental Disorders
R G Romanczyk and E FI Callahan
243
Autobiographical Remembering and the Self
] J Skowronski and W R Walker
251
Avoidant Personality Disorder
C A Sanislow, E E Bartolini, and E C Zoloth
257
Behavior Analysis
H S Roane and A M Betz
267
Behavior Genetics of Personality
S C South and R F Krueger
274
Contents xxxiii
Behavior Measurement in Psychobiological Research 280
G I Norman, G G Bemtson, and J T Cacioppo
Behavioral Economics 288
S Dhami and A al-Nowaihi
Behavioral Genetics 301
C T Nagoshi
Behavioral Medicine 311
7 A Jimenez and P J Mills
Behavioral Pharmacology 319
W A McKim
The Behavior-Genetics of Intelligence 326
M A Runco
Big Five Model and Personality Disorders 330
T A Widiger
Bilingualism and Multilingualism 338
L Wei
Biofeedback Therapy 344
K W Jung, D-H Yang, and S-J Myung
Bipolar Disorder 348
/ D Pettigrew
Birth Order, Effect on Personality, and Behavior 353
C Salmon
Body Dysmorphic Disorder: A Review 360
/ E Grant and B L Odlaug
Body Image 365
N L Burke, L M Schaefer, and J K Thompson
Borderline Personality Disorder 372
7 Paris
The Brain 377
R G Mair
Brain and Behavior Relationships 386
B T Legesse, B H Price, and E D Murray
Brain Chemicals: Global Projections of Ancient Aromatic Neuro transmitters 396
E C Azmitia
Brainwashing and Totalitarian Influence 406
S M Andersen and S A Saribay
Bulimia Nervosa 413
7 E Mitchell, K Steffen, T Ertelt, and J Marino
Bullying 417
S M Swearer, A Collins, and B Berry
Caffeine 423
W Revelle, D M Condon, and J Wilt
Career Development 430
S H Osipoiu
Catecholamines and Behavior 434
L A Trevino, K Lorduy, M Natishyn, A L Dougall, and A Baum
xxxiv Contents
Central Auditory Processing Disorders
G A Stefanatos and A T DeMarco
441
Central Nervous System
B P Daly, D M Eichen, B Bailer, R T Brown, and C L Buchanan
454
Child Abuse
M J Lawler and E B Talbot
460
Childhood Mental Disorders
A Scarpa and L Wilson
467
Chunking
A L Gilchrist and N Cowan
476
Classical Conditioning
B O Ploog
484
The Clinical and Cognitive Psychology of Conflict
A Parker
492
Clinical Assessment
E Gambrill
499
Clinical Psychology: An Information Processing Approach
P Fisher
510
Cognition and Personality
G Mattheivs
517
Cognitive Behavior Therapy
R Hagen and O Hjemdal
525
Cognitive Bias
A Wilke and R Mata
531
Cognitive Disorders
M E Berryhill, D Peterson, K Jones, and R Tanoue
536
Cognitive Dissonance Theory
E Harmon-Jones
543
Comparative Primate Psychology
D Maestripieri
550
Competition
S H Liening, P H Mehta, and R A Josephs
556
Confabulation and Reality Filtering
A Schnider
563
Conflict Communication
D D Cahn
572
Conformity and Obedience
D J Packer
580
Consumer Psychology
S S Posavac, F R Kardes, and S Hoeffler
589
Coping
S Blum, M Brow, and R C Silver
596
Creative and Imaginative Thinking
M A Runco
602
Creativity
V Drago and K M Heilman
606
Contents xxxv
Crisis Management 618
A M Legg and K Sweeny
Cross-Cultural Adaptation 623
V Y Kim
Crowd Psychology 63 1
S Reicher
Crowding: Effects on Health and Behavior 638
S J Lepore
Cultural Psychology 644
/ G Miller and V Chauhan
Decision Making (Individuals) 651
C F Chick, S T Pardo, V F Reyna, and D A Goldman
Defense Mechanisms 659
G E Vaillant
Delusions 667
P R Corlett
Dependent Personality (Disorder) 674
T A Morgan and L A Clark
Depression 682
R Ingram
Depth Perception 690
M R Watson and I T Enns
Developmental Psychopathology 697
A M Klahr, S A Burt, and M Nikolas
Disgust 702
V Curtis
Dissociative Disorders 710
E Cardeha
Divorce 716
S Stack
Drugs, the Brain, and Behavior 724
D E Presti
Dyslexia 732
M I Snowling, D C Gooch, and L M Henderson
VOLUME 2
Eating Disorder Not Otherwise Specified 1
K M von Ranson
Ecological Psychology 7
E P Charles and R Sommer
Electroconvulsive Therapy and Transcranial Magnetic Stimulation 13
V Madaan, D P Bestha, and D R Wilson
Electroencephalography 15
R Srinivasan and P L Nunez
Embarrassment and Blushing 24
R I Edelmann
xxxvi Contents
Empathy
E L Stocks and D A Lishner
32
Empirical Challenges to Conventional Mind-Brain Theory
E F Kelly and E W Kelly
38
Environmental Cognition
C Spencer and K Gee
46
Environmental Psychology
R Gifford
54
Episodic and Semantic Systems of Autobiographical Memory
D R Beike and J W Ransom
61
Episodic Memory
G Gillund
68
Equity Theory
E Hatfield, R L Rapson, and L Bensman
73
Event-Related Potentials (ERPs)
M M Bradley and A Keil
79
Evidence-Based Practice
B Spring, K Neville, and S W Russell
86
Evolutionary Clinical Psychology
LEO Kennair
94
Evolutionary Developmental Psychology
A C Jordan, G L Schlomer, and B J Ellis
103
Evolutionary Personality Psychology
A / Figueredo, J A Sefcek, C J Black, R A Garcia, and W J Jacobs
111
Evolutionary Psychology
R Kurzban
118
Evolutionary Social Psychology
D T Kenrick and Y J Li
125
Exercise and Sport Psychology
A LeUnes
132
Expectation
V Hoorens
142
Extraversion-Introversion
D H Saklofske, H J Eysenck, S B G Eysenck, R M Stelmack, and W Revelle
150
Eye Movements
B Bridgeman
160
Eyewitness Identification
M McGrath
167
Facial Expression of Emotion
A Freitas-Magalhdes
173
Family Systems
W H Watson
184
Forensic Psychology in Contemporary Society
T D Crespi and L A O'Leary
194
Free Will
H Rachlin
199
Contents xxxvii
Friendship 205
B Fehr
Generalized Anxiety Disorder 215
M E Portman, J H Riskind, and N A Rector
Genius, Eminence, and Giftedness 221
/ E Davidson
Gerotechnology 228
D C Burdick
Gestalt Psychology and the Development of Perceptual Organization 238
P C Quinn and R S Bhatt
Gestures 248
PBull
Glial Cells 255
R D Fields
Grandparenthood 261
B Hayslip Jr. and K S Page
Grief and Bereavement 268
M Stroebe and W Stroebe
Group Dynamics 276
P B Paulus, J Kenworthy, and FI Coskun
Hallucinations 283
K A Bujarski and M R Sperling
Handedness 290
S D Christman
Hippocampal Formation 297
H J Spiers
History of Film and Music 305
K A Force
Histrionic Personality Disorder 312
S F Smith and S O Lilienfeld
HIV/AIDS: The Role of Behavior and the Social Environment in a Global Pandemic 316
J B F de Wit and P C G Adam
Homeostasis 324
L Asarian, V Gloy, and N Geary
Homicide 334
/ Duntley
Hope and Optimism 342
D S Bailis and J G Chipperfield
Hormones and Behavior 350
7 B Becker, J A Cummings, A N Perry, and C Westenbroek
Human Factors/Ergonomics 358
P A Hancock
Human Intelligence 364
R J Sternberg
Human Mating 371
P K Jonason, K A Valentine, and N P Li
xxxviii Contents
Hypnosis 378
S / Lynn
Hypochondriasis or Health Anxiety 385
I Wilhelmsen
Id, Ego, and Superego 393
D K Lapsley and P C Stey
Implicit Memory 400
H E Schendan
Impression Formation 410
C N Macrae and S Quadflieg
Individual Differences in Temperament: Definition, Measurement, and Outcomes 418
K I Mathewson, V Miskovic, and L A Schmidt
Individualism 426
I Park and S Kitayama
Inner Speech 436
A Morin
Intellectual Disabilities 444
M M Griffin, M M Burke, and R M Hodapp
Intention 452
C Parkinson and T Wheatley
Internet Behavior 458
P Wallace
Interpersonal Perception and Communication 465
J Omarzu and J H Harvey
Interpersonal Psychotherapy 472
A 1 Raines and M M Weissman
Jealousy 479
P N Stearns
Judgment 487
S R Charlton and K V Sobel
Jury Psychology 495
B E Turvey and J L Freeman
Kinship 503
D Jones
Language Development 511
J C Wise and R A Sevcik
Leadership 517
F / Yammarino
Learned Helplessness 525
/ N Swanson, A L Dougall, and A Baum
Leisure 531
K A Henderson
Logic: Development of Logical Operations 538
J Pascual-Leone, E MR Escobar, and J Johnson
Love and Intimacy 550
E Hatfield, L Bensman, and R L Rapson
Contents xxxix
Marital Dysfunction
557
S C South
Mate Selection
K A Valentine and N P Li
564
Media Influence on Behavior
B J Bushman and J L Whitaker
571
Meditation: The Science and the Art
D Shannahoff-Khalsa
576
Memory
G A Radvansky and A K Tamplin
585
Memory, Neural Substrates
R Gupta and D Tranel
593
Mental Imagery
601
G Ganis
Military Psychology
M D Matthews and J H Laurence
608
The Mind-Body Problem
615
D E Presti
The Mirror Mechanism
G Rizzolatti and M Fabbri-Destro
622
Molecular Genetics and Human Behavior
T Partridge
630
Moral Development
F C Power
637
Moral Emotions
T Malti and B Latzko
644
Motivation
O C Schidtheiss, A Strasser, A G Rosch, A Kordik, and S C C Graham
650
Motor Control
S P Swinnen
657
Near-Death Experiences
B Greyson
669
Negotiation
P 1 Carnevale and Y K Kim
677
Neuroesthetics: The Body in Esthetic Experience
V Gallese and C Di Dio
687
Neuroethics
K L Sifferd
694
Neuroexecutive Function
D S Levine
701
Neuroimaging of Dementia
J C Masdeu
707
Neurotechnologies
A Kayser and M D'Esposito
718
Neuro theology
A B Newberg
724
xl Contents
Nonverbal Communication 731
M L Patterson
Obsessive-Compulsive Disorder 739
K R Brown, M A Stanley, and T Bjorgvinsson
Obsessive- Compulsive Personality Disorder 746
M E Owens and D C Beidel
Operant Conditioning 749
E Fantino and S Stolarz-Fantino
Organizational Behavior 757
P F Martelli, M K Stimmler, and K H Roberts
Our Cognitive Map 765
S Bell
VOLUME 3
Pain 1
Y Lampl
Paranoia 14
A Fenigstein
Parenting 21
D Laible and / Eye
Parent-Offspring Conflict 28
M Goodman, S F Waters, and R A Thompson
Peer Relationships and Influence in Childhood 34
K L McDonald and K H Rubin
Perceived Control 42
/ G Chipperfield, R P Perry, and T L Stewart
Perceptual Development 49
L A Thompson and G L Strosser
Personal Relationships in Everyday Life 57
S Duck
Personality Assessment 61
R E McGrath
Personality Development 68
M B Donnellan and R W Robins
Personality Development and Aging 74
M Allemand and R Lehmann
Personality Disorders 82
C M Weaver, C Mok, and R G Meyer
Personality, Structure 90
K Naragon-Gainey and D Watson
Persuasion 96
J Barden and R E Petty
Phobias 103
S E Cassin, J H Riskind, and N A Rector
Phonetics 109
P Case and B Tuller
Contents xli
Planning
M V Pezzo, 1 A Litman, and O Wilder
119
Play
A D Pellegrini
126
Pornography
T Kohut and W A Fisher
132
Positive Psychology
S M Schueller
140
Posttraumatic Stress Disorder
M B Powers, N Nayak, M C Capozzoli, and E B Foa
148
Preference fudgments (Individuals)
E R Igou, F van Dongen, and WAP van Tilburg
153
Prejudice, Discrimination, and Stereotypes (Racial Bias)
D M Marx and S 1 Ko
160
Premenstrual Syndrome (PMS)
S T Sigmon, J Craner, K L Yoon, and G L Thorpe
167
Primate Cognition
A E Parrish and S F Brosnan
174
Problem Solving
R E Mayer
181
Psychodynamic Psychotherapy: Theory and Practice
G O Gabbard and F Rachal
187
Psychological Predictors of Heart Disease
H S Friedman and M L Kern
195
Psychology and Religion
R W Hood Jr.
201
Psychology of Reading
C A Weaver and A E Holmes
209
Psychology, Science, and Astrology
I W Kelly, B L Janzen, and D H Saklofske
218
Psychopathology: Diagnosis, Assessment, and Classification
A M Epp, K S Dobson, and D Pusch
225
Psychophysics
F A A Kingdom
234
Psychotherapy
D Pusch, K S Dobson, and P J O'Brien
240
Reading and Phonological Processing
D Shankweiler
249
Reasoning
R B Ricco and W F Overton
257
Retirement
H L Sterns, J A Headley, and J K Kreuz
265
Risk-Compensating Behavior
M Thayer and C Garbacz
271
Risk-Taking Behavior (Young Male Syndrome)
M D Baker
276
xlii Contents
Savant Syndrome
F Happe and G L Wallace
281
Schizoid and Schizotypal Personality Disorder
P Roussos, E Chemerinski, and L J Siever
286
Schizophrenia
D R Combs , / Waguspack, and M R Basso
295
Selective Mutism
M M Nieves, F Mesa, and D C Beidel
302
Self-Defeating Behaviors
R C Curtis
307
Self-Efficacy
M W Gallagher
314
Self-Esteem
D R Kille and J V Wood
321
Self-Fulfilling Prophecy
R Rosenthal
328
Self-Regulatory Skills and Behavior Change
A Luszczynska and A B Durauia
336
Self-Relaxation
J C Smith
343
Semantic Memory
H E Schendan
350
Sensation Seeking
J H Patton
359
Sense of Smell
R L Doty
366
Sense of Taste (Effect on Behavior)
B P Daly, M P Daly, N Minniti, and J M Daly
373
The Sense of Touch
S Bensmaia and L Manfredi
379
Sentence Processing
E-K Lee and D G Watson
387
Separation Anxiety
M P Mychailyszyn, P C Kendall, ] Domingues, and K Treadwell
396
Sex Differences
D C Geary
403
Sex Roles
j D Sinnott and J S Rabin
411
Sexual Behavior
J D Baldwin and J I Baldwin
418
Sign Languages
J E Pyers
425
Sleep, Biological Rhythms, and Performance
K Cote
435
Sleep Disorders
A Culebras
442
Contents xliii
Social Anxiety Disorder
S G Hofmann, C A Gutner, and A Fang
450
Social Cognition
C Hunt, E Borgida, and H Lavine
456
Social Comparison
E A Pomery, F X Gibbons, and M L Stock
463
Social Development (Attachment, Imprinting)
R Goodvin and B A Sarb
470
Social Exchange
K S Cook and A Gerbasi
479
Social Loafing (and Facilitation)
S J Karau
486
Social Support
S J Lepore
493
Social Values (Influence on Behavior)
7 Beggan and S Allison
497
Socioemotional Development
R A Thompson and E A Virmani
504
Somatoform Disorders
E A Fors, T C Stiles, and P C Borchgrevink
512
Spatial Orientation
7 7 Rieser, G Narasimham, and A Erdemir
519
Spatial Perception
M E Beriyhill, C Hoelscher, and T F Shipley
525
Stress and Blood Pressure Dysregulation
W Gerin and M J Zawadzki
531
Stress and Illness
T R Schneider and R R Riffle
536
Studying
M Rabinowitz and j Z Ingber
540
Subjective Culture
H C Triandis and M Gelfand
544
Subliminal Perception
P K Smith and K C McCulloch
551
Suicide
D Lester and K Krysinska
558
Surprise
R Reisenzein, W-U Meyer, and M Niepel
564
Synesthesia
7 Simner
571
Syntax
77 Lasnik
578
Temperament and Individual Differences
7 Kagan
589
Terrorism
A F Lemieux and M J Boyle
595
xliv Contents
Test Behavior
604
F M Zaromb, J Bumis, and R D Roberts
Tickling
611
C R Harris
Touch
616
C Spence and A Gallace
Uncertainty
621
M Smithson
Violence
629
E K Englander and P Snell
Visual Motion Perception
637
S Grossberg
Visual Neglect
652
P W Halligan and P Bartolomeo
Visual Perception
665
L de-Wit and J Wagemans
Vocational Choice
672
G W Peterson and I G Lem,
War
683
/ H Liu, M W Allen, and D Paez
Word Retrieval
690
L E James and S Oberle
Work Efficiency and Motivation
697
E B Goldsmith
Index
703
Academic Achievement
B Spinath, Heidelberg University, Heidelberg, Germany
© 2012 Elsevier Inc. All rights reserved.
Glossary
Big-fish-little-pond effect This effect refers to the fact that
the evaluation of students' performance differs depending
on the average achievement of the class. The same
achievement will be perceived as better and will get a better
grade when the average achievement of the class is low and
will get a poorer grade when the average achievement is
high. This effect is observed in teacher grades and makes it
difficult to compare grades from different classes or schools.
Carrier variable A carrier variable is a factor that is just an
indicator or proxy for other, more specific variables.
A typical example for a carrier variable is age, which is not in
itself the cause of certain developmental events but
encompasses psychological and biological processes that
co-occur with age. Another example for a carrier variable is
socioeconomic status which does not per se influence
something else but is an indicator of education, intelligence,
educational aspirations, and the like.
Organisation for Economic Co-Operation and
Development (OECD) OECD is a union of 30 countries
around the world who are committed to democracy and the
market economy. The OECD provides their members with
opportunities to compare policy experiences, seek answers
to common problems, identify good practice, and
coordinate domestic and international policies. Among
other activities, the OECD collects data about educational
systems and outcomes and commissions international
comparative studies of scholastic achievement.
Socioeconomic status (SES) SES of a person or
family indicates their possession of economic and
social capital or resources. Most frequently SES is
assessed by occupation, education, or income or a
combination of all.
Standard deviation The standard deviation is a statistical
parameter that characterizes the variability within a
population or sample. It is the average difference of scores
from the mean of the distribution. Together with the
arithmetic mean, the standard deviation describes the
distribution of scores. A small standard deviation indicates
that the majority of scores scatter closely around the
arithmetic mean whereas a large standard deviation
indicates that the scores spread out widely.
Academic achievement is the general term for performance
outcomes in intellectual domains taught at school, college, and
university. By confining the definition to intellectual domains,
fields such as music and sport are excluded for which certain
motor abilities are more important than intellectual abilities.
Academic achievement indicates the level of intellectual educa-
tion of a person, a group, or a whole nation.
This article begins with pointing out the importance of
academic achievement for individuals, societies, and as a field
of psychological research. After this, it looks at different indi-
cators for and the measurement of academic achievement.
An important question refers to the prerequisites or predictors
of academic achievement. Because of their special significance
for academic attainment, this article focuses on individual char-
acteristics, specifically intelligence, motivation, and personality.
Finally, we deal with the issue of equal opportunities in educa-
tion with regard to gender and socioeconomic status (SES).
Importance of Academic Achievement
The importance of academic achievement can be viewed from
different perspectives. Here, we distinguish an individual and a
societal perspective as well as the perspective of psychological
and educational research.
On an individual level, academic achievement is the most
important predictor of vocational careers and, therefore, indi-
vidual socioeconomic prosperity. This is true because school
grades and scholastic achievement tests are used as selection
criteria for jobs and higher education. The strength of the
association between academic achievement and indicators
of life success is moderate (r= 0.40-0.50). In undeveloped
societies that do not use such selection criteria or have no
formal education the association between education and
individual life success is much weaker or nonexistent. Higher
academic achievement opens choices as to what university
and course of study one enrolls in and, later, what job one
pursues, where, and with whom. These greater numbers of
choices make it more likely that a person finds working con-
ditions that fit his/her needs and contribute to personal well-
being. At the same time, a higher level of education comes
along with increased aspirations and expectations, so that
higher academic achievement is not necessarily associated
with higher life satisfaction.
On a societal level, the same is true as on an individual level:
academic achievement is the most important prerequisite for
societal prosperity. The more educated a society is, the higher
the chances for a positive socioeconomical development. Because
of this established association, the OECD (Organisation for Eco-
nomic Co-Operation and Development) initiated worldwide
comparative studies on scholastic achievement and provides
their members with information to monitor their educational
systems. National administrations use the results of international
scholastic achievement studies to analyze sttengths and weak-
nesses of their educational systems. These large-scale studies
have led to an empirical turn in educational policy. Politicians
in many countries ground their decisions on data from school
achievement studies.
1
2 Academic Achievement
The high individual and societal importance makes aca-
demic achievement one of the most investigated issues in
psychological and educational research. With the aim of
finding ways to foster academic achievement, research tries to
identify central prerequisites of learning. On the side of the
individual learner, research focuses on characteristics such as
intelligence, motivation, and personality. On the environmen-
tal side, research investigates how characteristics of the learning
conditions to which the learner is exposed (e.g., teaching prac-
tices, media use) and characteristics of the overall educational
system (e.g., early, late, or no separation into educational tracks)
influence academic performance. Besides fostering learning and
performance, one further aim of research on academic achieve-
ment is to close gaps between groups of learners and, thus,
contribute to equal opportunities and a fairer educational sys-
tem. Much research is done on the issue of the gender gap in
education as well as on the gap between learners from different
socioeconomic backgrounds.
Taken together, academic achievement is the most impor-
tant prerequisite for personal and societal prosperity. More-
over, education empowers people to participate in and shape
their cultures and societies which is vital for democracy. There-
fore, it is an important issue of psychological and educational
research to look for the prerequisites of education and ways to
improve them. The overall aim is to maximize each learner's
academic achievement according to their potentials. Interven-
tions may either focus on individual prerequisites or on char-
acteristics of the learning environment and educational system.
Indicators of Academic Achievement
Academic achievement can be measured in two different ways.
On the one hand, there are grades and educational degrees. On
the other hand, academic achievement can be measured by
standardized achievement tests. Although both types of mea-
surement capture mostly the same underlying characteristic,
that is, educational achievement, they are not completely
equivalent. It is important to understand the difference in
what grades and standardized achievement tests measure
because findings of empirical studies vary depending on the
indicator of academic achievement.
Grades
Academic achievement in schools and higher education insti-
tutions is typically evaluated and quantified by grades. The
grade point average (GPA) is the arithmetic mean of all grades
that have been received during a certain time. Grades are
ecologically valid measures of academic achievement because
allocation and selection decisions for higher education and job
positions are, to a large extent, based on grades. This makes
grades a very important issue for psychological research.
The following box illustrates the functions of grades. As can
be seen, grades are supposed to fulfill many different func-
tions. Some of these functions demand that grades measure
students' performance as accurately as possible (the functions
of feedback, information, and allocation). Other functions
might be fulfilled better if grades diverge slightly but
systematically from the actual performance level. For example,
when teachers want to reward an increase in performance and
motivate students, grades will not reflect the exact performance
but will be slightly better. To evaluate the quality of grades,
all different functions need to be taken into account. An
overall judgment whether grades fulfill their functions also
depends on the weight that is given to each function. A person
who emphasizes the motivation function of grades will
evaluate the quality of grades differently from a person who
puts emphasis on the allocation function. This example illus-
trates that the question of how good grades fulfill their func-
tions cannot be answered by empirical studies alone but that
the answer depends on ideological points of view.
Functions of Grades
1. Give students feedback
• Information about strengths and weaknesses
• Familiarize with performance evaluations
2. Motivate students
• Good grades are incentives to continue the good work
• Poor grades are incentives to improve
3. Discipline students
• Lack of discipline is reflected by poor grades
• Some grades specifically evaluate learning behavior
4. Inform third parties
• Parents, employees, and others get information about students'
performance
5. Allocate, select, and legitimize
• Grades are criteria for the allocation of desired resources (e.g.,
university admission, scholarships, jobs)
Much research has been devoted to the question of whether
teachers' grades actually measure students' performance. The
most convincing evidence for the validity of grades is their
strong association with standardized achievement tests. Typi-
cally this correlation is between r = 0.60 and 0.70. Another
strong evidence for the validity of grades is that they are the
best predictors of future success in academic domains. For
example, high school GPA is the best known predictor of
college GPA. This is true although across classes and schools
different grading standards are used. A major problem of
grades is that they depend on the average achievement level
of the class. When teachers evaluate students' performance,
they use a standard that is adapted to the overall achievement
level of the class. The same student will receive better grades in
a low performing class and poorer grades in a high achieving
class. This phenomenon is also known as a frame-of-reference
effect, specifically the big-fish-little-pond effect.
Despite this known distortion of grades due to teachers'
different grading standards, school grades are often found to
outperform standardized achievement tests in the prediction of
future success. This can be explained by the fact the GPA is a
highly aggregated measure of performance collected from dif-
ferent teachers, over different subjects, and over a relatively
long period of time. Many different single measures of student
performance make the GPA very reliable. The fact that grades
capture not only pure intellectual capacity but also motiva-
tional and personality aspects is not necessarily a disadvantage
but contributes to the predictive validity of the GPA for future
achievements.
Academic Achievement 3
Educational degrees
Educational degrees are another type of indicator of academic
achievement. Educational degrees depend directly on the
grades accumulated over the educational career. They are the
most important prerequisite for admission to higher education
and job positions. The first attempt to collect international
data about educational outcomes was to compare graduation
rates and educational degrees among nations. Because of the
differences in the educational systems, such comparisons did
not turn out to be very meaningful. Today, educational out-
comes among nations are compared by means of standardized
achievement tests. Nevertheless, graduation rates and educa-
tional degrees are also monitored by international studies on
educational achievement because they are important ecologi-
cally valid indicators of educational outcomes. For example,
the annual study Education at a Glance, commissioned by the
OECD, reports international comparisons of indicators of the
quality of education, such as who participates in education,
what is spent on it, and how many students graduate in basic
and higher education.
Standardized Achievement Tests
Standardized academic achievement tests measure abilities
that are acquired in educational institutions or are important
for future success in academic and other domains. Standardized
achievement tests vary with regard to the degree to which they
are curriculum based. A test that is meant to assess knowledge
and skills that have been acquired at school has to be curriculum
based, that is, include contents taught at school (e.g., SAT II
Subject Tests). Other achievement tests are not curriculum
based but assess whether certain competencies have been
acquired at a certain age and whether certain criteria are met,
that is, certain tasks are mastered. A well-known example of a
standardized achievement test that is not curriculum-based is the
SAT I Reasoning Test (formerly Scholastic Aptitude Test and
Scholastic Assessment Test). The SAT I is designed to measure
students' general ability or aptitude for learning. Many colleges
in the United States of America base their admission decisions
on SAT scores in combination with grades. Because there are
wide differences between secondary schools in the United States,
tests like the SAT are needed to compare applicants against a
common standard.
Different from grades, the only function that standardized
achievement tests have is to measure students' achievement as
purely as possible. This means that other factors that influence
teachers when grading students' performance, such as how
much effort a student invested, or whether there is a positive
trend that is worth rewarding, must not influence the results of
an achievement test. Therefore, standardized achievement tests
are a purer measure of achievement compared to grades.
Other examples of standardized achievement tests are the
measurements within international large-scale studies of edu-
cational achievement (see following box). Typically these stud-
ies assess domain-specific abilities, which means that they have
to be curriculum based to some extent. Nevertheless, the extent
to which international large-scale assessments of scholastic
achievement rely on the curriculum depends on the domain
in question. In domains in which knowledge is especially
dependent on school-based learning, such as in mathematics
and science, the tests are more strongly curriculum based.
In other domains, such as languages, learning is relatively
more influenced by experiences outside of the school, so that
tests for reading comprehension need not be valid according to
school curriculum to assess students' reading competencies.
International Large-Scale Assessments of School
Achievement
Since the 1950, the OECD collects data about the educational systems and
outcomes of their member countries. The data collected in these studies do
not only guide educational policies but have made outstanding contributions
to the scientific knowledge about academic achievement. This is both due to
enormous sample sizes, which provide a solid basis for conclusions, and
sophisticated methods that have been advanced in the context of these studies.
In the following, three of the largest and most important international
large-scale assessments are listed and characterized by who and what is
investigated:
PISA (Programme for International Student Achievement)
• Who: 15-year-olds
• What: Literacy in different domains (mother language, mathematics,
science) and cross-curricular competencies (e.g., problem solving)
TIMSS (Third International Mathematics and Science Study)
• Who: Elementary school children, 7th— 8th grades and students in
last year of mandatory education
• What: Literacy in mathematics and science
PIRLS (Progress in International Reading Literacy Study)
• Who: 4th graders
• What: Mainly Reading Literacy
Do standardized achievement tests measure academic
achievement or intelligence or both?
A comparison between results of large-scale assessments of
school achievement and intelligence tests indicates a close
correlation between the two. This is true both on the level of
the individual student as well as on the level of nations. This
imposes the question of whether large-scale studies such as PISA
actually measure school achievement or whether, instead, they
measure intelligence. This is an important question because
if it were not mainly school-based learning outcomes that
were measured but intelligence, then PISA and the like would
not assess the quality of educational systems and would not
inform countries about shengths and weakness of their schools.
Because intelligence is known to have a strong genetic basis,
which means that a large portion of the differences observed in
intelligence can be traced back to different genetic endowments,
intelligence is often considered to be a relatively stable character-
istic. If international large-scale studies of school achievement
measured a stable trait instead of educational outcomes, their
results would be more of a verdict about which nations are smart
and which are not and not a means to assure educational quality.
This line of argumentation is flawed in several ways. First,
because of the definition of what intelligence is, it is not
reasonable to assume that any complex cognitive performance,
such as academic achievement, can be measured without also
measuring intelligence. Second, intelligence tests and standar-
dized achievement tests share many methodological character-
istics which contribute to a strong association. Therefore, the
close correlation does not only reflect shared contents but also
shared methods. Third, if the correlation among intelligence
4 Academic Achievement
tests and school achievement tests are considered to be too
high to justify two distinct constructs, this does not necessarily
compromise the concept of academic achievement as measured
in large-scale assessment but raises questions about whether
intelligence tests function as they are supposed to. It might be
argued that extant intelligence tests rely too heavily on school-
based learning contents and fail to measure an underlying cog-
nitive potential. Fourth, it is wrong to assume that intelligence is
heritable but school achievement is not or less so. Intelligence
and student achievement are heritable to a comparable extent.
Fifth, there is a widespread misunderstanding according to
which heritability means that something is not malleable.
Because heritability is a characteristic of the population but not
of a single individual, high heritability does not inform us about
the limits of individual growth of intelligence or academic
achievement given an optimal education.
In view of these considerations, it becomes apparent that
the discussion about what exactly is measured with standar-
dized achievement tests is an important one. After all, the
answer to be found might be less critical for the measurement
of school achievement but more so for the measurement of
intelligence.
Standards-based assessment of students’ achievement
Standards-based assessment is a special kind of standardized
achievement testing. It has all the advantages of standar-
dized tests and their results can be used instead of or as a
supplement to traditional teacher grades. The emergence of
standards-based assessments is closely related to international
large-scale assessments of school achievement. As a consequence
of the feedback from international comparative assessments,
many countries have changed their educational policies from
an input orientation to an output orientation. Input orientation
means that the main instrument to assure the quality of educa-
tion is a detailed description of the contents to be taught (i.e.,
the curriculum). In conUast, an output-oriented education puts
emphasis on the competencies to be acquired and regularly
evaluates these competencies by standardized tests. Because
these standardized tests are developed with an aim to assess
competencies according to educarional standards, these tests
are called standards-based assessments. Different than for
grades, standards-based assessments are usually not or at least
not only evaluated by the same teacher who taught the class.
Standards-based assessments have several advantages.
Because the evaluation of students' achievement depends on a
common, typically nation-wide standard, results of standards-
based assessments are comparable across different classes and
schools. Different from grades, standards-based assessments do
not depend on the general level of achievement in the class. This
makes allocation decisions on the basis of standards-based
assessments fairer than decisions made exclusively on the basis
of grades. Moreover, the evaluation by common standards can
harmonize teachers' evaluations across different classes and
schools and therefore make teacher-given grades better in terms
of comparability. Furthermore, it is argued that standards-based
assessments foster students' motivation and their sense of justice.
This should be so because standards are more transparent than
the evaluation by grades and students often perceive good grades
as confined to only a few students. Because standards are based
on task mastery rather than social norms, the notion is conveyed
that every student who reaches these criteria can get good evalua-
tions. Finally, standards-based assessments evaluate not only
students' performance but also the quality of the work of teachers
and schools. If two schools, that are equally funded and whose
students come with comparable prerequisites (e.g., SES) differ
with respect to their results in standards-based assessments, then
the conclusion might be drawn that these different outcomes are
a result of better and poorer teaching. Therefore, standards-based
assessments also fulfill the function to evaluate schools and to
make the quality of schools comparable.
One important danger that is associated with standards-
based assessment is the phenomenon called teaching-to-the-
test. Once teachers realize what is measured in standards-based
tests, they might focus too much on conveying only the
knowledge and skills measured by the test and neglect other
important areas of learning. Because the tests can only assess
competencies that are testable by means of paper and pencil
assessments, there is the threat that only these kinds of con-
tents will be taught. As a matter of fact, it is especially difficult
to measure cross-curricular competencies, such as complex
problem solving or social competences in a group-testing situ-
ation with paper and pencil. Therefore, teaching-to-the-test
poses a serious problem for standards-based assessment.
Predictors of Academic Achievement
In the following section, we take a look at those factors known
to be most closely related to differences in academic achieve-
ment. Because most of the underlying research uses designs
that do not allow for unanimous causal interpretations, these
factors are called 'predictors' rather than determinants or influ-
ences of academic achievement.
Individual Student Characteristics
Individual student characteristics are by far the most important
predictors of differences in academic achievement. Taken
together, individual characteristics such intelligence, motiva-
tion, and personality explain about half of the differences in
scholastic performance. Among these individual characteris-
tics, differences in intelligence are responsible for half of the
differences in scholastic achievement. This makes intelligence
the most important single predictor of academic achievement.
Intelligence explains about 25% of the differences in school
achievement. The strength of association between intelligence
and scholastic achievement differs depending on whether
grades or standardized tests are used as indicators. Intelligence
is more closely correlated with results from standardized tests
than with grades.
Knowing the history of intelligence testing, the strong asso-
ciation between intelligence and school achievement is unsur-
prising. Intelligence tests have been developed with an aim to
assess children's cognitive potential and guide decisions about
the most adequate schools and treatments. Therefore, every
new intelligence test has to prove its usefulness and validity
by being a good predictor of school performance. But of course
these associations cannot and must not be perfect because
in such a case the test would not measure anything besides
scholastic achievement. Instead, intelligence tests are meant to
Academic Achievement 5
assess the cognitive potential of a person largely independently
from educational influences. Therefore, intelligence tests use
tasks that should be largely free of prior knowledge and school-
based learning, such as anagrams, matrices, and the like. Never-
theless, intelligence tests are not completely free of educational
influences and any measurement of intelligence always assesses,
to some extent, educational influences. It has been shown that
attending school has positive effects on intelligence. Therefore,
intelligence and academic achievement are closely related not
only because intelligence influences academic achievement but
also because academic achievements foster intelligence.
Besides cognitive abilities, personality plays an important
role for academic achievement. Among the famous Big Five
factors of personality (extraversion, neuroticism, openness to
experience, agreeableness, and conscientiousness), for some
time now, the leading model to describe personality, conscien-
tiousness has proved to be the best predictor of academic
achievement. The other factors of personality show only
weak or no association with academic achievement. A person
who is conscientious can be described as planned, organized,
achievement-oriented, ambitious, dependable, and persistent -
all of which contributes to better scholastic achievement.
Although the relationship between conscientiousness and aca-
demic achievement is consistently shown, it is far weaker than
the relationship between cognitive abilities and performance.
One reason for this is that the Big Five are meant to describe
personality in all kinds of contexts and are not specifically
designed to explain achievement-related behavior.
Besides these broad personality factors, psychologists who
are especially interested in learning and achievement have
identified personality dimensions that are more specifically
important in achievement situations. These personality con-
structs can be subsumed under the term achievement motiva-
tion. Some of the most investigated constructs of achievement
motivation are the need for achievement, ability self-concepts,
self-efficacy, task values, interest, intrinsic and extrinsic moti-
vation, and goals. Overall, there is a moderate association
between the expression of these motivational constructs and
academic achievement. Motivational characteristics are by
and large independent of intelligence and are often found to
contribute to the explanation of academic performance over
and beyond cognitive ability. The strongest relationship is con-
sistently found between domain-specific ability self-concepts
and academic achievement. This association is often as strong
as the association between intelligence and academic achieve-
ment. The relationship between interest or intrinsic motivation
and academic achievement is moderately close. This might be
surprising because both researchers and practitioners are often
convinced that these motivational factors are of paramount
importance for performance. It needs to be taken into account
that in achievement contexts, learners are often not free to
choose whether to engage in a subject or task but feel a pressure
to do well regardless of their interest or intrinsic motivation.
This explains why interest and intrinsic motivation are often
to be found less predictive of academic achievement than
expected.
Besides cognitive ability and personality, there is much
research on the role of students' study behavior and attitudes,
often subsumed under the term self-regulation abilities. Sur-
prisingly, empirical studies often showed no or only weak
associations between the use of learning strategies and learning
outcomes. A recent meta-analysis tries to clarify which study
habits, skills, and attitudes actually are associated with better
outcomes and what might be the reasons why so many studies
fail to show these effects. For example, study habits, skills, and
attitudes have been shown to be better predictors of college
achievement than of earlier school achievement. Nevertheless,
even in college the association between self-regulated learning
strategies and achievement is weak.
Classroom Characteristics
Surprisingly few classroom and teacher characteristics show
consistent positive associations with academic achievement.
This is primarily due to difficulties separating single character-
istics and investigating their effects. For example, regular and
private schools often not only differ with respect to one single
characteristic but in several respects (e.g., applied grading stan-
dards, teacher education and selection, pedagogical approach,
financial, and other resources). Therefore, it is impossible to
trace differences in students' academic performance back to
one single classroom or teacher characteristic. A second reason
is that different teaching styles might lead to similar results
so that there is not just one way to enhance students' per-
formance. Teachers might impress their students with their
subject knowledge or might place great value on good student-
teacher relations and both approaches might serve similarly
to motivate students for school-based learning.
Nevertheless, there is one cluster of classroom or teacher
characteristics that is consistently connected with differences in
academic achievement: Students who attend a class with a high
level of disciple and strict classroom management have better
learning outcomes. A high level of order in the classroom
allows teachers to spend more time-on-task and on contents
rather than on nonsubject-related matters. Time-on-task is the
single-most important classroom characteristic that explains
differences in school achievement. Teachers who have good
classroom management skills provide clear conduct rules and
consistently sanction positive and negative student behavior.
In well-managed classrooms, teachers formulate clear assign-
ments so that students know what to do and do not lose time
solving uncertainty. All of these characteristics contribute to
better student learning and achievement.
Educational System Characteristics
International large-scale studies on student achievement pro-
vide solid data regarding the association between characteris-
tics of the educational system and student achievement.
However, the same problems that apply to comparisons of
classrooms also apply to the level of the educational system.
Educational systems usually differ in many aspects so that it is
difficult to trace better student achievement back to one single
characteristic of the educational system. Often only the combi-
nation of different characteristics goes along with enhanced
achievement. For example, Finland - one of the leading
countries with regard to student achievement - does not sepa-
rate students into different tracks. Children with and without
special needs are taught in the same class. Copying this charac-
teristic would make no sense without also copying the small
6 Academic Achievement
class sizes and good student-teacher relation with extra special
needs teachers. Neither a comprehensive school nor small clas-
ses are per se associated with better student performance; but in
the right combination, these characteristics make a difference.
On the level of the educational system, there are some
characteristics that are consistently associated with better
mean levels of academic achievement. One of these is the
percentage of children attending kindergarten and preschool:
the more children get preschool education, the better a nation's
mean level of student achievement later in school. Institutio-
nalized preschool education is especially important for chil-
dren who lack a stimulating environment at home. Preschool
education helps to compensate unequal prerequisites and to
detect special needs or other ability deficiencies, such as dys-
lexia or language problems, for which special aid can be given
even before school age.
A second characteristic of the educational system that is
consistently associated with better student achievement is the
use of high-stakes tests, such as central exams and standards-
based assessment. There are many ways to explain this associa-
tion. One can argue that with a national curriculum and
curriculum-based tests teachers have better goal clarity and
place more importance on teaching what students need to
know to do well on the tests. In its extreme, this goal clarity
might turn into teaching- to -the- test, which is undesirable,
because competencies not assessed in the test, that still might
be important, are neglected. Therefore, teaching-to-the-test is a
serious danger that comes along with high-stakes testing.
Another reason for the association between use of high-stakes
tests and better school achievement might be that countries
employing high-stakes tests place more importance on educa-
tion and invest more effort in raising the educational level.
Establishing high-stakes testing is usually a consequence of a
country's participation in international large-scale studies of
academic achievement and the wish to monitor educational
outcomes more closely. Therefore, a plausible explanation for
the association between high-stakes testing and student
achievement is that nations more concerned with the quality
of education make more use of high-stakes tests. The last reason
might be that the use of high-stakes tests trains students to do
better on standardized achievement tests as used in interna-
tional scholastic achievement studies. For example, the typical
type of tasks and answering formats in high-stakes tests and
standardized achievement tests are very similar, so that regularly
taking high-stakes tests at school will train students to do better
on standardized achievement tests.
Besides the mean level of student achievement, a second
important indicator of educational quality is the extent of
differences between students' achievement. It is desirable that
students' achievement is not only high on average but also that
the differences between students' achievement are not too
large. This can be explained by the example of early tracking.
One characteristic of the educational system that has been
shown to enlarge differences between students is an early
separation of children into educational tracks stratified for
level of achievement. Empirical evidence indicates that in
countries with an early separation into educational tracks, the
standard deviation of scholastic achievement as measured with
standardized achievement tests increases from elementary
school to secondary school. In countries with late or no
separation into educational tracks, the standard deviation
decreases over the school years. Another undesirable effect of
earlier tracking is a sttonger association between children's
school achievement and family background characteristics
such as SES. For example, Germany is one of the few countries
that separate students as early as grade 4. In Germany, the upper
and lower 5% of the achievement spectrum show the greatest
divide compared to all other nations. At the same time, in
Germany the association between children's school achieve-
ment and their families' SES is the sttongest. These results
from international school achievement studies have led to a
discussion to abolish the early separation either by prolonging
elementary school or by merging different types of secondary
schools.
Equal Opportunities and Academic Achievement
One important aim of psychological research on academic
achievement is to shed light on reasons for differences in
education that come along with characteristics that, per se,
have no bearing on education. As two examples for such char-
acteristics, we pick gender and SES.
Gender Differences
At the end of the twentieth century, in many countries it is girls
and not boys who earn better school grades and higher educa-
tional degrees. The gender gap in educational achievement is
both a theoretically challenging and practically relevant topic.
A connected question is why girls are not using their educa-
tional advantages and transferring them into higher achieve-
ments later in life. Women are still unrepresented in higher job
positions and earn less than their male colleagues.
A variety of reasons for the gender gap in education have
been put forward. The influences of changing cultures on
children's educational outcomes are visible in similar trends
taking place throughout the developed world: the more devel-
oped a country is and the more it aspires equal opportunities
for both sexes, the better girls are at school. For example, in
rural areas of China, girls are still found to do worse than boys
in school. In Hong Kong, girls outperformed boys in school
only after a Western outlook had been adopted in addition to
the traditional Chinese culture. Evidence from international
large-scale studies shows that the gender gaps in math and
reading are associated with the degree of women's emancipa-
tion that is realized in a country. For example, in Scandinavian
countries, which score high on women's emancipation, the
gender gap for mathematics is smaller and in some cases even
nonexistent, whereas in reading, girls have an even bigger head
start than on OECD average. In contrast to this, in countries
that have no culture of women's emancipation, such as Turkey,
girls do markedly worse in math than boys and their advantage
in reading is comparably small. Thus, girls' better school
achievement is related to societal changes in attitudes toward
equal opportunities for men and women.
Besides such societal influences, gender differences in indi-
vidual characteristics, such as cognitive ability, personality, and
motivation, have been proposed as explanations for the gender
gap in academic achievement. Gender differences in cognitive
Academic Achievement 7
ability can be excluded as an explanation for girls' better aca-
demic attainment for several reasons. First, most researchers
agree that there are no gender differences in general intelli-
gence. If differences in general intelligence are found, then they
are in favor of boys. Second, gender differences in specific
intelligence components are too weak to explain the observed
differences in school achievement. It is often found that girls
have an advantage in verbal intelligence and boys' in numeric
intelligence. Nevertheless, these ability differences are too
small to explain the observed differences in school achieve-
ment. Third, even when differences in intelligence are statisti-
cally controlled, girls have been shown to get better grades.
Therefore, differences in cognitive ability are not the explana-
tion for gender differences in academic achievement.
The most promising candidates that explain gender differ-
ences in academic achievement are gender-typical differences
in personality and motivation. Girls are typically more intrin-
sically motivated for school-based learning. Boys, on the hand,
show on average lower school-related learning motivation and,
at the same time, have fewer self-control resources to counter-
balance their low motivation. Their higher self-control allows
girls to regulate their emotions and behavior, and respond to
demands at school even if they are not interested in what they
have to do. Moreover, whereas in girls' peer groups it is abso-
lutely acceptable to be good at school and conform to school
rules, boys' peer groups rather reward oppositional behavior
and sanction boys who do too well at school. Another assump-
tion is that girls succeed in school because of their more
agreeable personalities. Teachers are known to reward agree-
able behavior with better grades. Taken together, these differ-
ences in girls' and boys' personality are important explanations
for girls' better school achievement.
If a better fit of girls' personality and motivation to the
school environment is the reason for girls' advantages in aca-
demic achievement, then there are two possibilities to enhance
boys' achievement: either change boys' characteristics or
change the school environment. According to the first alterna-
tive, interventions that enhance boys' intrinsic motivation for
school-based learning and foster their self-regulatory compe-
tencies are a viable way. According to the second alternative,
the school environment needs to be cleared of such character-
istics that unfairly favor girls but do not actually foster or assess
achievement. For example, if grades are used to reward agree-
able behavior, then teachers need to be trained to keep their
evaluations free of such influences and find other ways to
ensure disciple in the classroom. Another way to ensure that
boys conform to school rules is to sanction violations of rules
consequently and strictly. If boys do not have the chance to
violate rules, then the differences between girls' and boys'
behavior become smaller and might, as a consequence, dimin-
ish differences in academic achievement. At this time, these
suggestions are not tested empirically. More research is needed
to see whether these are viable ways to decrease the gender gap
in school achievement.
Socioeconomic Status
The SES of a person, family, or other social group indicates
their possession of economic and social resources. Most fre-
quently, SES is assessed by occupation, education, income, or a
combination of all. It is a general finding that the SES of a family is
associated with children's academic achievement and their high-
est educational degrees. According to meta-analyses and interna-
tional large-scale studies of school achievement, the strength of
the association between SES and children's academic achieve-
ment is moderate (r= 0.30-0.40). The strength of association
reported varies markedly depending on the indicators used to
measure SES and academic achievement, age of children, and
whether the full range or a restricted range of SES and achieve-
ment was investigated.
Because SES has per se no bearing on a child's achievement,
the association between SES and academic achievement is com-
monly considered as evidence for social injustice and unequal
opportunities. The most convincing evidence that SES actually
is associated with injustice comes from studies showing that
two students, one from a family with low SES the other from a
high SES home, have different chances to acquire higher educa-
tional degrees even if they have the same ability according to
standardized tests.
But why is the family SES associated with children's aca-
demic achievement? The way to find an answer to this question
is to look behind the label SES and find those factors that are
known to be prerequisites for academic achievement. SES itself
is just a carrier variable that is associated with other factors
that are actual causes of differences in academic achievement.
For example, it is well established that a higher SES, on average,
goes along with higher intelligence and higher aspirations
and engagement for education. Parents who possess such
favorable characteristics pass these on to their children, either
by providing them a stimulating environment or via genes.
Parents who have high aspirations for their children provide
them with a more supportive environment and, at the same
time, have children who are more likely to have high aspira-
tions themselves. It can be shown that large parts of the associ-
ation between family SES and children's school achievement
are explained by parents' and children's sharing of personal
and environmental characteristics that contribute to better
academic achievement. This is an important finding because
it shows that the association between family SES and children's
achievement is not per se an indicator of social injustice.
Instead, it indicates that desirable resources are distributed
unequally among families and that societies must make delib-
erate choices whether to compensate for different starting posi-
tions or whether to let those differences become even larger.
What can be done to avoid unequal opportunities with
regard to SES? On the level of the educational system, the
evidence is clear that early tracking increases dependability of
academic achievement on family attributes. One recommenda-
tion to reduce the association between SES and academic
achievement is to keep students together in one type of school
throughout middle school and even high school. Another
recommendation is to use not only grades for admission deci-
sions to the next level of education but also standards-based
assessments. To assess something with different measures is
always a good way to validate each single measure. On the one
hand, teachers might be influenced by their knowledge of the
families' background whereas standardized tests are not. On
the other hand, some standardized tests are even more strongly
associated with SES than grades. Therefore, it is important
to base vital decisions on more than just one source of
8 Academic Achievement
information. Again, this is a field in which more research is
needed to find the best way to produce equal opportunities
with respect to families' social and economic background.
See also: The Behavior-Genetics of Intelligence; Cognition and
Personality; Equity Theory; Genius, Eminence, and Giftedness;
Human Intelligence; Motivation; Self-Efficacy; Studying; Sex
Differences.
Further Reading
Atkinson RC and Geiser S (2009) Reflections on a century of college admissions tests.
Educational Researcher 38: 665-676.
Bartels M, Rietvelt JH, van Baal GCM, and Boomsma Dl (2002) Heritability of
educational achievement in 12-year-olds and the overlap with cognitive ability.
Twin Research 5: 544-553.
Bradley RH and Corwyn RF (2002) Socioeconomic status and child development.
Annual Review of Psychology 53: 371-399.
Ceci SJ, Williams WM, and Barnett SM (2009) Women’s underrepresentation in
science: Sociocultural and biological considerations. Psychological Bulletin
135: 218-261.
Cliffordson C and Gustafsson J-E (2008) Effects of age and schooling on intellectual
performance: Estimates obtained from analysis of continuous variation in age and
length of schooling. Intelligence 36: 143-152.
Crede M and Kuncel NR (2008) Study habits, skills, and attitudes: The third pillar
supporting collegiate academic performance. Perspectives on Psychological
Science 3: 425-453.
Duckworth AL and Seligman ME (2006) Self-discipline gives girls the edge: Gender in
self-discipline, grades, and achievement test scores. Journal of Educational
Psychology 98: 198-208.
Frey MC and Detterman DK (2004) Scholastic assessment or cfl The relationship
between the scholastic assessment test and general cognitive ability. Psychological
Science 15: 373-378.
Guiso L, Monte F, Sapienza P, and Zingales L (2008) Culture, gender, and math.
Science 320: 1164-1165.
Hanushek EA and Kimko DD (2000) Schooling, labor-force guality, and the growth of
nations. American Economic Review 90: 1 1 84-1 208.
Hanushek EA and WoBmann L (2006) Does educational tracking affect performance
and inequality? Differences-in-differences evidence across countries. Economic
Journal 116: C63-C76.
Hoge RD and Coladarci T (1989) Teacher-based judgements of academic achievement:
A review of literature. Review of Educational Research 59: 297-313.
Hyde JS (2005) The gender similarity hypothesis. American Psychologist
60: 581-592.
Johnson W, McGue M, and lacono WG (2007) Socioeconomic status and school
grades: Placing their association in broader context in a sample of biological and
adoptive families. Intelligence 35: 526-541.
Kuncel NR, Hezlett SA, and Ones DS (2004) Academic performance, career potential,
creativity, and job performance: Can one construct predict them all? Journal of
Personality and Social Psychology 86: 1 48-1 61 .
Organization for Economic Co-Operation and Development (OECD) (2004) What Makes
School Systems Perform? Seeing School Systems Through the Prism of PISA .
Paris: OECD Publications.
Organization for Economic Co-Operation and Development (OECD) (2005) School
Factors Related to Quality and Equity: Results from PISA 2000. Paris: OECD
Publications.
Organization for Economic Co-Operation and Development (OECD) (2007) PISA 2006
Volume 2: Data. Paris: OECD Publications.
Rindermann H (2007) The ^-factor of international cognitive ability comparisons:
The homogeneity of results in PISA, TIMSS, PIRLS and IQ-tests across nations.
European Journal of Personality 21 : 667-706.
Rindermann H and Ceci SJ (2009) Educational policy and country outcomes in
international cognitive competence studies. Perspectives on Psychological Science
4: 551-557.
Roberts BW, Kuncel NR, Shiner RL, Caspi A, and Goldberg LR (2007) The power of
personality: The comparative validity of personality traits, socioeconomic status,
and cognitive ability for predicting important life outcomes. Perspectives on
Psychological Science 2: 313-345.
Sirin SR (2005) Socioeconomic status and academic achievement: A meta-analytic
review of research. Review of Educational Research 75: 417-453.
Spinath B, Spinath FM, Harlaar N, and Plomin R (2006) Predicting school
achievement from general cognitive ability, self-perceived ability, and intrinsic
value. Intelligence 34: 363-374.
Weinert FE and Helmke A (1995) Interclassroom differences in instructional quality
and interindividual differences in cognitive development. Educational Psychologist
30: 15-20.
Addictions and Adolescence
K C Winters, University of Minnesota Medical School, Minneapolis, MN. USA
T Chung, Pittsburgh Adolescent Alcohol Research Center, Pittsburgh, PA, USA
R Stinchfield, University of Minnesota Medical School, Minneapolis, MN, USA
J D Kassel and M Conrad, University of Illinois, Chicago, IL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cognitive-behavior therapy (CBT) A common type of
counseling based on the belief that using both cognitive
therapy and behavioral therapy is more effective than just
one of these types. The two components of the therapy seeks
to help the client identify and change harmful or
dysfunctional thinking, emotional and behavior responses.
Diagnostic and Statistical Manual of Mental Disorders
(DSM) The DSM is published by the American Psychiatric
Association and provides a common language and standard
criteria for the classification of behavioral and mental
disorders, including the addictive disorders discussed in this
article. The current version is the fourth edition, known as
DSM-IV.
Nicotine dependence Nicotine is present in tobacco
products (e.g., cigarettes, cigars, tobacco snuff) and like
other physically addictive drugs such as alcohol and cocaine,
can lead to dependence. Signs of dependence include
frequent use, repeated urges to use, and withdrawal
symptoms when the person cuts down or stops using a
tobacco product.
Pathological gambling As defined by DSM-IV, pathological
gambling is a persistent and recurrent maladaptive disorder
that disrupts one's life as evidenced by five or more of the
following diagnostic criteria: preoccupation with gambling;
needing to gamble with larger amounts of money in order
achieve the same level of excitement; repeated unsuccessful
attempts to cut down or stop gambling; feelings of
restlessness or irritability when trying to cut back or stop
gambling; gambling to escape problems or unpleasant
emotions; chasing losses; lying to family member or hiding
evidence of one's gambling; committing illegal acts to
support one's gambling; jeopardizing significant
relationships; and relying on others to relieve a desperate
financial situation.
Problem gambling Whereas pathological gambling is a
clinical diagnosis with specific diagnostic criteria, the term
'problem gambling' is a broader constmct that identifies
individuals who exhibit some problems with their gambling
but who do not meet diagnostic criteria for pathological
gambling. This level of problem severity is sometimes
referred to as subthreshold or subclinical problem gambling.
Substance use disorders DSM-IV recognizes two
substance use disorders (SUDs) Abuse and dependence.
The two diagnoses are defined by distinct sets of symptoms.
Substance abuse is generally considered to represent a
milder SLID, compared to substance dependence, and its
symptoms focus on life disruptions and negative
consequences to the user. DSM-IV substance abuse requires
meeting 1 of 4 criteria that represent recurrent psychosocial
consequences related to substance use (e.g., school grades
dropped due to substance use, interpersonal relations
damaged due to substance use, substance-related legal
problems), and hazardous substance use (e.g., driving when
intoxicated). DSM-IV substance dependence requires
meeting 3 of 7 criteria within the same 1-year period, and
identifies a compulsive pattern of substance use despite the
continuation of negative consequences. Dependence criteria
include physical symptoms (i.e., a high level of tolerance, a
recurrent withdrawal syndrome), prioritizing substance use
over other activities (i.e., much time spent using, reducing
activities in favor of drug use), and impaired control over use
(i.e., used more or longer than intended, difficulty cutting
down or abstaining from use, use despite adverse physical
and psychological consequences of substance use). In
keeping with the notion that dependence is a more severe
condition than abuse, meeting criteria for dependence
overrides a diagnosis of abuse for a given substance.
Although there are differences across dmgs in the extent to
which certain criteria may be relevant, the same criteria are
used across all substances, with the exception that DSM-IV
does not include a nicotine 'abuse' diagnosis, and
withdrawal is not a criterion for dependence on cannabis or
hallucinogens.
Introduction
In the adolescent literature, the past 20 years have been char-
acterized by a rapid growth of research in the understanding
of developmental psychopathology and adolescent develop-
ment. This emerging research area has greatly advanced our
understanding of the extent and nature of addictive disorders
among adolescents. We recognize that addictive behaviors, as
they pertain to youth, represent a broad construct as it does
for adults. However, there are relatively large research lit-
eratures that address three broad domains of addictive disor-
ders for adolescents: nicotine dependence, other substance
use disorders, and pathological gambling. Thus we have
organized this article around the description of the three
addictive disorders from clinical, epidemiological, and treat-
ment perspectives. But first we address several developmental
considerations that pertain to the construct of addiction as
applied to adolescents.
9
1 0 Addictions and Adolescence
Developmental Issues
Historically, addictions have been defined with reference only
to the compulsive - like the use of a psychoactive substance -
disorders (including nicotine) that temporarily alter reward-
based chemicals in the brain. Many experts now consider
behavioral addictions as legitimate members of the spectrum
of addictive disorders; these include 'psychological dependen-
cies' on gambling, food, sex, and shopping.
All addictive disorders are reflected by individuals pursuing
rewards or relief by substance use or other behaviors, and are
characterized by impairment in behavioral control, craving,
inability to consistently abstain, and diminished recognition
of significant problems with one's behavior and interpersonal
relationships. Thus, the essence of addiction is the recurring
compulsion felt by an individual to engage in some specific
activity in the presence of harmful consequences to the indivi-
dual's health, mental state, or social life. Like other chronic
diseases, addiction involves cycles of relapse and remission,
and many individuals do not achieve stable recovery until after
several courses of treatment. Also, some individuals obtain
successful recovery without formal treatment.
Several issues are relevant here as we consider the impor-
tance of the context of adolescence and addictions. We briefly
discuss the following three contextual issues: etiology, identifi-
cation of clinical significance, and applicability of addiction
diagnostic criteria for adolescents.
Etiology
A traditional view in the developmental literature is that ado-
lescents may be more vulnerable than adults to addiction.
Indeed, the vast majority of adults with an addictive disorder
began their indulgence during the critical developmental
period of adolescence. Also, recent and converging evidence
suggest that the substance-based addictions develop more
quickly and with less exposure to the particular substance
during adolescence compared to the same during adulthood.
Multiple factors clearly play roles in fostering the development
of addiction, including genetics, peer influence, management
of stress, parenting, and the motivational push to explore the
world to learn to be an adult. Also, delays in social and emo-
tional functioning, diminished respect toward authority, and
tendencies to be self-absorbed and to minimize negative
consequences may contribute to involvement in addictive
behaviors.
Brain Development
One emerging view is that adolescent vulnerability to addictive
disorders is significantly influenced by the way their brain
grows. Basic and clinical data support the stance that adoles-
cence is a neurodevelopmentally critical period of great vulner-
ability for risk-taking, experimentation with substances, and
acquisition of addictive disorders.
Recent studies on this subject support the view that the
neurodevelopmental basis for this vulnerability is associated
with the maturation of a particular region in adolescent
brains. The region of the brain that monitors impulse and
motivation, the frontal cortex, is not fully formed during
adolescence (see Figure 1). Adolescent impulsivity and nov-
elty seeking as a transitional trait behavior may be explained
in part by maturational changes in the frontal cortical region.
These developmental processes may not only advantageously
promote learning drives for adaptation to adult roles, but may
also confer greater vulnerability to the addictive actions of
drugs. This brain region's quick-fire change that is relatively
unique to adolescence may translate to teenagers being more
likely than children and adults to try out new experiences and
to be more impulsive and take risks. These findings suggest
that for adolescents, addictive disorders are neurodevelop-
mental disorders.
Figure 1 Gray matter maturation. Reproduced from Gogtay N, Giedd JN, Luck L, et al. (2004) Dynamic mapping of human cortical development during
childhood through early adulthood. Proceedings of the National Academy of Sciences of the United States of America 101:81 74-81 79.
Addictions and Adolescence 1 1
Inherited Characteristics
A wealth of family and twin studies provide strong support to
the fact that genetic factors play a major role in the develop-
ment of addictive disorders. An additional support for this is
the theory that states that during childhood, the genetic liabil-
ity for addictive disorders manifests as an underlying deficit in
the young person's ability to regulate and control impulses,
often referred to as the dysregulation syndrome. This liability is
also seen as an antecedent to several behavioral disorders that
commonly coexist with addictive disorders, such as conduct
disorder, oppositional defiant disorder, and attention-deficit
hyperactivity disorder (ADHD). A related group of antecedents
to addictive behaviors that may also have a genetic basis are
negative affect disorders, such as anxiety disorders and depres-
sive disorders. It has been hypothesized that young people
suffering from these disorders engage in addictive behaviors
to alleviate the underlying negative emotions of anxiety and
depression.
Environmental Influences
Many nonbiological factors also influence the likelihood of
developing an addictive disorder, and there is converging
data indicating that these factors generalize across a range of
addictive disorders. Key environmental factors that increase risk
for an addictive disorder for youth are affiliation with delin-
quent peers, lack of involvement in prosocial activities (e.g.,
participation in extracurricular school activities), lack of con-
ventional values, and poor parenting practices (e.g., distant
relations with the child; poor supervision and monitoring;
poor role modeling). Another environmental factor pertains
to the ease of access and availability of the addictive 'agent'
(i.e., drug, gambling venue, etc.). Two dramatic examples of
this pertaining to youth in the United States are the following:
( 1 ) the substances abused the most by youth are those that are
the most available and accessible to them - legal substances
such as tobacco and alcohol, and (2) the dramatic expansion
of legalized gambling in the United States is consistent with the
recent increase in gambling by teenagers.
Identifying Clinical Significance
It is true that relatively few teenagers who use drugs or gamble,
even regularly, will become addicted. Nonetheless, it can be
difficult to determine when involvement in addictive behaviors
by adolescents will have no negative consequences, when it may
involve short-term and minimal health effects, or when it may
be associated with serious negative long-term repercussions.
As will be discussed later in this article, the high prevalence
rate of drug use and gambling by teenagers gives some credence
to the notion that these behaviors are a normative part of youth.
When adolescents use drugs, it is typical that they use legal drugs
(alcohol or tobacco) and use them within a social context.
Gambling by adolescents often occurs with peers and involves
making informal, social bets (e.g., who will get the better test
score). The large majority of drug-using youth will not progress
beyond the use of the so-called gateway drugs of alcohol and
tobacco, and we can say with similar certainty that most youth
gamblers will not develop a gambling problem.
However, it is also the case that adolescence may be
a particularly vulnerable period to develop an addictive
disorder or an addition problem. For example, young people
show higher rates of alcohol problems compared to older
age groups. Figure 2 below shows that among youth aged
15-20years old, 12.2% met an official definition of an alcohol
dependence disorder within the past 12 months. This rate
was much higher than the other age groups.
There is a similar association between early onset gambling
and increased risk for adult gambling problems. Data collected
during 2001-2003 from the US National Comorbidity Survey
Replication Study found that adults with a pathological gam-
bling disorder reported first gambling significantly earlier than
gamblers without a history of pathological gambling (mean
14
i i i ! ! i ! i i
15-20 21-24 25-29 30-34 35-39 40-44 45-49 50-54 55-59 60+
Figure 2 Prevalence of past year DSM-IV alcohol dependence: United States, 2000-2001. Reproduced from Grant BF, et al. (2004) Drug and
Alcohol Dependence 74: 223-234.
1 2 Addictions and Adolescence
age 16.7 vs. 23.9years). Then there is the issue of current
problems associated with addictive behaviors during youth.
Later in this article we provide more details about prevalence
rates, but in a general sense the rates among adolescents of
each of our target addictive disorders are appreciable. One
example: Based on the National Household Survey data, it is
estimated that 1 1% of youth ( 12-18 year olds) meet a current
DSM-IV diagnosis for a substance use disorder.
Applicability of Addiction Diagnostic Criteria for
Adolescents
A related issue is the applicability of formal diagnostic criteria
for addictive disorders to adolescents. The three addictive dis-
orders we discuss in this article - nicotine dependence, sub-
stance use disorders and pathological gambling - are defined
by DSM-IV. However, adolescent-specific definitions of these
disorders do not exist in the DSM system, a concern that does
not sit well with many experts who can point to some existing
criteria that are not relevant for youth and can identify other
symptoms that are adolescent specific and clinically relevant
and yet are not included in the current diagnostic nomencla-
ture. We will return to definitional issues in upcoming sections.
Three Adolescent Addictive Disorders
We now turn our attention to the three domains of adolescent
addictive behaviors for which there is a relatively substantial
empirical literature - nicotine dependence, other substance use
disorders, and pathological gambling. Each subsection will
provide an overview of clinical characteristics and definition,
epidemiology, etiology, and treatment.
Nicotine Dependence
Clinical Characteristics and Definition
The primary tobacco product used by adolescents is the ciga-
rette, so this subsection will focus solely on this source of
nicotine. Experimentation with cigarette smoking is a common
phenomenon among adolescents, with upward of 46% of high
school seniors reported having previously tried smoking.
While not all adolescents who try smoking progress to habitual
smoking, data from the Monitoring the Future Study (MTFS)
reveal that over 12.3% of high school seniors reported smoking
daily. Rates of smoking continue to increase during late ado-
lescence/emerging adulthood (from age 18 to 24), resulting in
the highest rate of smoking (23.9%) among all age groups.
Factors that influence progression beyond initial experimenta-
tion may be quite distinct from those that influence first trials
with cigarettes. For example, social influence processes are
well-established predictors of initial experimentation, but
mood/emotional factors may be more important than social
influences in determining progression beyond initial trials
with smoking.
We know that the vast majority of current adult smokers
began their smoking careers by the age of 18. Given the associ-
ation between smoking in adolescence and subsequent health
problems in adulthood, the initiation and maintenance of
smoking during adolescence and young adulthood represent
genuine public health concerns. In addition, mounting evi-
dence links smoking with health problems (e.g., respiratory
and cardiovascular) during adolescence. The notion that ado-
lescent smokers are addicted to nicotine and thus driven by
compulsive drug-seeking behavior represents a serious concern
above and beyond that posed by the associated health risks.
Indeed, the Surgeon General in 1994 concluded in a published
report that "most adolescents [who smoke] are addicted to
nicotine and report that they want to quit but are unable to
do so; they experience relapse rates and withdrawal symptoms
similar to those reported by adults."
The notion that most adolescent smokers are nicotine
dependent should have a considerable impact on prevention,
treatment, and policy at multiple levels. However, the vast
majority of research on nicotine dependence has focused on
adult smokers, and relatively few attempts have been made
to document the emergence of nicotine dependence among
adolescents. Indeed, Alexander Prokhorov and colleagues, nic-
otine research experts, speculated that "The lack of interest in
the study of nicotine dependence in adolescents might be
explained by the prime orientation of adolescent research on
the prevention of smoking onset along with skepticism with
respect to the ability to develop an appreciable degree of nico-
tine dependence during the relatively short period of adoles-
cence." More than a decade later a paradigm shift in the field is
placing increasing attention on the need for research to identify
the factors that may promote smoking behavior and to exam-
ine the emergence of nicotine dependence in young people.
Epidemiology
Reducing smoking prevalence is an important national health
objective in the United States. Whereas cigarette smoking
had steadily decreased between 1997 and 2004, the National
Health Interview Survey (NHIS) indicates that cigarette smok-
ing prevalence has remained steady at 20.8% since then. Smok-
ing prevalence is higher among men (23.9%) than women
(18.0%), and among different racial groups, those of Asian
(10.4%) and Hispanic (15.2%) descent have lower smoking
rates than Caucasians (21.0%), African-Americans (23.0%), or
American Indians/Alaskan Natives (32.4%). There are also
higher rates of smoking among those with the lowest socioeco-
nomic status.
As noted earlier, smoking behavior most often begins in
adolescence, with 20% of 8th graders reporting some experi-
ence with cigarettes. Although adolescents may under-report
smoking behavior for a number of reasons, including social
undesirability, fear of disclosure, or inaccurate recall, tobacco
has remained the substance most often used on a daily basis by
high school students for over 30 years, according to responses
from successive cohorts of the MTFS. To track trends in ciga-
rette smoking among adolescents and young adults, the MTFS
researchers have administered a series of national surveys that
assess cigarette, alcohol, and illicit drug use over time. Results
show that in the 1980s, smoking rates remained steady among
adolescent smokers, though they declined in adults. In the
early 1990s, cigarette smoking among 8th and 10th graders
Addictions and Adolescence 1 3
rose by 50% before declining steadily in the second half of the
decade. The decline in smoking among high school students
has decelerated in recent years, however, and most daily ado-
lescent smokers report smoking half a pack or more per day.
A closer examination of these data reveals some factors that
might go some way toward explaining such trends in cigarette
smoking among youth. Increased cigarette prices reduce ciga-
rette smoking among adolescents, and not smoking sharply
reduces risk of using alcohol and illicit drugs. Importantly,
men are more likely to progress from occasional to regular
smoking, as are those who hold more positive smoking out-
come and affect regulation expectancies.
Whereas regular (e.g., daily) smoking is a hallmark of nico-
tine dependence in adults, adolescent smokers can apparently
exhibit symptoms of dependence with only occasional use,
often before the onset of daily smoking. To date, despite vari-
ous validated nicotine-dependence questionnaires, there is no
universally accepted tool for identifying or measuring nicotine
dependence in adolescents. Joseph DiFranza and his collea-
gues suggest that loss of autonomy over one's behavior might
be the most salient feature of nicotine dependence in adoles-
cents, rather than quantity or frequency of tobacco use. Indeed,
the development of a single symptom of nicotine dependence
appears to predict continued use, indicating that loss of auton-
omy and dependence might have a strong temporal (and per-
haps causal) relationship.
Etiology
The conventional description of the natural history of depen-
dent smoking is that individuals first try smoking in early to
mid-adolescence and gradually escalate their frequency of
smoking over a 2- or 3-year period to the point where they
are smoking daily. Daily smokers continue to escalate cigarette
use over the next several years as they develop into chronic,
heavily dependent smokers. Though this description is consis-
tent with national survey data, this pattern represents an aver-
age that does not convey the wide variety of patterns that can
occur during the emergence of nicotine dependence. To date,
little research has contradicted this pattern, though we expect
this may be due to the expectation that heavy daily smoking
is necessary for the emergence of nicotine dependence. That
is, the vast majority of smoking studies have measured depen-
dence symptoms only among respondents who meet clearly
established smoking patterns (e.g., daily use), and this ap-
proach precludes the examination of dependence symptoms
at nondaily levels of use. This approach derives from an im-
plicit assumption governing the field; that is, that an isomor-
phism exists between smoke exposure (i.e., amount smoked)
and symptoms of dependence. However, emerging evidence
suggests the relationship between smoking exposure and the
symptoms of withdrawal and dependence is far from perfect.
Research findings from the lab of one of the authors (Kassel
and Conrad) bear on the question of whether adolescent
smokers - even those who smoke relatively infrequently -
manifest signs and symptoms of nicotine dependence. They
assessed the acute effects of nicotine on positive (PA) and
negative affect (NA) in a group of 15- 18 year olds. A matched
group of nonsmokers served as a comparison group. This is
probably the first study conducted in the United States in
which nicotine was administered to adolescent smokers
under controlled laboratory conditions. The findings con-
formed to a negative reinforcement model of nicotine effects
and strongly suggested that, even among young (mean age
= 17.5), light smokers (5.5 days a week smoked; 3.5 cigarettes
a day), nicotine dependence and resultant withdrawal symp-
toms appear to serve as motivating factors governing smoking
behavior.
Several interesting and important findings also emerged
from correlational analyses. Craving was significantly corre-
lated with nicotine dependence (r=0.60), NA management
expectancies (r= 0.37), boredom relief expectancies (r = 0.53),
boost in expired air carbon monoxide (CO; an index of smoke
exposure) derived from smoking the cigarette (r=0.40), and
pre-post-smoking reduction in NA (r=— 0.38). In addition,
the measure of nicotine dependence demonstrated significant
correlations with recency of smoking (the more dependent the
participant was, the more recently he or she had smoked;
r=-0.40), number of cigarettes smoked on the study day
(r=0.49), and NA management expectancies (r=0.32).
Taken together, then, these findings suggest that even young,
relatively light smokers exhibit signs and symptoms of nicotine
dependence (albeit at low levels) and that symptoms of nico-
tine dependence are correlated in meaningful ways with other
smoking-related characteristics.
Nevertheless, the pathways to becoming a smoker are clearly
complex and governed by a host of genetic, social, pharmaco-
logical, and psychological determinants. While social factors
(e.g., peer affiliations and peer socialization) have emerged
as potent and reliable predictors of smoking initiation and
experimentation, some evidence suggests that heightened levels
of stress and negative affect, manifestations of psychopathol-
ogy associated with affective distress and anxiety, are linked to
progression in smoking status and nicotine dependence among
adolescents. Thus, distinguishing the processes that govern
smoking initiation and experimentation from those that under-
lie progression to nicotine dependence is of particular impor-
tance in examining smoking patterns in adolescents and young
adults.
Findings from cross-sectional and longitudinal studies
show that adolescents who experience significant levels of
affective disttess are more likely to smoke, and many researchers
have inferred that these adolescents smoke because it alle-
viates negative affect. However, ascribing causal (within-subject)
mechanisms based on results derived from cross-sectional
(between-subject) analyses may not always be appropriate.
Although only limited empirical work has been devoted to
answering the question of whether smoking genuinely influ-
ences emotional response in adolescents, several studies do
offer some insight into the motives associated with adolescent
smoking. Though not a direct indicator of nicotine dependence,
the subjective effects experienced by smokers are likely to reflect
the role of nicotine (and other constituents of tobacco smoke)
in governing smoking behavior. A number of studies point to
the importance of affect management as motivation for smok-
ing among young smokers. For example, several studies have
found that adolescent smoking initiates commonly reported
'smoking to relax' and 'stress reduction' as important motives
for smoking. More recent studies by Johnson and colleagues
1 4 Addictions and Adolescence
also reveal significant endorsement of emotional reinforcement
motives among adolescent smokers, with a large proportion
attributing their smoking to its reputed ability to facilitate affect
regulation. Correspondingly, a number of recent investigations
found that smoking in response to negative affect was the most
frequently endorsed motive among adolescent smokers.
Treatment
There are a number of obstacles to overcome when considering
treatment options for adolescent smokers. First, intervention
efforts are most often impeded by irregular smoking patterns
among this group, as most teenagers smoke under limited
circumstances and, perhaps most importantly, do not view
their behavior as a serious problem. At this transitional life
stage, adolescent smokers are also often concerned with pri-
vacy and autonomy. Most smoking cessation programs are
tailored to adult regular smokers and are therefore less accessi-
ble to an adolescent occasional smoker. Ultimately, successful
interventions are careful to tailor the treatment to the individ-
ual and protect confidentiality.
Currently, cognitive-behavioral therapy (CBT) is the treat-
ment most often recommended for adolescent smokers who
want to quit, in the absence of youth-specific treatment
options. Previous research has focused on the frequency and
duration of treatment sessions, comparing the effectiveness of
longer sessions offered once per week to shorter sessions
offered several times per week, often with a contingency man-
agement component that provides tangible reinforcement for
maintaining abstinence. Studies show that abstinence rates do
not vary as a function of session length or frequency. As such,
this finding might represent an opportunity to tailor treatment
to the individual, based on convenience and preference.
Medical-based or pharmacological treatments used in con-
junction with CBT have also garnered increasing empirical scru-
tiny in adolescents. Generally, the nicotine patch appears safe to
use with smokers as young as 13 years old. The patch, as com-
pared with placebo, decreases cigarette craving and withdrawal
symptoms, though it ultimately appears to have no effect on
abstinence rates. Bupropion, an antidepressant deemed effica-
cious as a smoking cessation aid in adult smokers, has also been
found effective in adolescent smokers, although it is usually
employed in conjunction with brief counseling or family support.
Alternative smoking cessation programs have recently fo-
cused on using different media outlets to offer counseling
services. In the short-term, Internet-based virtual reality appli-
cations, monitored by a trained counselor, result in enhanced
1-week abstinence rates, fewer smoking days in the past week,
and fewer cigarettes smoked in the last week for those treated
than those receiving no treatment. Web-assisted tobacco inter-
vention programs that combine interactive quizzes with self-
assessment and feedback show promising results in reducing
intentions to smoke, increasing resistance to smoking, and
decreasing heavy cigarette use.
Other studies indicate that brief visits with the pediatrician
in a medical office setting can increase the adolescent's readiness
to quit. One study offered four brief individual weekly sessions
in the medical office, while a home-based Internet intervention
was accessible for 24 weeks. Whereas no group differences in
abstinence rates were observed, those who continued to smoke
showed fewer smoking days in the office condition compared to
the Internet intervention. Most important was the rapid decline
in accessing the Internet-based intervention: by week 3 of the
study; access had dropped to less than one third of all partici-
pants. Among the most popular Web pages were those offering
discussion support groups, followed by quit plan pages. Inter-
action pages were accessed more often than informational
pages, and men accessed fewer interactive pages than women.
These patterns of use hold important implications for the
design of future Web-based applications and their inclusion
as a component of smoking cessation treatment.
On an individual treatment level, brief self-help interven-
tions appear generally effective with adolescent smokers who
are less nicotine dependent. School-based programs, however,
are both prevention and treatment focused in nature, and are
thus able to target a larger range of smokers. Standard support
within school settings usually comprises abstinence-based les-
sons. Adding eight, 1-hour no-smoking lessons over the course
of 2 years, in addition to support from school nurses, reduces
the likelihood that students will smoke regularly or at all in the
past month. School-based tobacco use interventions are also
effective in reducing overall smoking prevalence, smoking ini-
tiation, and smoking intentions in the short term. Unfortu-
nately, relatively few studies perform long-term follow-up, and
among those that have, little evidence has been found for
school-based prevention programs. In light of these mixed
findings, some researchers suggest that schools frame cigarette
smoking as an example of a real-world problem in order to
further develop critical thinking skills. The best school-based
programs are interactive in nature and teach skills to resist drug
offers and prodrug influences, correct misperceptions that drug
use is normative, and increase social and personal competence.
On a more community-based level, previous studies indi-
cate that well-funded mass-media campaigns targeted at the
general population and implemented at the state level, in
addition to a comprehensive tobacco control program (e.g.,
restricting easy access to tobacco products by removing ciga-
rette vending machines; compliance checks at points of pur-
chase to enforce the minimum age law) are associated with
reduced smoking rates in adolescents. Also, as we noted above,
raising the cost of tobacco products is generally associated with
reductions in tobacco use among youth.
Other Substances Use Disorders (Excluding Nicotine)
Clinical Characteristics and Definition
Substance involvement ranges along a continuum of severity
that includes abstinence, experimentation, occasional use, reg-
ular use, and heavy use and associated problems. Most of the
population engages in minimal to occasional use, with increas-
ingly smaller proportions involved in more regular and heavy,
chronic levels of substance use. For individuals at the higher
end of the substance involvement continuum, the presence of a
substance use disorder (SUD) reflects a categorical determina-
tion (i.e., disorder is present or absent) that an individual's
substance use results in significant impairment in functioning
or subjective distress associated with need for treatment.
DSM-IV SUDs were developed based on clinical experience
and research involving adults. When applied to adolescents,
Addictions and Adolescence 1 5
DSM-IV SUDs have demonstrated validity as markers of an
adolescent's level of substance involvement and associated
impairment in functioning. However, certain limitations of
DSM-IV SUDs when applied to youth also have been identified,
highlighting the importance of developmentally informed
assessment of substance-related problems. An important limita-
tion of DSM-IV SUDs when applied to adolescents involves the
threshold used to determine the presence of SLID. Some ado-
lescents may report substance-related symptoms, but fall short
of meeting criteria for a DSM-IV SLID, either because the symp-
toms reported are not included among the diagnostic criteria or
because the number of symptoms does not meet the threshold
for diagnosis. Some youth who report substance-related pro-
blems do not meet SUD criteria, and are known as 'diagnostic
orphans.' The diagnostic orphans typically report one to two
dependence symptoms, and do not meet criteria for a diagnosis
of abuse. Research suggests that although diagnostic orphans
do not meet criteria for a SUD, they report similar levels of
substance use and impairment in functioning as youth with
an abuse diagnosis. Diagnostic orphans appear to represent
'missed' cases of SUD that are particularly important to consider
when assessing adolescent substance involvement.
Another limitation of DSM-IV SUD criteria when applied to
adolescents involves adequate coverage of substance-related
problems that are relevant to youth. For example, substance
dependence criteria focus on failed attempts to quit or cut
down, behaviors that are commonly reported among adults
in addictions treatment. However, due to adolescents' typically
shorter histories of substance use compared to adults, adoles-
cent substance users may report a pattern of frequent, heavy
substance use, that has not yet progressed to attempts to cut
down or stop use. In this regard, DSM-IV SUD criteria are
limited in capturing a compulsive pattern of use among those
who have not yet tried to cut down or stop substance use,
resulting in possible under-identification of youth who may
benefit from intervention. As another example, the abuse crite-
rion regarding substance use in a hazardous situation is often
met by recurrent episodes of driving while intoxicated among
adults. However, this criterion applies less to youth who are
not driving or who have limited access to a vehicle. Among
younger adolescents, symptoms such as recurrent substance-
related blackouts, passing out, or risky sexual behavior, may
indicate hazardous use, and although they are not explicitly
included in DSM-IV SUD criteria, they need to be considered
when evaluating adolescent substance users.
Developmentally informed assessment of substance in-
volvement aims to minimize false positive and false negative
symptom assignments in order to increase overall validity of
SUD diagnoses, particularly when applied to adolescents. Spe-
cifically, certain SUD symptoms may manifest differently in
adolescents (e.g., drop in school grades due to marijuana use),
compared to adults (e.g., marijuana-related impairment in job
performance). Other symptoms may have a different meaning
in adolescents and adults due to the developmental context in
which the symptom occurs. For example, youth may report the
dependence symptom as 'spending much time trying to obtain
alcohol, drinking, or getting over its effects' due to difficulties
in obtaining alcohol as a result of their status as minors, rather
than due to a compulsive pattern of use, which is the intended
meaning of the symptom. In addition, some youth may report
symptoms of dependence (e.g., withdrawal) at relatively low
levels of use. Careful querying is needed to ensure that the
adolescent understands what is being asked, particularly for
relatively complicated phenomena (e.g., clinically significant
tolerance, withdrawal) in order to avoid false positive symp-
tom assignments and over-diagnosis.
Epidemiology
The prevalence of substance use rises steadily over ages 12-21,
then levels off after young adulthood. There are two periods of
peak risk for substance use initiation. One risk period occurs in
early adolescence, around ages 13-14, coincident with puber-
tal maturation, and the other period occurs during the transi-
tion from late adolescence to early adulthood, around ages
18-21. The substances most commonly used by adolescents
(ages 12-18) include alcohol, tobacco, and marijuana. The
sequence of substance use initiation typically begins with use
of alcohol and tobacco, which may be followed by use of
marijuana, and possible progression to use of other illicit
drugs. The 'gateway hypothesis' has been proposed to explain
the fairly predictable sequencing of substance use initiation, and
states that use of a substance early in the sequence leads
to the use of the next substance in the sequence. Support for
the gateway hypothesis has been debated, since environmental
factors such as drug availability may influence the sequence of
drug use initiation for a given individual. Regardless of the
actual temporal ordering of substance use initiation, progression
to higher levels of use within a type of substance, or to the use of
other types of substances, is probabilistic, and not inevitable.
Initiation of alcohol use by early adolescence (e.g., prior to
age 15), although relatively infrequent, has been associated
with greater risk for subsequent development of an alcohol
use disorder. Results from Monitoring the Future (MTF), a
national school-based survey, indicated that in 2009, 15% of
8th graders, 30% of 10th graders, and 43% of 12th graders
reported alcohol use in the past month. Early exposure to
alcohol may reflect alcohol use in a religious context, or only
sips or tastes of alcohol. However, the particular concern with
respect to early exposure is when a young person engages in
a risky pattern of alcohol use, which involves consump-
tion of five or more drinks in a row, a pattern described as
'heavy episodic drinking' or 'binge drinking.' Data from the
2008 National Household Survey on Drug Use and Health
(NSDUH) indicate that 9% of youth (ages 12-17) reported
an episode of heavy drinking in the past 30 days. Although the
prevalence of episodic heavy drinking among adolescents has
gradually declined in the past decade, the absolute prevalence
of this risky pattern of alcohol use among youth in recent years
remains cause for concern.
According to the 2008 NSDUH, the prevalence of illicit
drug use among 12- 17 year olds has declined significantly in
recent years, from 12% in 2002 to 9% in 2008. However, 2009
MTF data suggest that the prevalence of marijuana use may be
increasing slightly among adolescents, in tandem with a
decline in adolescents' beliefs regarding the risks associated
with marijuana use. Among illicit drugs, marijuana is the
substance used most often by adolescents. In 2008, 7%
of 12- 17 year olds reported marijuana use, whereas use of
cocaine or heroin was reported by <1% of youth. Although,
1 6 Addictions and Adolescence
overall, marijuana is the illicit drug used most often by adoles-
cents, there are important age differences in the type of sub-
stance most often used. Among 12- 13 year olds, nonmedical
use of prescription medication (e.g., pain relievers, tranquili-
zers) was slightly more prevalent than marijuana use (in 2008:
1.5% vs. 1.0%, respectively). In contrast, among 14-17year
olds, marijuana was the illicit drug most often used (in 2008:
5.7% among 14-15year olds, and 12.7% among 16-17year
olds), followed by nonmedical use of prescription medication
(3.0% among 14-15year olds, and 4.0% among 16-17year
olds). The age differences in the relative prevalence of types of
illicit drug use emphasize the importance of evaluating non-
medical use of prescription medication, particularly among
younger adolescents (ages 12-13), who may have access to
prescription medications (e.g., opiate-based pain relievers,
sedatives) in their homes.
Among adolescents who progress from initial use to SUD,
the time to onset of SUD occurs, on average, within 3 years
of first use of a substance. The relatively rapid progression to
SUD among adolescents underscores the importance of early
detection and intervention to prevent escalation of substance
use. The prevalence of SUD increases with age through adoles-
cence, and reaches a peak in young adulthood. In 2008, 7.6%
of 12-17 year olds met SUD criteria in the past year, compared
to 20.8% of 18-25 year olds. The most common SUDs among
youth aged 12-17 involve alcohol, marijuana, and tobacco.
From 2002 to 2008, the prevalence of alcohol abuse or depen-
dence among 12-17year olds declined from 5.9% in 2002 to
4.9% in 2008. Over the same time period, abuse or depen-
dence on illicit drugs also declined from 5.6% to 4.6% among
12-17year olds, primarily due to a decline in marijuana use
disorders. Among past month cigarette users aged 12-17, one
out of three adolescents was estimated to be dependent on
nicotine. In addition to youth who met criteria for an SUD,
an additional 17% of adolescents were estimated to be 'diag-
nostic orphans,' who reported substance-related problems, but
did not meet criteria for a SUD. Limited information from
national survey data exists on the proportion of youth who
meet criteria for more than one SUD. Youth who meet criteria
for at least one SUD may engage in a pattern of polysubstance
use, typically involving use of alcohol, marijuana, and tobacco.
Adolescent polysubstance use suggests a concentration of
substance-related problems within a subgroup of youth who
often experience problems with several substances that have
negative effects on multiple areas of functioning (e.g., at home,
at school, legal involvement).
The gender gap in substance use prevalence, whereby males
generally had higher prevalence of substance use compared to
females, has narrowed in recent years, particularly among ado-
lescents. In the 2008 NSDUH, among 12-17year olds, the pro-
portion of current drinkers was similar for males and females
(14.2% and 15.0%, respectively). Likewise, there was little dif-
ference in the prevalence of current cigarette smoking among
males and females (9.0% and 9.2%, respectively) in this age
group. Male and female youth also had similar rates of illicit
drug use (9.5% and 9.1%, respectively). Although males had
slightly higher prevalence of marijuana use compared to females
(7.3% and 6.0%, respectively), females had higher prevalence of
nonmedical use of prescription drugs (3.3% and 2.5%, respec-
tively). With regard to prevalence of SUD, adolescent females
had slightly higher prevalence (8.2%) of past year substance
abuse or dependence compared to males (7.0%), although
there was little difference by gender in rates of illicit drug use.
The narrowing of the gender gap in rates of substance use, with
females surpassing males in rates of use for certain substances
(e.g., misuse of prescription drugs) and in meeting criteria for
SUD, calls for increased efforts to reduce adolescent substance
use, particularly among adolescent females.
National survey data indicate some ethnic differences in
substance use prevalence. Among 12- 17 year olds, for most
substances, American Indian and white adolescents generally
had the highest rates of use, followed by Hispanic and black
youth, and Asian youth tending to report the lowest rates of
substance use. In 2008, 17.2% of American Indian adolescents
and 16.3% of white adolescents reported recent alcohol use,
compared to 14.8% of Hispanic youth, 10.1% of black adoles-
cents, and 5.7% of Asian youth. For illicit drugs, past year use,
which primarily involved marijuana, was reported by 3 1.3% of
American Indian youth, 19.7% of white adolescents, 17.6%
of black youth, 18.8% of Hispanic adolescents, and 9.8% of
Asian youth. In parallel with the relative prevalence of sub-
stance use across ethnic groups, white adolescents had higher
rates of past year SUD compared to black and Hispanic youth
(9%, 5%, and 7% respectively). The higher rate of SUD among
white youth reflects the higher prevalence of alcohol use dis-
orders in this subgroup. Ethnic differences in the prevalence of
substance use and SUDs suggest the potential utility of cultur-
ally tailored prevention and intervention efforts.
Etiology
Developmental contexts and transitions in biological, psycho-
logical, and social domains that are unique to adolescence
influence the emergence and progression of substance involve-
ment. In particular, ongoing brain development during ado-
lescence involves a normative delay in the maturation of
behavioral inhibitory systems in relation to neural systems
associated with reward processing. The relative immaturity of
inhibitory control systems during adolescence has been pro-
posed to underlie a developmentally normative increase in risk
taking and sensation-seeking behavior, such as substance use
during adolescence. Pubertal maturation during early adoles-
cence also may contribute to risk for initiating substance use,
particularly among early maturers, who may have greater
access to substances through affiliation with older peers. In
addition, animal models suggest that adolescents tend to be
less sensitive to the sedative effects of certain substances (e.g.,
alcohol) compared to adults. Adolescents' lower sensitivity to
alcohol's sedative effects helps to explain the pattern of heavy
episodic drinking observed among adolescents in both animal
models and humans. These biologically based aspects of devel-
opment, which occur during adolescence (e.g., relative imma-
turity of inhibitory control systems, early pubertal maturation,
adolescent sensitivity to drug effects), foster conditions asso-
ciated with peak risk for initiation to substance use. Although
experimentation with substance use typically begins in
adolescence, heavy substance use at a time when the brain
is continuing to develop has been associated with possible
neurocognitive impairment and multiple adverse outcomes
(e.g., substance-related injury, impaired school performance).
Addictions and Adolescence 1 7
A developmental model of substance involvement pro-
poses that risk and protective factors influence the onset and
escalation of substance use during adolescence. Risk factors
(e.g., substance using peers) predict onset and progression of
use, whereas protective factors (e.g., parental monitoring)
interact with risk factors to delay the onset of substance use
and reduce the likelihood that substance use will escalate.
The main domains of risk and protective factors include
those associated with the individual (e.g., genetic liability for
substance use disorder, temperament, co-occurring psychopa-
thology, reasons for use, stressful life events), family environ-
ment (e.g., parental substance use, availability of substances in
the home), peer relations (e.g., substance using peers), and
school and neighborhood contexts (e.g., drug dealing that
occurs in the neighborhood). These factors are organized in a
hierarchical, nested structure, such that some individual factors
(e.g., temperament, attitudes toward substance use) are gener-
ally more strongly predictive of substance use than the immedi-
ate social environment (e.g., family, peer group), and the school
or neighborhood contexts within which an adolescent is
embedded. These interacting systems operate at multiple levels
(e.g., ranging from public policies and media exposure to peer
norms and family rules regarding substance use) in influencing
an adolescent's risk for substance involvement over time.
Risk and protective factors interact dynamically to influence
adolescent substance use. For example, risk and protective
factors may evolve over time (e.g., parental monitoring may
decrease with age), may have developmentally specific effects
(e.g., transition to high school or college) or persistent effects
(e.g., sequelae of sexual or physical abuse), or be differentially
salient at certain points in development (e.g., increasing
salience of peers, relative to family, with age). Some risk factors
may be specific to a certain substance (e.g., positive attitudes
toward alcohol use), whereas others may increase risk or protec-
tion across substances more generally (e.g., parental neglect).
Factors that predict transitions between levels of use within a
given substance also may differ (e.g., social reasons for use may
predict early onset of substance use, whereas coping reasons for
use may predict progression to heavy use). Some research sug-
gests that the cumulative number of risk factors, more than any
specific type of risk factor, predicts level of substance involve-
ment in adolescents. A developmental model of substance use
recognizes individual differences in the timing and pace of devel-
opment, and multiple pathways toward and away from sub-
stance involvement. In addition, the risk and protective factors
that influence the course of substance involvement may differ
across subgroups, defined for example, by gender and ethnicity.
Treatment
Few adolescents meet criteria for a DSM-IV SUD report receiv-
ing treatment. Among youth ages 12-17, <1% received treat-
ment for substance use, although 8% had a DSM-IV SUD. In
part, the discrepancy between the proportion of youth with an
SUD and the proportion who receive treatment occurs because
adolescents seldom initiate treatment. Most youth in publicly
funded treatment are court-mandated to treatment, or are
referred by the school or family. In accord with typically
short histories of substance use and emerging substance-
related problems, the majority of adolescents are treated in
outpatient programs. Youth with more chronic, heavy patterns
of substance use, who have not been able to reduce use as
outpatients, may benefit from more intensive intervention
offered in residential treatment. Adolescents (ages 12-17)
represented 12% of admissions to publicly funded substance
abuse treatment in 2007. Two out of three youth who pre-
sented to treatment for substance use in 2007 reported canna-
bis as their primary drug, 8% reported alcohol, 2% reported
opiates, and 2% reported cocaine as their primary drug.
Substance use treatment for adolescents has largely been
adapted from programs shown to be effective with adults.
The most widely used treatments for adolescent substance use
include the Minnesota Model, which is based on the Twelve
Steps of Alcoholics Anonymous; CBT focused on relapse preven-
tion, motivational enhancement interventions, and family-based
interventions. Specialized programs, such as multi-systemic ther-
apy, also have been developed to address the multiple treat-
ment needs (e.g., family, psychiatric, legal) of substance using
youth in the juvenile justice system. Studies of the effectiveness
of pharmacological treatment for adolescent substance use are
limited. No single type of treatment for adolescent substance
use has been shown to be clearly superior to any others. The role
of external pressure in bringing adolescents to treatment sug-
gests the importance of increasing adolescents' readiness to
change substance use behavior, and of motivational enhance-
ment interventions in facilitating reductions in substance use.
For youth who experience problems in multiple areas of func-
tioning, a comprehensive and coordinated set of interventions
is needed to facilitate positive outcomes. Adolescents may
require more than one episode of treatment to facilitate the
long-term maintenance of treatment gains.
Treated adolescents generally show reductions in substance
involvement compared to pretreatment levels of use, as well as
improvements in psychosocial functioning. However, indivi-
duals show different trajectories of posttreatment substance
involvement. Following treatment, prototypical short-term tra-
jectories of substance involvement include stable low, stable
high, increasing, and decreasing patterns of use. Most youth are
in stable low or decreasing use trajectories immediately follow-
ing treatment. Longer-term trajectories treated youth followed
into young adulthood include stable abstinence, infrequent
use, gradually decreasing substance involvement ('slow im-
provers'), and persistent high substance involvement. Most
adolescents who followed into young adulthood were 'slow
improvers' or infrequent users, with smaller proportions in the
stable abstinence and persistent high substance involvement
trajectories. For polysubstance users, treatment tends to have a
generalized, rather than substance-specific effect on substance
use (e.g., alcohol and marijuana tend to change together). Indi-
viduals in stable abstinence and low substance involvement
trajectories after treatment generally had better emotional,
interpersonal, and family functioning in young adulthood
compared to those in chronic use trajectories. Improvements
in school functioning tend to occur early, roughly within a
year after treatment, while improvements in other areas, such
as family relations, tend to emerge 2 years after treatment.
Although many treated adolescents show reductions in sub-
stance involvement and improvement in psychological and
social functioning, treated adolescents continue to show greater
overall impairment in functioning compared to a community
1 8 Addictions and Adolescence
comparison sample over long-term follow-up. Adolescent-onset
SUD can delay or disrupt the development of cognitive and
social skills, with effects that persist into adulthood.
Predictors of adolescent treatment outcomes have been
examined at the time of treatment admission, during treat-
ment, and following treatment. At treatment admission, the
presence of co-occurring psychopathology, such as conduct
problems, depression, and symptoms related to sexual or phys-
ical abuse, signal risk for a persistent pattern of substance use.
During treatment, factors that have been associated with better
outcomes among adolescents include longer duration of treat-
ment, treatment completion, greater readiness to change sub-
stance use, and family involvement in treatment. Although
treatment may help to increase or maintain an adolescent's
motivation to reduce substance use, the nature of the ado-
lescent's environment (e.g., family, peers) and activities (e.g.,
leisure time activities) outside of treatment is essential to
maintaining treatment gains over time. Posttreatment factors
associated with better outcomes among adolescents include
aftercare involvement, low levels of peer substance use, and
continued commitment to abstain from substance use.
More chronic and severe trajectories of substance involve-
ment may reflect the impact of co-occurring psychopathol-
ogy (e.g., conduct problems, depression) in complicating the
course of recovery. Not only does co-occurring psychopathol-
ogy affect treatment outcomes, but it also may play a role in the
timing of SUD onset, the rate of SUD development, severity,
and duration. The co-occurring conditions most commonly
associated with adolescent substance involvement include con-
duct problems, mood disorders (e.g., depression), ADHD, and
physical or sexual trauma. A majority (roughly 60%) of ado-
lescent substance users in a community sample were found
to have a coexisting psychiatric condition, such as ADHD or
conduct disorder. Among youth in substance use treatment,
more than half are estimated to have one or more of these
coexisting disorders. Comprehensive interventions that ad-
dress the substance problems and co-occurring problems are
indicated to support long-term positive outcomes in youth
with coexisting conditions.
Pathological Gambling
Clinical Characteristics and Definition
Gambling is the act of wagering or betting money or some-
thing of value on an event with an uncertain outcome with
the intent to win more money or things of value than was
wagered. Gambling involves risking something of value,
including money, for the chance of winning more than you
risked. For adolescents, gambling is viewed as an adult activity
that they can participate in fairly easily, such as playing poker
for money with their friends, and without upsetting their par-
ents too much, "Oh, they are only playing cards, good, I was
afraid they were smoking or drinking." Gambling behavior
among adolescents ranges from no gambling to experimenta-
tion to occasional or regular social gambling to excessive and
problematic gambling. This continuum is skewed toward most
youth gambling rarely or occasionally and some youth gam-
bling excessively. Adults play commercial or legal forms of
gambling such as slot machines at casinos or buy lottery
tickets, whereas youth tend to play informal games amongst
themselves such as poker or betting on sports or games of skill
such as video games. The legal age for gambling varies across
jurisdictions but is most commonly between 18-21 years of
age. Some youth celebrate reaching the legal gambling age
birthdays by visiting a casino. Some underage youth play com-
mercial games such as the lottery by obtaining lottery products
from legal-age gamblers, including family members.
Adolescents can exhibit pathological gambling or gam-
bling addiction. There is not a separate or different definition
of pathological gambling for adolescents; however, it is exhib-
ited slightly differently in youth. For example, an adult path-
ological gambler may be absent from work in order to
gamble, whereas an adolescent may be absent from school
in order to gamble. Adults may lie to their spouse about
their gambling, whereas adolescents may lie to their parents
about their gambling. Adults may spend their paycheck on
gambling when the money is supposed to pay for food and
housing, whereas Adolescents may wager their pocket money
or their iPod or video game player. Adolescent gamblers can-
not lose their house, or spouse or family, or file for bank-
ruptcy, but they can exhibit adolescent-specific adverse
consequences. Gambling at any group is considered to be a
problem when it interferes with the individual's relationships
with friends and family or their school and work obligations.
Adolescent problem gambling has been characterized as per-
sistent gambling behavior that leads to negative personal
consequences, including the loss of money and things of
value, and negative consequences to the person's social net-
work including the alienation of family and friends. Not only
can early gambling lead to later gambling problems, gambling
during the formative years can lead to current gambling
problems.
A small number of youth gambling assessment instru-
ments have been developed to assist clinicians in identifying
youth who may have a gambling problem. For example, a
recently developed instrument was created specifically for
adolescents, the Canadian Adolescent Gambling Inventory
(CAGI). It contains items associated with symptoms of path-
ological gambling, such as preoccupation, loss of control, and
chasing one's gambling losses by accelerating one's gambling
involvement.
Epidemiology
It is important to examine the larger cultural context in which
gambling occurs. The gambling industry has experienced
unprecedented expansion in the past three decades. Commer-
cial gambling in the United States has expanded out beyond
Atlantic City, New Jersey and Las Vegas, Nevada to nearly every
convenience store across America and nearly every computer
with access to the internet. Gambling is advertized in all forms
of media and youth are exposed to this promotion on a daily
basis. The availability of gambling venues varies by region,
such as state lotteries or tribal casinos, but commercial gam-
bling has become much more accessible to the general public
in the past 30 years. The gambling industry is also enjoying a
more favorable public image as a socially acceptable activity for
the general public as opposed to its public image in the past as
a vice. Citizens are encouraged to gamble by their state
Addictions and Adolescence 1 9
governments that benefit from the proceeds. With this rapid
growth of the gambling industry and the shift from a negative
to a positive public sentiment toward gambling have come
concerns about youth gambling.
Most youth have gambled at least once in their lifetime. The
National Research Council recently reviewed existing youth
surveys and reported that between 39% and 92% of youth
reported gambling at least once in their lifetime and that the
median rate was 85%. In terms of recent gambling, the NRC
concluded that between 52% and 89% of youth have gambled
in the past year and that the median rate was 73%. Whereas
the majority of youth have gambled in their lifetime and in the
past year, most do not experience any adverse consequences
from their gambling. Gambling for most youth is an infrequent
and inconsequential past time. Gambling for some youth is
part of adolescent experimentation with adult behaviors, for
example, some adolescents celebrate reaching the legal gam-
bling age with a visit to a casino.
While the literature is fairly consistent about the extent of
youth gambling participation, there is much less agreement on
the prevalence of problem and pathological gambling among
youth. Rates of past year pathological gambling range from as
low as 0.3% to as high as 9.5% with a median of 6%. Rates of
combined past year problem and pathological gambling range
from 1 1.3% to 27.7% with a median of 20%. The estimates of
lifetime problem and pathological gambling range from 7.7%
to 34.9% with a median of 15.5% (NRC, 1999). For lifetime
PG only, estimates range from 1 .2% to 1 1.2% with a median of
5%. These wide-ranging estimates are thought to be largely due
to using different methodologies, such as telephone versus in-
school survey administration, different assessment tools, and
different cut scores.
Nevertheless, these and other surveys provide some consis-
tent findings: the majority of youth have gambled in the past
year, more boys engage in gambling than girls, older youth
tend to gamble more than younger youth, a small percentage
of youth are frequent gamblers, and an even smaller percentage
may be considered problem or pathological gamblers.
A related epidemiological issue is whether rates of youth
gambling is on the rise. A recent review of youth gambling
surveys around the globe found that the evidence for changes
in youth gambling rates over time is mixed and varies by
location and methodology. For example, an examination of a
series of youth surveys in the state of Minnesota concluded that
there has been a decline in general gambling participation rates
from 1992 to 2007. Another informative survey is from the
Annenberg Foundation; since 2002 they have conducted
annual telephone surveys of about 900 youth in the United
States between the ages 14 and 22 (www.annenbergpublicpo-
licycenter.org). They recently reported weekly card playing
among 14- 17 year old boys peaked in 2005 and has continued
to decline in 2006, 2007, and 2008. The Minnesota survey
found a similar trend with respect to card playing. It would
appear that youth gambling, including problem gambling,
while showing occasional fluctuations for specific games, for
the most part has not been on the rise over the past two
decades. While this is good news, there still remains a small
but significant portion of the youth population who are exces-
sive gamblers and who are experiencing adverse consequences
as a result of their gambling.
Etiology
Research into the etiology of youth problem gambling is in its
infancy. The most common etiology studies are those that look
at psychosocial correlates of youth problem gambling. These
studies have found a number of correlates including being male,
antisocial behavior, peer deviance, tobacco use, alcohol use,
drug use, parental/familial gambling, school behavior problems
and academic failure, and impulsivity.
Also, there is emerging evidence that genetic factors may
conttibute to adolescent gambling behavior. Family studies
have shown that problem gambling is familial; adolescent prob-
lem gamblers are more likely to report having at least one parent
with a history of problem gambling compared to nonproblem
adolescent gamblers, and twin studies have shown that genetic
factors influence gambling behaviors among young adult males.
Another etiological factor to consider with youth gambling is
indirectly provided by research that has shown that youth prob-
lem gambling is associated with other psychiatric disorders,
including other addictions, mood disorders, and ADITD. It is
thought that problem gamblers seek out the stimulation of
gambling because their natural resting state is one of being either
under stimulated or over stimulated. For example, depressed
individuals may seek out gambling for stimulation and hyperac-
tive individuals may seek out gambling to calm their senses by
focusing on the gambling activity. One study found that young
problem gamblers also reported having ADHD symptoms and
another study found that young adults with childhood ADHD
symptoms that persisted into young adulthood reported greater
gambling problem severity than young adults with no ADHD or
those with childhood ADHD that did not persist into young
adulthood. This research suggests that young problem gamblers
may have impulse control deficits and these deficits may have a
neurological or biological basis. Further research will be required
to examine the neurological basis of impulse control deficits in
young problem gamblers, but there appears to be a link between
impulse control and executive function deficits and adolescent
problem gambling.
Another possible determinant of youth problem gambling
is the widespread and growing availability of legal forms of
gambling. Gambling is now available in nearly every conve-
nience store and every computer connected to the Internet. This
widespread availability has the added effect of a growing social
acceptance of gambling. Where gambling was previously con-
sidered a vice and participation was discouraged, it is now
considered entertainment and participation is encouraged.
While most jurisdictions have age restrictions for legal forms
of gambling, underage adolescents have found ways to partici-
pate, such as having legal-age adults buy lottery tickets for them.
The adult etiological literature points to the importance of
cognitive distortions or illusions in the development of problem
gamblers, and these factors have been discussed in light of
adolescent problem gambling as well. Cognitive distortions
associated with gambling involve the following: an illusion of
control where the individuals believe they can control the out-
come of a gambling event; the false or mistaken belief that the
person can win money playing games of chance; a failure to
understand the laws of probability, odds, and independence of
random events; and a tendency to underestimate one's losses
and overestimate one's wins.
20 Addictions and Adolescence
A final etiological consideration pertains to neurodevelop-
ment. As we noted in the introduction, recent advances in
understanding important neurodevelopmental events during
adolescence occur in brain regions associated with motivation,
reward, and decision making. It has been hypothesized that the
relative immaturity of these developing brain regions may con-
tribute to adolescent impulsivity and risk taking, including gam-
bling. Clearly, a common factor thought to underlie gambling is
the tendency toward general risk taking and a lack of control of
one's impulses. Gambling is a risky activity, and in that respect is
similar to other high-risk behaviors such as tobacco use, alcohol
use, drug use, and unprotected sexual activity.
Treatment
There has not been much attention given to treatment for youth
problem gambling in the literature. This would seem contradic-
tory given that on the one hand, we have estimates of youth
problem gambling that are higher than adult estimates. On the
other hand, we have few if any adolescents presenting them-
selves for treatment services. There are several possible explana-
tions for this disparity: youth problem gambling rates may be
inflated due to biases in survey tools; youth are less likely to
present themselves for treatment of any disorder, problem
gambling included; and parents - a main referral source for
teenagers receiving clinical care - are not very likely to recognize
problem gambling in their child or they do not consider gam-
bling as a risk behavior that merits concern. Regardless of the
apparent discrepancy in youth and adult prevalence rates, there
have been some efforts to examine adolescent treatment
options for problem gambling. Treatment options for adoles-
cents, like assessment tools, are typically adaptations of adult
methods. Current research work in this area is on the develop-
ment of behavioral treatment methods.
The reality is that there are few treatment programs for
adolescent problem gamblers. However, there are several efforts
to protect youth from the negative consequences of problem
gambling with primary prevention efforts. These efforts have
fallen into two camps, either abstinence or harm reduction.
Abstinence approaches attempt to delay the onset of gam-
bling in order to reduce gambling problems. Harm reduction
approaches attempt to promote responsible gambling behavior
and enhance skills needed to maintain control when gambling
through teaching youth about the facts of gambling as well as
the risks associated with gambling. This approach also addresses
the cognitive distortions and misperceptions that youth may
have about gambling, such as a belief that they are 'lucky' and
therefore can win money gambling. Also, some programs take a
broader approach by targeting a number of related high-risk
behaviors such as tobacco use, alcohol use, drug use, along
with gambling. Most youth problem gambling prevention pro-
grams have not been rigorously evaluated and therefore, there is
a need for the systematic evaluation of these programs.
Summary
Adolescence is a developmental period that involves significant
biological, psychological, and social changes, including rapid
growth in physical and mental capabilities. This is also a period
of increased sensation-seeking behavior, such as use of drugs
and alcohol. Nearly half of all teenagers have tried at least one
illicit drug, and three out of four have used alcohol by their
senior year.
Recent advances in developmental psychopathology have
lead to a greater understanding of many knowledge areas of
adolescent addictive disorders, including prevalence, clinical
course, and treatment. We have focused here on three core
addictive disorders for which a relatively large empirical litera-
ture exists - nicotine, other psychoactive substances, and gam-
bling. Beyond these disorders, other adolescent behavioral
'indulgences' are emerging in the research and clinical litera-
tures. These include internet addiction, excessive video game
playing, and sex addiction. Clinical, epidemiological, and
treatment perspectives on the full spectrum of addictive dis-
orders in adolescents will benefit from continued research and
clinical work.
Acknowledgments
Portions of this article have been supported by grants
K02DA0153478 (to Winters) from the National Institute on
DrugAbuse, R01AA014357 (to Chung) from the National Insti-
tute on Alcohol Abuse and Alcoholism, and 1P01CA98262
(to Kassel) from the National Cancer Institute.
See also: Adolescence; Attention Deficit Hyperactivity Disorder;
Childhood Mental Disorders; Developmental Psychopathology; Drugs,
the Brain, and Behavior.
Further Readings
Brown SA, McGue M, Maggs J, Schulenberg J, Sher K, Winters KC, and Lowman C
(2008) A developmental perspective on alcohol and youth ages 16-20. Pediatrics
121: S290-S310.
Chung T and Martin C (2010) Diagnosis, epidemiology, and course of youth substance
use and use disorders. In: Kaminer Y and Winters KC (eds.) Clinical Manual of
Adolescent Substance Abuse Treatment. Washington, DC: American Psychiatric
Association.
Dahl RE and Spear LP (eds.) (2004) Adolescent brain development: Vulnerabilities and
opportunities. Annals of the New York Academy of Sciences 1021.
Dierker L and Mermelstein R (2010) Early emerging nicotine-dependence symptoms:
A signal of propensity for chronic smoking behavior in adolescents. Journal of
Pediatrics 156: 818-822.
Kassel JD (2000) Are adolescent smokers addicted to nicotine? The suitability of the
nicotine dependence construct as applied to adolescents. Journal of Child and
Adolescent Substance Abuse 9: 27-49.
Kassel JD, Evatt DP, Greenstein JE, Wardle MC, Yates MC, and Veilleux JC (2007)
The acute effects of nicotine on positive and negative affect in adolescent smokers.
Journal of Abnormal Psychology 116: 543-553.
Stinchfield R (2002) Youth gambling: How big a problem? Psychiatric Annals
32: 197-202.
Stinchfield R and Winters KC (2004) Adolescents and young adults. In: Grant JE and
Potenza MN (eds.) Pathological Gambling: A Clinical Guide to Treatment,
pp. 69-81. Washington, DC: American Psychiatric Publishing.
Welte JW, Barnes GM, Tidwell M0, and Hoffman JH (2008) The prevalence of problem
gambling among U.S. adolescents and young adults: Results from a national
survey. Journal of Gambling Studies 24: 1 1 9-1 33.
Winters KC (2009) Adolescent brain development and alcohol abuse. The Journal of
Global Drug Policy and Practice 3: http://www.globaldrugpolicy.org/.
Winters KC, Botzet AM, Fahnhorst T, and Koskey R (2009) Adolescent substance abuse
treatment: A review of evidence-based research. In: Leukefeld C, Gullotta T, and
Addictions and Adolescence 21
http://www.monitoringthefuture.org/ - Monitoring the Future: National school based
survey of alcohol and substance use.
http://www.oas.samhsa.gov/nhsda.htm - National Household Survey on Drug Use and
Health: National annual survey of substance use.
www.drugfree.org/teenbrain/ - Partnership for a Drug Free America.
http://youthgambling.mcgill.ca/ - International Centre for Youth Gambling Problems
and High-Risk Behaviors.
Staton Tindall M (eds.) Handbook on the Prevention and Treatment of Substance
Abuse in Adolescence, pp. 73-96. New York: Springer Academic Publishing.
Relevant Websites
Adolescence
S L Gray, C L Culpepper and D P Welsh, University of Tennessee, Knoxville, TIM, USA
© 2012 Elsevier Inc. All rights reserved.
Adolescence is the transitory period between childhood and
adulthood. This period is characterized by biological, cogni-
tive, social, and psychological development. Historically, this
period of development has been described as tumultuous,
which is sometimes attributed to increased emotions and an
undeveloped prefrontal cortex. Empirical research has shown
that adolescence is not characterized by the degree of 'storm
and stress' that was once believed. Nonetheless, adolescence is
a period in which there is a tremendous amount of growth and
change in adolescents. The purpose of this article is to provide
an accurate account of adolescent development. This article
will review the primary changes that occur within an adoles-
cent as well as the contextual factors that influence and facili-
tate those changes.
Primary Changes in Adolescence
During adolescence, youth experience a multitude of changes.
These changes affect their physical appearance, cognitive devel-
opment, and emotional development. Moreover, they not only
affect the individual, but also the relationships and networks
built with others. The first part of this article examines the
primary changes that take place in adolescence, and the possi-
ble effects of those changes. It is important to note that
although these changes apply to all adolescents, the rate and
pattern of changes varies between individuals.
Biological Foundations
The transition from childhood to adolescence is clearly identi-
fied by the biological changes occurring on the inside and
outside of a child's body. These hormonal and bodily changes
make up the developmental stage called puberty, which typi-
cally occurs during early adolescence. The onset of puberty is
predominantly determined by genetic patterns, but can also be
affected by environmental factors. Historically, the onset of
puberty has been known to begin between 9 and 16 years of
age. Most commonly, puberty begins at age 1 1 for females and
at age 12 for males. The onset of puberty has decreased mark-
edly over the past decades but has recently leveled off.
Puberty is characterized by a fluctuation in hormones,
which are chemicals secreted by the endocrine glands. The
endocrine system controls the hormonal changes associated
with puberty using a negative feedback system. When hor-
monal levels fall too low, the endocrine system reacts to
the low level of hormones by causing the release of more
hormones into the bloodstream. Specifically, the hypothala-
mus, located in the brain, responds to the decreased level of
hormones in the blood system by releasing substances that
stimulate the pituitary gland, which is also located in the
brain. The pituitary gland then stimulates the gonads (the
ovaries in females and the testes in males) to release more sex
hormones. When the level of sex hormones falls too low, the
hypothalamus and pituitary gland stop stimulating the gonads
and they stop producing sex hormones. With the onset of
puberty, the levels of sex hormones in the body increase sub-
stantially. At puberty, the hypothalamus becomes desensitized
to sex hormones and requires a significantly higher level of
hormones in the body before the hypothalamus and pituitary
gland stop stimulating the gonads to release more hormones.
There are two types of hormones that are known to affect
puberty differently in males and females: androgens and estro-
gens. Androgens are male sex hormones and estrogens are female
sex hormones. Testosterone, an androgen, plays a major role in
male pubertal development; while, estradiol, an estrogen, plays
a major role in female pubertal development. Nonetheless, both
males and females have androgens and estrogens.
Weight, body fat, and a hormone called leptin are corre-
lated with the onset of puberty. Higher weight and a higher
percentage of body fat have been linked to early pubertal
development, especially in females. Adolescents living in
developing countries without enough food for adequate nutri-
tion begin puberty earlier than adolescents in developed
nations. Leptin is also thought to signal the beginning and
the progression of puberty. Levels of leptin, which are higher
in females than males, are linked to the amount of body fat in
females and the amount of the hormone androgen in males.
An increase in leptin is thought to signal an adequate amount
of body fat for reproduction.
There are some sociocultural and environmental factors
that have been linked to the onset of puberty. Researchers
have found that cultural differences and early experiences
may be correlated with the early onset of puberty. Empirical
findings suggest that the absence of a father, geographic dis-
tance between family members, child mistreatment, and low
socioeconomic status (SES) have also been linked to early
onset of puberty.
Cognitive Development
The transition from childhood to adolescence is also marked
by changes in cognitive development. There are distinct struc-
tural changes in the brain that occur during adolescence. These
changes are seen in the corpus callosum, prefrontal cortex, and
amygdala. The corpus callosum connects the right and left hemi-
spheres. During adolescence, the corpus callosum thickens,
which improves the adolescent's ability to process information.
An adolescent's prefrontal cortex, which houses reasoning,
decision-making ability, and self-control, continues to develop
during adolescence. Responsible for emotion regulation, the
amygdala matures earlier than the prefrontal cortex. Empirical
evidence has shown that although adolescents experience a vast
array of emotions, the prefrontal cortex continues to develop
well into adulthood; therefore, adolescents may not have as
much control over their emotions as they will when their pre-
frontal cortex is fully mature.
22
Adolescence 23
In the past, researchers thought that brain cells could not be
generated after early childhood, but recent research has discov-
ered that people produce new brain cells throughout their
lives. Factors such as exercise can aid the brain in producing
new cells. The adolescent brain also has some plasticity. In
childhood as well as adolescence, the brain is able to repair
itself. However, the earlier a brain injury occurs, the more likely
that recovery will be successful.
Several theories have been used to explain brain develop-
ment from childhood through adolescence. One of the most
prominent theories, articulated and investigated by Jean Piaget,
proposed that adolescents work to understand their world
because the making of meaning is hardwired and serves a
biological purpose. Adolescents construct their world using
schemas, defined as mental concepts or frameworks that help
organize and understand information. Adolescents use assimi-
lation (incorporation of new information) and accommoda-
tion (the adjustment to new information) to organize their
schemas.
Piaget articulated four stages that describe cognitive devel-
opment: sensorimotor, preoperational, concrete operational,
and formal operational stages. The formal operation stage is
typically reached during adolescence. This stage consists of
reasoning in abstract, idealistic, and logical ways and is char-
acterized by the development of problem-solving skills and
formation of hypotheses.
Lev Vygotsky introduced a theory that views knowledge as
situated and collaborative. Specifically, Vygotsky believed that
knowledge is distributed among people and their environ-
ments and knowledge can best be attained through interaction
with others through cooperative activities. Vygotsky empha-
sized the social contexts of learning. One of his most important
concepts is the zone of proximal development (ZPD). ZPD
describes a range of tasks that require assistance from adults
or mature peers.
In later years, other theories were developed that spoke of
the use of information processing and psychometric testing to
assess adolescents' cognitive development. However, Piaget
and Vygotsky's theories are still commonly used to explain
adolescent cognitive development. All of these views take
into account the genetic as well as the environmental influence
on adolescent cognitive development.
Identity
As puberty occurs and cognitive processes are developing, ado-
lescents are also forming a sense of self. An adolescent's sense of
self consists of his or her identity. Identity incorporates self-
understanding self-esteem, and self-concept. Self-understanding
is an individual's cognitive representation of the self. The devel-
opment of an adolescent's self-understanding is complex
and can be affected by many internal and external factors.
Adolescents, unlike children, describe themselves within various
contexts. Research on adolescence has shown that an adoles-
cent's sense of self can fluctuate. This sense of self can also
be contradictory, because of the multiple roles adolescents play
in different relationship contexts. One explanation for this
contradiction can be found in the real versus ideal theory.
This theory states that adolescents have an ideal or imagined
self and this ideal view of themselves contradict their actual view
of themselves. Some theorists believe that this contradiction is
maladaptive, while other theorists believe that this contradiction
can provide motivation and drive toward future goals.
Adolescents, unlike children, often compare themselves to
their peers, are more self-conscious, and more preoccupied
with their self-understanding. Even though most adolescents
employ introspection, they also look to their friends and other
peers for self-clarification and self-definitions. In Eastern
countries, adolescents may also tend to look to family mem-
bers to aid with this process as their cultural norms place a
high value on family relationships and social categories. This
search for self-clarification and self-definition can affect the
adolescent's self-esteem and self-concept. There is evidence to
suggest that adolescents have a lowered sense of self-esteem
when compared to children, and this decline usually occurs
in middle school. Male adolescents usually report higher
self-esteem than female adolescents, but this gender-gap typi-
cally decreases as adolescents move into adulthood.
Erik Erikson proposed a theory of human development
that consists of eight stages. Most adolescents, he believed, fall
within his fifth stage: identify versus identity confusion. Erikson
proposed that during this stage, adolescents have to decide their
identity, their goals and causes, and the direction of their life.
Erikson believed that adolescents experiment with many
different roles and identities while they are trying to 'find them-
selves'. He labeled this exploratory period as a moratorium period.
If adolescents use adaptive methods to explore their identity, they
will achieve identity formation. If adolescents use maladaptive
methods to explore their identity or they do not explore their
identity at all, they will experience identity confusion.
James Marcia empirically investigated Erikson's ideas and
identified four statuses, which help to simplify Erikson's theory.
The four statuses, identity diffusion, identity foreclosure, iden-
tity moratorium, and identity achievement, capture the adoles-
cents' exploration of alternatives and commitments toward a
direction in life. This constmct of identity is biased toward the
Western, masculine ideals of individualism, as opposed to
relatedness. Adolescents from Western countries may concep-
tualize identity achievement as a self-directed objective in
which individualization and independence is highly valued.
However, adolescents from Eastern countries may conceptual-
ize identity achievement as more relational and interdepen-
dent. Some research has shown that individual characteristics
are valued in Eastern countries; however, these characteristics
tend to be more collectivistic in nature and place higher values
on attending to others, fitting in, and harmonious interactions.
Although different cultures place differing values on an
individual's characteristics, the function of self-definition in
regard to identity development can be seen across cultures.
Recent views of identity development propose that it is a
gradual process that continues into adulthood. It is important
to study adolescent identity development because during ado-
lescence, youth begin to integrate the physical, cognitive, and
emotional development, forming a better understanding of
their sense of self.
Moral Development
Moral development can be defined as thoughts, behaviors, and
feelings regarding standards of right and wrong. There have
24 Adolescence
been numerous theories of adolescent moral development, but
one of the most influential theories was created by Lawrence
Kohlberg. Kohlberg's theory of moral development empha-
sizes adolescents' reasoning about moral issues. He proposed
that the way in which an individual reasons changes through-
out life and his model includes six sequential stages of
moral development. Kohlberg studied the moral reasoning
of fifty males over the course of 20 years and there was clear
evidence that each participant followed the same developmen-
tal sequence through the stages of moral development. His
participants did not skip any of the developmental stages and
there was no regression. Empirical evidence also suggests that
the stages are culturally universal, especially the first four
stages. This research also suggests that the rate of development
and the final stages that people achieve vary between indivi-
duals and groups.
Kohlberg's model consists of three levels of reasoning about
moral issues: preconventional reasoning, conventional rea-
soning, and postconventional reasoning. Within each level
are two stages. In the preconventional reasoning level, moral
reasoning is controlled by external, concrete consequences.
People functioning at this level of development make moral
decisions based on whether they will be rewarded or punished
by their decisions. Youth, in the beginning stages of adoles-
cence, are usually at this preconventional level of reasoning. In
the conventional reasoning level, moral reasoning is controlled
by internal and external factors. Moral decisions at the conven-
tional level are based on conforming to and upholding the
rules, expectations, laws, and conventions of society. Most
adolescents are said to be at this level of reasoning. The third
level of reasoning, postconventional reasoning, is thought to
be achieved at or after the age of 20. This level is characterized
by moral principles that are internalized. Moral decisions at the
postconventional level are based on understanding society's
rules, then formulating and internalizing a personal moral
code that emphasizes principles of justice: individual rights,
equality, and human dignity and a concern for the welfare of
the larger community.
Carol Gilligan has argued that Kohlberg's orientation toward
justice represents a male perspective that is more commonly
found in Western cultures. Gilligan proposed that moral devel-
opment is built on a concept of care. This type of morality
focuses on responsibility to others, values relationships, and
seeks to help. The concept of care also emphasizes sensitivity
to social context, whereas justice emphasizes the application
of rules and principles. Gilligan believes that conceptualiza-
tions of morality as care based represents a more female per-
spective, as well as the perspectives of societies that value
collectivism.
Other theories of moral development emphasize moral
behavior instead of moral reasoning. The social cognitive theory
of moral development assesses the difference between the ado-
lescent's ability to produce moral behavior (moral competence)
and his or her ability to implement moral behaviors in spe-
cific situations (moral performance). Social cognitive theorists
believe that the adolescent's moral performance is not guided by
abstract thinking, but by rewards, punishments, and motivation.
Behaviorists also emphasize moral behavior. These theorists
use imitation, rewards, and punishment to explain why adoles-
cents display different moral behaviors. When adolescents are
positively reinforced (rewarded) for a behavior that is consid-
ered moral, they are more likely to repeat that behavior.
When adolescents are punished for a morally unacceptable
behavior, they are more likely to decrease that behavior. There
are other factors and situations that mediate these processes.
Empirical research has found that adolescents will not display
the same moral behavior when put in diverse social settings.
For example, an adolescent is more likely to cheat if being
pressured by a peer. Even though other factors may affect the
outcome, overall adolescent moral behavior can be predicted
by reinforcement.
Moral development, like so many other areas, does not
begin in adolescence and will not end in adolescence. It is
important to study moral development during adolescence
because cognition, identity, and other parts of the adolescent
are actively developing during this stage as well. Adolescence is
an important developmental stage because there are so many
internal and external influences that are at work during this
period of life.
Contextual Areas of Interest Regarding Adolescent
Development
Adolescent development can be affected by numerous contex-
tual factors. The context in which an adolescent grows up, the
experiences they face, and the situations they encounter will
frame their development as they approach adulthood. Because
adolescents will experience different contextual factors under
different situations, it is important that each of these factors be
explored. The remainder of this article highlights contextual
factors regarding family, peers, romantic relationships, school,
work, and cultural concerns.
Family
One of the most important contextual factors that can affect
adolescent development is family dynamics. Research has
shown that parents have a significant effect on the ways in
which adolescents develop. Previously, adolescent relation-
ships with parents were conceptualized as tumultuous and
generally full of conflict. However, research over the past sev-
eral decades has revealed that relationships are more positive
and less tumultuous than previously considered. Although
adolescence is a time of self-exploration and emerging auton-
omy, parents who respond in an understanding manner to the
changes adolescents face generally experience less turmoil in
the home. In addition, parents who expect that adolescence
will be a tumultuous experience are more likely to have ado-
lescents who experience more conflictual family relationships.
The guidance provided by parents during this time period is
crucial in helping the adolescent transition into becoming a
responsible adult. Parents aid in this transition by helping the
adolescent form his/her own moral standards, in addition to
supporting, encouraging, and providing learning opportunities
for them.
Parenting styles
Different parenting styles may be an important contextual factor
in terms of adolescent development. Parenting styles have been
Adolescence 25
conceptualized into four different categories (authoritarian,
authoritative, indulgent/permissive, neglectful) that range on
levels of responsiveness and control. Authoritarian parents
tend to take a more unresponsive, or parent-centered role, and
are generally demanding while exerting a high level of control
over their children. Their adolescent children tend to be depen-
dent, submissive, and overly conforming in the presence of their
parents and other authorities. However, when out of the pres-
ence of authorities, these adolescents tend to be rebellious,
defiant, hostile, and resentful. In contrast to authoritarian par-
ents, indulgent/permissive parents tend to take a more respon-
sive, or child-centered role, and are generally undemanding
while exerting a very low level of control over their children.
Adolescent children of indulgent/permission parents tend to
lack self-regulation skills, and often disregard rules and regula-
tions. Because they are used to getting their way, they generally
are not as socially adept as other adolescents. Neglectful parents
are generally unresponsive to their children, while also being
undemanding and showing little control. Adolescents who grow
up in these families tend to suffer the most as they may get into a
lot of trouble and engage in more risky behavior. These adoles-
cents also tend to be self-rejecting and may feel inferior to
others. In contrast to neglectful parents, authoritative parents
are responsive, accepting, and child-centered, while also setting
clear limits for their children. This authoritative parenting style
has been found to be associated with more positive adolescent
developmental outcomes. Adolescents from these families tend
to be more socially and academically competent. They also
exhibit higher levels of self-esteem and demonstrate more per-
son control than other adolescents. It is important to note that
these parenting styles are not always completely independent of
one another. Parents may mesh different styles together and/or
use different styles with different children. Interestingly, even
within the same family unit, parents may be utilizing completely
different parenting styles. For example, an indulgent/permissive
mother, paired with an authoritarian father, may cause an ado-
lescent to experience a great deal of uncertainty and confusion.
Attachment styles
During adolescence, early attachment styles begin to shift in
accompaniment to adolescents' need for autonomy. Although,
it is normal for adolescents to feel a desire to remain close to
their parents, their emerging adolescent attachment style does
not require as much dependency on their caretaker. However,
having a secure base in which to return is key as adolescents
make the journey toward becoming more autonomous, or
independent. Early attachment theorists John Bowlby and
Mary Ainsworth put forth that developing a secure attachment
to caregivers as an infant serves as a foundation that adoles-
cents use for healthy psychological development. The develop-
ment of a secure attachment allows children to feel safe while
exploring their environment. Insecurely attached adolescents
avoid or distance themselves from their parents, show an
unusual amount of fear toward their parents, or show ambiva-
lence or resistance toward their parents. However, it is impor-
tant to note that normal adolescent attachment functioning
can falsely resemble this insecure attachment style (which has
developed from infancy and early childhood). Adolescent
attachment theorists highlight that adolescence is a period
in which attachment styles transform from hierarchical
parent-child attachments to codependent peer-peer attach-
ments (in which both persons offer and receive support).
In adolescence, with the goal of autonomy and self-other
differentiation in mind, attachments with peers become a
main focus. Even under distressing situations, adolescents
may purposefully resist seeking comfort from parents as they
are striving for autonomy. This purposeful distancing may be
experienced as uncomfortable by both parents and adolescents
at times and represents a transforming shift in attachment style
from infancy and early childhood. Because adolescents are in
a transition period, this shift in attachment needs can be
conflicting and confusing. As adolescents distance themselves
from adult caregivers, they are able to explore more of their
emotional interior and problem-solving techniques on their
own. Knowing that adult attachment figures are readily avail-
able when necessary helps facilitate this process. Early attach-
ment styles and adolescent attachment experiences with peers
can color the way in which adolescents experience the world
around them and typically guide adolescents' expectations
regarding future relationships with other people. Adolescents
who developed a secure attachment to their parents from early
childhood show more self-confidence, have better peer and
romantic relationships, and are less likely to engage in behav-
ior problems such as juvenile delinquency and drug abuse
compared with insecurely attached adolescents.
Siblings and birth order
Having siblings can greatly impact the development and
experiences of an adolescent. In American households, about
80% of adolescents have siblings. Research has shown that
siblings are often utilized as a source of support for adoles-
cents. Communication between siblings usually occurs more
naturally, and the exchange of information is more direct than
with parents. Adolescents may confide in a sibling in order to
discuss difficult or taboo topics, relying on the experience and/
or opinions of a sibling. However, although siblings may serve
as sources of support for each other, sibling conflict is not
uncommon. Chronic conflict among siblings can have a nega-
tive effect on adolescent development, especially when the
parent's relationship with the adolescent is also tumultuous.
Studies have found that depression among adolescents is
higher when sibling conflict is reported at greater frequencies.
Sibling conflict tends to be higher when parents treat siblings
very differently and favor one sibling over another. Sibling
conflict tends to decrease in adolescence compared with child-
hood probably because siblings spend less time together dur-
ing adolescence than in childhood. As adolescents begin to
emerge into adulthood, conflictual relationships with siblings
tend to further decrease, and the intensity of tension between
siblings may diminish.
Birth order has also been noted by many for its impact on
adolescent development. Although there is some inconsistency
in the findings regarding the effects of birth order, a few trends
are noteworthy. Typically, firstborns are held to a higher stan-
dard of social, academic, and professional accomplishments
than their younger siblings. This is commonly felt and
expressed by firstborns as a theme of their childhood and
adolescent years. Conversely, lastborns tend to be perceived
as the perpetual 'baby' of the family. Some siblings may com-
plain about the lastborns' ability to become self-sufficient and
26 Adolescence
independent, especially when parents are indulgent toward
their lastborn. In general, middle children are characterized
as being the negotiators of the family. Additionally, middle
children are generally noted to have problems gaining equal
attention from their parents; thus, they may engage in more
problematic or attention-seeking behavior. These trends in
birth order are not always representative of all families, and
many exceptions can be found across siblings.
Family structures: Divorce and stept amities
With the divorce rate being over 50%, many adolescents are
reared in single-family homes, blended homes, or other non-
traditional living situations. One study estimated that nearly
40% of children born into intact families will eventually
become children of divorced families. Divorce can disrupt the
connection between adolescents and parents, causing a rift
during key adolescent developmental phases in which parent
connection is desirable. Most research regarding adolescent
development in the context of divorce has shown negative
outcomes for adolescents, especially when compared to ado-
lescents from intact homes. Although most adolescents go
through a difficult period, in which their functioning is
impaired immediately after the divorce, most return to normal
functioning after a brief period of disruption. The majority
of adolescents from divorced families are functioning and
developing normally after adjusting to the family transition.
However, small subsets of adolescents continue to experience
prolonged distress. Research suggests that high continued
parental conflict following divorce is associated with poorer
adolescent functioning.
School
Adolescents spend a very large percentage of time at school
with teachers and peers. The type of school an adolescent
attends, and the experiences he or she derives from that school
will have an effect on their development and future transition
into adulthood. Specifically, school size, teachers, and peers
are factors that may contribute to an adolescent's school expe-
rience. Previously, educational psychologists focused heavily
on classroom climate and how students fit with their class-
mates. Now, researchers study the social climate of the school,
in addition to individual factors that may affect students. For
example, smaller schools have been linked to increased
amounts of prosocial behavior. Arguably, smaller classrooms
also provide an environment that encourages a higher standard
of teaching and learning. Many students may feel lost in the
crowd at very large middle and high schools. This feeling may
overwhelm many students, discouraging them from finding
their niche, or becoming involved in activities. Isolated stu-
dents may be left behind educationally and socially, which
leaves these students more susceptible to joining antisocial
peer groups or cliques.
Teachers not only serve as persons who provide formal
education for adolescents, but many times these individuals
also serve as additional role models and mentors for students.
Their classroom management strategies help students learn
time management skills and self-monitoring techniques, in
addition to learning actual classroom material. Skilled teachers
plan classroom activities in a manner that challenges students,
and generally prevents academic and emotional problems.
Adolescents who are in well-managed classrooms may show
greater levels of autonomy, self-confidence, and academic
achievement.
Peers make up a large amount of an adolescent's school
experience. While it is not uncommon for adolescents to
occasionally report feeling ostracized at school, or feeling
isolated, it is of concern if these negative experiences become
common. Peers may engage in teasing or other antisocial beha-
viors, especially in larger schools with less monitoring capabil-
ities. Bullying is the act of physically or emotionally hurting
another peer, generally at school. A national survey showed
that approximately one-third of adolescents were either being
bullied, or were themselves bullies. The most common form
of bullying is disparaging comments about looks or speech.
Almost equal to that, adolescent girls report being the subject
of sexual comments and gestures. The effects of bullying
can be harsh for some adolescents. A few studies have found
that common consequences of continued bullying involved
depression, lack of interest in school, suicidal ideation, somatic
complaints, and/or avoidance techniques. Research has also
found that a large number of boys who bully others in middle
school are likely to be convicted of a criminal offense in early
adulthood.
Although school can be a negative experience for many
adolescents, these experiences may be altered with individual
attention and focus on academic and social competence. This
is especially true for adolescents from lower SES backgrounds.
Because these adolescents may have underlying barriers to
academic achievement, additional tutoring and social integra-
tion is required. Some students may find school difficult due to
learning difficulties, or problems with attention disorders.
While larger schools tend to be less desirable due to sub-
sequent large classroom and less individualized programs,
some larger schools may be able to provide more resources to
students with these types of learning difficulties. In addition to
the formal education process, schools are an important con-
textual factor regarding adolescence because they provide a
platform for career exploration, friendship development, and
extracurricular activities.
Work
Most late adolescents have a part-time job in addition to
attending school. Working can allow adolescents to develop
professional skills, a strong work ethic, and can help with
money management. The money earned from a part-time job
also can also provide adolescents the freedom to purchase
personal consumer items (e.g., trendy clothes, music, etc.).
Depending on the place of employment, working may also
serve as another source of social integration. Many adolescents
work in establishments that also employ many of their peers.
Thus, work may allow adolescents the opportunity to socialize
with peers who are not in their network of friends.
Despite the fact that numerous benefits to working exist,
several negative factors are also associated with working. In
general, for most adolescents, as the number of hours spent
working increases, school grades suffer. This is especially true
for adolescents working more than 20 h per week. Typically,
Adolescence 27
adolescents who must balance school and work often must
sacrifice one for another. A homework assignment may go
unfinished in order to work a shift; or conversely, a work
shift may be skipped in order to pull an all-nighter on a
homework assignment. Although having a balance between
work and school can be beneficial, it starts to become prob-
lematic when ongoing sacrifices must be made in either
domain. Adolescents who work may not be able to participate
in extracurricular activities or sports that require a large time
commitment.
Peer Relationships
Friendships
Peer relationships, including close friends and romantic part-
ners, are an extremely important contextual factor for adoles-
cents. Different from childhood, adolescents spend a much
greater amount of time interacting with their peers. One
study found that adolescents were spending about 2 times as
much time with their peers than their parents, even over week-
ends, outside of formal school time. Adolescents have the most
contact with their peers during school hours; however, they
also spend time with friends after school, and may develop
friendships through church, work, or extracurricular activities.
Peers relationships serve numerous purposes for adoles-
cents. A basic purpose of peer interaction is social integration,
ft is important for adolescents to feel wanted and included in
social groups. Persons who are socially integrated show fewer
psychological signs of distress, including depression. Peer rela-
tionships also provide a context for developing social skills.
Adolescents learn how to interact with others by learning how
to maintain their own friendships. Peer relationships allow
adolescents to navigate through conflict management, listen-
ing, empathy, and intimacy skill building.
The importance of friendship can be summarized within six
basic domains: companionship, stimulation, physical support,
ego support, social comparison, and intimacy. The variety of
these domains demonstrates the importance of close friendships
for adolescents. Contributing more heavily to the developmen-
tal goals for adolescence, ego support, social comparison, and
intimacy are particularly noteworthy. Friends provide reassur-
ance, support, and encouragement during times of uncertainty
for many adolescents. This ego support allows adolescents to
maintain a reasonable level of self-esteem. Friends also provide a
source of social comparison, such that adolescents can gauge
themselves against friends. Adolescents engage in social compar-
isons for a variety of reasons, including identity development,
and maybe interested in how their academic achievement, body
image, popularity, and romantic involvement compares to
others. Lastly, adolescents are starting to develop intimate,
close relationships with their peers. These intimate relationships
allow adolescents the ability to explore self-disclosure and trust.
Because adolescent peer relationships serve so many impor-
tant functions, it is also important to understand why many
adolescents do not achieve close friendships. Sociometric sta-
tus refers to the degree to which adolescents are liked or dis-
liked by their peers. This concept is gauged by asking
participants to rank order and nominate other peers for
'most-liked' and 'least-liked' roles. Five different categoriza-
tions of peers have been established: popular, rejected, and
neglected. Popular individuals are often nominated as some-
one's best friend, and typically do not receive many 'least-liked'
votes. They tend to be good communicators, enthusiastic, confi-
dent, and happy. Their internal states and external behaviors
both contribute to their ease in developing close friendships.
Conversely, rejected individuals receive many 'least-liked' votes
and few best friend votes and tend to be characterized as aggres-
sive, arrogant, tactless, and annoying. Neglected adolescents do
not receive either 'best friend' or 'least-liked' votes and tend to be
overly timid, shy, or lacking in enthusiasm. Rejected adolescents
tend to be lonelier than neglected adolescents and are more
likely to develop psychological problems. As loneliness is a
rather subjective experience, some research suggests that there
are cultural differences regarding how isolation is experienced.
However, adolescents do universally demonstrate a need for
belonging and acceptance.
Groups, crowds, and cliques can develop as a means of
inclusion, and exclusion, for many adolescents. Groups tend
to form around a similar purpose, activity, or special interest.
For example, athletic teams, religious groups, and clubs can all
make up a group. These groups have social norms and roles
that members follow in order to be included. Groups can
infuse a sense of companionship, purpose, and accomplish-
ment among adolescents. Group membership may provide a
certain level of intrinsic satisfaction, in addition to extrinsic
rewards for participation. Groups comprising different clubs,
organizations, and teams are typically associated with positive
psychosocial developmental outcomes.
Crowds are entities that tend to form based on reputation,
and membership may be involuntary. These classifications make
up terms such as the, 'jocks,' 'goths,' 'preppies,' etc. Adolescents
may belong to a crowd based on the type of behaviors they
engage in (e.g., 'druggies'), their SES standing (e.g., 'preppies'),
or their level of social integration (e.g., 'nobodies'). These large
collections of individuals may not interact and/or socialize
together, although smaller branches of friendships (groups or
cliques) may develop within the crowd.
Cliques are smaller, more intimate, sets of friends that may
have formed based on inclusion in a larger group or crowd.
Cliques tend to involve no more than 12 people (typically
about 5 or 6 people), and mostly comprise adolescents who
attend school together. Both positive and negative outcomes
can be associated with the development of cliques. Because of
the in-group and out-group processes, some adolescents may
feel socially isolated if they are not members of a social clique.
Cliques may also encourage antisocial behavior, such as teas-
ing, toward others who are not members of a particular clique.
However, cliques do allow small groups of friends to become
closer, which can serve as an additional support system for
many adolescents.
Peer relations are also influenced by gender, SES, and culture.
Adolescents tend to select friends who are similar to themselves
in age, gender, SES, and ethnicity. Girls tend to focus more on
interpersonal interactions within their friendship, striving to
maintain high levels of intimacy and self-disclosure; while
boys tend to focus more on activities, giving less attention to
self-disclosure and intimacy. Adolescents from middle and
upper SES groups tend to be more socially integrated and take
more leadership roles within friend groups. Those from lower
SES groups, or those from minority groups, may seek friendship
28 Adolescence
among themselves. Groups and cliques that form based on
minority or cultural status may serve as a source of social inte-
gration and to ward off isolation.
Romantic relationships
In addition to the development of more intimate and consis-
tent bonds with peers for the purpose of developing friendship,
adolescence is a time in which peers can also serve as possible
romantic partners. It is not uncommon for adolescents to
identify themselves as 'dating,' 'having a boyfriend/girlfriend/
or 'going-out with someone.' One study found that among
17- year-olds, ~70% reported that they were currently in a
romantic relationship, or had been within the previous year
and a half. Thirty-six percent of 13-year-olds, and 53% of
15-year-olds, also reported romantic relationships in that
study. Comparisons across studies have been difficult due to
differing language usage among researchers. For example,
some studies may ask, 'Are you in a serious relationship,'
while other studies may ask, 'Are you involved romantically
with another person?' These questions may elicit different
responses, thus making the empirical comparison of data diffi-
cult. Nevertheless, numerous studies have emphasized the
importance of understanding the development of romantic
relationships in adolescence.
Adolescent romantic relationships are characterized as hav-
ing many different functions. In addition to courtship for the
purposes of marriage, these relationships can also serve as a
platform to further explore the socialization process. Dating
may also assist with building intimacy, companionship, iden-
tity, and achievement. Further, adolescent romantic relation-
ships provide an arena for sexual experimentation and sexual
identity development. Adolescence is a time of exploration
within numerous fields, and romantic relationships can aid
in that exploration process.
Early adolescents (ages 10-13) tend to spend more time
simply thinking about potential romantic interests rather than
actually interacting with them. At this time, adolescents are still
acquiring basic skills regarding how to interact with potential
romantic partners. As they are exploring their initial attractions
to romantic partners in general, they are also making social
comparisons to their friends' 'progress' in this dimension. Early
adolescents who get involved early in dating and sexual rela-
tionships are at risk for problematic developmental pathways
involving problems with alcohol, drugs, truancy, and educa-
tional underachievement. Middle (ages 14-17) and late (ages
18- 21) adolescents tend to engage in more formal dating
practices and sexual behaviors. Adolescents who begin dating
later rather than earlier, and engage in sexual behaviors only in
the context of dating relationships, are not more likely to show
problematic behaviors than their nondating peers. Findings
also have suggested that 'light' sexual behaviors (e.g., hugging,
kissing, etc.) among adolescents are linked to more committed
relationships and healthier relationships with parents. Heavy
sexual behaviors (e.g., intercourse) were not linked to any
higher levels of psychological distress or behavior problems
in late adolescence. Late adolescents may engage in sexual
behaviors as a sign of their commitment and intimacy.
Most research regarding adolescent romantic relationships
and sexual behaviors focus on mixed-gender relationships.
However, although many sexual minority adolescents in
same-gender romantic relationships will undergo similar
experiences, these adolescents may also experience more diffi-
culties interpreting their feelings and behaviors within the
context of societal norms and expectations. Research has
shown that many individuals identify their first same-sex
experience (i.e., attraction, behavior, self-labeling) during
adolescence. Unfortunately, many sexual minorities risk
experiencing verbal and physical abuse, which may encourage
sexual minority adolescents to feel uncomfortable proclaiming
their romantic and sexual orientations.
Culture and Social Trends
From a holistic perspective, the concept of culture encompasses
any specific group's way of thinking and being including their
cultural norms, values, and beliefs that are passed on from one
generation to the next. Not all individuals who associate with a
particular culture will continuously identify with their group's
norms, values, and beliefs. It is not uncommon for individuals
to question or reject certain aspects of their cultural back-
ground. These rejections, or cultural violations, may not be
well received by others within their cultural group. However,
individuals may develop a strong sense of ethnocentricity
especially during adolescence. In other words, they may
begin to favor their own cultural group over others.
SES and ethnicity
Historically and globally, two groupings that have been com-
monly used to establish culture are ethnicity and SES. Ethnicity
is typically based on an individual's race, religion, heritage,
and/or language. An individual's SES, or class, is typically
based on an individual's economic, occupational, and/or edu-
cational status. Because adolescents often have similar occupa-
tional and educational statuses, their SES classification is based
upon their parent's status within these areas. Research has
shown that an adolescent's ethnicity and SES may have an
effect over their development. One study found that adoles-
cents who come from impoverished families, and who reside
in crime-ridden neighborhoods, are at greater risk for negative
developmental outcomes, including higher levels of psycho-
logical distress and poor academic success. Additionally,
research has also shown that parents from higher SES groups
are less likely to utilize authoritarian parenting styles and
physical punishment. Conversely, lower SES parents were
found to utilize more directive parenting strategies and physi-
cal punishment.
The culture of SES can also affect an adolescent's develop-
ment by encouraging or discouraging self-efficacy and self-
gratification. For example, parents from higher SES groups
are more likely to encourage their children to explore, take
initiatives, and delay gratification. Self-exploration and delayed
gratification expectations are likely to encourage children to stay
in school longer, and obtain higher paying jobs that value
critical thinking skills, autonomy, and creativity. Conversely,
parents from lower SES groups are more likely to be concerned
with conformity and discipline. These trends within lower SES
communities may stem from concerns regarding employment,
safety, and a general ability to care for self and family. Addition-
ally, persons from lower SES communities typically hold jobs in
Adolescence 29
which autonomy and creativity are not valued. Therefore, the
ability to conform and be disciplined is highly desirable. It is not
uncommon for parents to pass these values and expectations,
based out of SES backgrounds, onto their children. Notably,
however, parents from lower SES groups who place a high
value upon education are more likely to have children who
achieve a high level of academic success, which can later provide
higher paying careers.
In addition to SES, ethnicity also may play a significant role
in adolescents' lives. Because ethnic minority groups are more
likely to experience poverty, many issues surrounding ethnicity
tend to be confounding factors that overlap with SES issues.
Ethnic minorities may be unjustly categorized as 'inferior' in
educational or occupational realms as a result of their ethnic
makeup. However, it has been commonly found that environ-
mental and economic factors account for a large amount of
ethnic minorities' difficulties within these areas. Nevertheless,
ethnicity within itself may also play a significant role in ado-
lescents' lives. Ethnic minority adolescents are often at risk for
experiencing the effects of negative stereotypes and prejudice.
These experiences, which may include unwarranted teasing
and harassment, can cause ethnic minority adolescents to
experience greater levels of psychological distress and feelings
of inadequacy.
Media
Today's adolescent is likely to be extraordinarily technologi-
cally savvy. Adolescents are large consumers of a variety of
different media outlets, including television, computers, cell
phones, video games consoles, CD/MP3 players, and digital
cameras. Nearly a culture within itself, technology has shaped
and guided the way in which adolescents experience the social
world around them. With approximately two-thirds of adoles-
cents having a television in their bedroom, it is not surprising
that time spent watching television still takes up the greatest
portion of an adolescent's daily activities outside of school.
Increased levels of television consumption are generally asso-
ciated with lower academic success, more passive learning styles,
and less active lifestyles. Generally, late adolescents make a shift
from television consumption toward other media outlets, such
as computers and CD/MP3 players. Researchers have recently
noted a significant increase in the amount of time adolescents
spend on the Internet. Many adolescents use the Internet for a
variety of reasons, including entertainment, social, and academic
purposes. Although Internet use can be extremely beneficial to
adolescents, certain concerns regarding exposure to inappropri-
ate material exist. Among adolescents, about 44% have visited
a sexually explicit adult website; and about 25% have visited a
website containing racially motivated hate material. Many ado-
lescents use the Internet to visit social networking websites,
such as Myspace and Facebook. These websites allow adoles-
cents the opportunity to socialize and interact with peers over
an electronic format. Although enjoyed by the majority of
adolescents, these websites can also facilitate passive-aggressive
'web-fights,' or cyber-bullying between peers, and may serve as
a source of stress for a portion of adolescents.
Overall, culture and social trends not only have an effect on
adolescent development independently, but also can affect how
other contextual factors are manifested. Families, friendships,
romantic relationships, work, and school can all be affected by
an adolescent's culture and prevailing social trends. For exam-
ple, an adolescent's culture may dictate certain expectations
regarding romantic relationships and dating scripts to which
adolescents must adhere. Social trends, such as the popularity
of cell phones, texting, and the use of social networking websites
(i.e., Facebook and Myspace) greatly affect the manner in which
adolescents communicate with parents and peers.
Summary
Adolescence is a transitory period in which an individual shifts
from childhood into adulthood. For most, this period is a time
of physical, social, and emotional changes. The development
that occurs within adolescence is key in setting the stage for
adulthood. The age range when an individual will experience
adolescence will vary depending on maturity and cultural
norms. However, researchers have recently begun to conceptu-
alize adolescence as spanning more years than previously
believed. Because adolescence is filled with ambiguity and
uncertainty in many areas, self-exploration and identity devel-
opment are important goals within this transitory period.
Although a portion of adolescents will experience a significant
amount of turmoil during this time that may interfere with
healthy development, most will make the transition into adult-
hood successfully.
See also: Addictions and Adolescence; Family Systems; Friendship;
Moral Development; Parent-Offspring Conflict; Risk-Taking Behavior
(Young Male Syndrome); Social Development (Attachment,
Imprinting).
Further Reading
Adams GR and Berzonsky MD (eds.) (2003) Blackwell Handbook of Adolescence.
Oxford: Blackwell Science.
Collins WA (2003) More than a myth: The developmental significance of romantic
relationships during adolescence. Journal of Research on Adolescence 13: 1-24.
Collins WA, Maccoby E, Steinberg L, Hetherington EM, and Bornstein M (2000)
Contemporary research on parenting: The case for nature and nurture. American
Psychologist 55: 218-232.
Collins WA, Welsh DP, and Furman W (2009) Adolescent romantic relationships.
Annual Review of Psychology 60: 631-652.
Engels R, Kerr M, and Stattin H (eds.) (2007) Friends, Lovers and Groups: Key
Relationships in Adolescence. London: Wiley.
Larson RW (2000) Toward a psychology of positive youth development. American
Psychologist 55: 170-183.
Larson RW, Clore GL, and Wood GA (1999) The emotions of romantic relationships: Do
they wreak havoc on adolescents? In: Furman W, Brown BB, and Feiring C (eds.)
The Development of Romantic Relationships in Adolescence, pp. 19-49. New York:
Cambridge University Press.
Lerner RM and Galambos NL (1998) Adolescent development: Challenges and
opportunities for research, programs, and policies. Annual Review of Psychology
49: 413-446.
Lerner RM and Steinberg L (eds.) (2004) Handbook of Adolescent Psychology.
Hoboken, NJ: Wiley.
Muuss RE (1995) Theories of Adolescence, vol. 6. New York: Random House.
Smetana JG, Campione-Barr N, and Metzger A (2006) Adolescent development in
Interpersonal and societal contexts. Annual Review of Psychology 57: 255-284.
Steinberg LT and Morris AS (2001) Adolescent development. Annual Review of
Psychology 52: 83-110.
Adrenal Glands
L A Trevino, M L Uhelski, A L Dougall, and A Baum 1 ', The University of Texas at Arlington, Arlington, TX, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Anabolic process A metabolic process that builds
molecules from smaller units.
Catabolic process A process that releases energy in the form
of ATP through the breakdown of larger molecules into
smaller units.
Catecholamines A class of neurohormones, including
epinephrine, norepinephrine, and dopamine, that are
involved in various central and peripheral processes.
Glucocorticoids A class of steroid hormones that act to
maintain glucose homeostasis and circadian rhythm,
contribute to carbohydrate metabolism, and are involved in
arousal, stress response, and cognitive functions such as
mood and memory.
Gluconeogenesis The production of new glucose molecules
that are utilized in metabolic processes.
Hyperplasia Abnormal increase in proliferation of
cells in organs or tissues that is above what is
considered normal.
Metabolism The sum of all catabolic and anabolic processes
that occur within an organism to help sustain life.
Type II diabetes A disorder that is characterized by insulin
resistance brought on by chronically elevated blood glucose
levels, stress, obesity, and other factors.
Introduction
The adrenal gland and its hormones are responsible for many
functions that ultimately impact human behavior. The release
of adrenal hormones into the peripheral bloodstream impacts
global functioning, starting with changes in cellular metabo-
lism and expanding to include alterations in mood, alertness,
and attention that in turn lead to measurable changes in
behavior. Examining the influence of hormones released
from the adrenal gland on homeostatic processes, as well as
the consequences of disease states that hinder normal adrenal
functionality, demonstrates the importance of this endocrine
gland and its many chemical messengers.
The purpose of this article is to discuss the role of the
adrenal gland and its function in human behavior. An over-
view of the organization and anatomy of the adrenal gland
is given, followed by an explanation of the hormones asso-
ciated with the different parts of the adrenal gland. Hormone
synthesis and release and the regulation of glucocorticoids,
aldosterone, androgens, and catecholamines are described.
The impacts of these hormones on metabolic functions,
mood, and reproduction are also discussed.
Anatomy and Physiology of the Adrenal Gland
The adrenal glands are small endocrine organs located directly
above the kidneys. The adrenal gland is made up of the inner
adrenal medulla encased by the outer adrenal cortex. These
regions are morphologically distinct and produce hormones that
play important roles in essential physiological processes, includ-
ing stress response, regulation of blood pressure, and production
of sex hormones. The following sections discuss the hormones
associated with the adrenal cortex and the adrenal medulla.
The synthesis of these hormones and their receptors is discussed,
as well as the connections of the adrenal cortex and adrenal
^Deceased.
medulla to the hypothalamus-pituitary-adrenal (HPA) axis and
the sympathetic nervous system (SNS), respectively (Figure 1).
Adrenal Cortex
The cortex of the adrenal gland is responsible for the synthesis
and secretion of glucocorticoids, aldosterone, and dehydroepi-
androsterone (DHEA). It consists of three layers: the zona
glomerulosa, the zona fasciculata, and the zona reticularis.
Each layer is responsible for different functions associated
with the synthesis and release of certain hormones. The zona
fasciculata is responsible for the storage and release of gluco-
corticoids; the zona glomerulosa is responsible for the synthe-
sis and release of mineralocorticoids; and the zona reticularis
is responsible for the release of androgens.
Glucocorticoids
Glucocorticoids (i.e., cortisol and corticosterone) are produced
in the zona fasciculata of the adrenal cortex and are synthesized
from plasma cholesterol. Glucocorticoids are vital to the circa-
dian rhythm, glucose metabolism, arousal, and cognitive func-
tions, such as memory and mood. Cortisol is also the main
hormone associated with the physiological stress response in
humans, whereas corticosterone is the hormone equivalent to
cortisol in most rodents. However, in humans, corticosterone
is a precursor in the synthesis of aldosterone, which is dis-
cussed later. When cortisol is synthesized, it begins with the
conversion of cholesterol to pregnenolone and then it is con-
verted tol7a-hydroxypregnenolone. This is then converted to
17a-hydroxyprogesterone and then to II-deoxycortisol, which
is converted to cortisol.
Glucocorticoids are released following activation of the
HPA axis and are considered an integral part of the body's
response to a stressor. Though there has been much debate
about what a stressor is in relation to the individual, for the
purposes of this article a stressor is defined as either an envi-
ronmental or psychological occurrence that activates the
stress response (i.e., the HPA axis and the sympathoadrenal
30
Adrenal Glands 31
Figure 1 Overview of the hypothalamus-pituitary-adrenal axis and
sympathetic nervous system in relation to the stress response. CRH,
corticotropin-releasing hormone; ACTH, adrenocorticotropin hormone;
GCs, glucocorticoids; E, epinephrine; NE, norepinephrine.
medullary (SAM) system). When a stressor is perceived, the
paraventricular nucleus (PVN) of the hypothalamus is activated,
causing the release of corticotropin-releasing hormone (CRH).
CRH is released through the blood into the hypothalamic-
pituitary portal system, where it travels to the pituitary gland.
Once in the anterior pituitary gland, adrenocorticotropin
hormone (ACTH) is released. ACTH travels via the bloodstream
to the adrenal cortex where it binds to the melanocortin 2 recep-
tors (MC2-R), activating glucocorticoid release via a second
messenger pathway.
The release of cortisol from the adrenal cortex is regulated
by the ACTH that is released from the anterior pituitary gland
following release of CRH from the hypothalamus. When ACTH
is released from the pituitary gland, ACTH binds to the MC2-Rs
located on the zona fasciculata cells of the adrenal cortex. This
receptor is a G protein-coupled receptor, which when activated
causes an increase in intracellular cyclic adenosine monopho-
sphate (cAMP) to activate protein kinase A (PKA). Activation of
PKA causes two distinct cellular responses that allow for more
cholesterol to be used for the synthesis of adrenal hormones.
First, this converts cholesterol esters into free cholesterol that
can be used for creating glucocorticoids. Second, the increase in
free cholesterol increases the steroidogenic acute regulatory
protein (StAR) expression, allowing for an increase in choles-
terol being transported to the mitochondria, which is then
converted into pregnenolone.
Once released from the adrenal cortex, glucocorticoids (pri-
marily cortisol in humans; corticosterone in rodents) travel
throughout the body via carrier proteins in the bloodstream.
Normally, the vast majority of cortisol in humans travels
bound to transcortin (also called corticosteroid-binding pro-
tein, CBP), an a globulin protein, or by serum albumin, a
nonspecific steroid hormone carrier. These carrier proteins
are produced primarily in the liver instead of the adrenal
gland and can be influenced by liver disease as well as estrogen
levels during pregnancy. In addition to cortisol, transcortin
also binds to progesterone, aldosterone, and its precursor
1 1 -deoxycorticosterone, while serum albumin is involved in
maintaining osmotic pressure as well as binding to substances
such as fatty acids, thyroid hormones, and other fat-soluble
compounds. Normally, cortisol remains bound and inactive
until reaching the target tissue, with less than 10% of available
hormone free to bind glucocorticoid receptors (GRs) so that
the pool of available hormone remains steady in spite of
changes in the level of hormone being released from the adre-
nal cortex. However, under conditions of severe stress or
during certain disease states (such as sepsis), the level of glu-
cocorticoid production may increase above the capacity of
available carrier proteins, increasing the proportion of free
hormone and enhancing its impact.
Once free glucocorticoids reach a target tissue, they bind to
intracellular transcription factor proteins known as GRs, which
are part of the family of nuclear receptors that bind steroid and
thyroid hormones, as well as vitamin A metabolites. Typically,
these receptors remain in the cytoplasm, bound to a hetero-
oligomeric group of proteins that occlude the DNA-binding
regions but leave the ligand-binding site open. When gluco-
corticoids enter the cell and bind to the GR, a transformational
change occurs and the GR dissociates from the protein com-
plex, exposing the DNA-binding regions. The receptor enters
the cell nucleus and binds to the promoter regions of target
genes that contain glucocorticoid-response elements (GREs)
that influence the expression of several gene products such
as those involved in immunosuppression, anti-inflammatory
activities, and the creation of adipose cells. After binding has
occurred, the receptor dissociates from the GRE and remains in
the nucleus for an extended period before returning to the
cytoplasm. There are also nongenomic effects of glucocorti-
coids, which occur rapidly and can affect cell membrane prop-
erties through activation of membrane-bound or intracellular
GRs. Although the mechanisms of these effects are not entirely
clear, studies of isolated synapses have revealed unique bind-
ing sites for steroid hormones in presynaptic terminals that
could directly modify neurotransmitter release and other activ-
ities dependent on calcium levels directly. For example, the
administration of exogenous glucocorticoids led to increased
aggression in male rats and enhanced lordosis responses in
females within minutes, an effect not seen when other steroid
hormones were administered (such as testosterone or estra-
diol). Such effects are considered nongenomic because they
occur much faster than processes that rely on the GR complex
and because pharmacological blockades of GRs do not elimi-
nate these effects.
Cortisol is one of the few hormones that have a circadian
rhythm associated with basal levels. This hormone has an
increase in levels just before a person wakes up in the morning,
roughly 1-2 h before awakening. Cortisol levels continue to
oscillate throughout the day, with their lowest levels being
around the beginning of the sleep cycle, usually the beginning
of the dark cycle. This rhythm was once believed to be regu-
lated by the suprachiasmatic nucleus (SCN), which received
input about the light/ dark cycle via input from the retina. It has
been shown that those who are completely blind do have
circadian rhythms controlled by the SCN; however, their
rhythms are usually either shorter or longer in duration than
those who have functioning retinas. In addition, lesion studies
of the SCN have shown that the cortisol circadian rhythm
continues to occur, indicating that this circadian rhythm is
32 Adrenal Glands
not solely regulated and controlled by the SCN. It is unclear
how this particular rhythm is maintained without the SCN and
what additional areas of the brain may be involved in the
maintenance of circadian rhythms, specifically the cortisol
rhythm.
The effects of acute stressors tend to differ from those seen
following chronic stress. The stress response system was de-
signed to maintain homeostasis during times of crisis, but
when chronically activated, dysregulation and damaging ef-
fects may occur. The hippocampus is particularly vulnerable
to the negative effects of chronic stress. Trauma in early life is
associated with dysregulation of the HPA axis such that these
individuals are more prone to high stress reactivity and blunted
negative glucocorticoid feedback, as well as increased vulnera-
bility to anxiety, mood, and substance abuse disorders. The
extent of this dysregulation can be seen in animal subjects
exposed to early life stress, where evidence of neuronal loss
and reduced dendritic volume is present in the adult hippocam-
pus. The loss of hippocampal neurons is also associated with
chronic illnesses such as depression, where it is associated
with cognitive dysfunction. Normally, the hippocampus can
regenerate neurons and dendrites to repair damage following
trauma, but in the brains of individuals under chronic stress
or disease, these processes are blunted and damage persists.
Glucocorticoids and behavior
Glucocorticoids not only play a vital role in stress response but
also produce widespread effects physiologically and behavior-
ally. As described earlier, the level of glucocorticoids in the
bloodstream fluctuates based on circadian rhythms. The main-
tenance of a normal circadian pattern of glucocorticoid release
is necessary for physical and mental health.
Individuals with inherited or acquired pathologies of the
glucocorticoid system suffer from a variety of negative symp-
toms and physical effects. The following section examines the
behavioral impact of glucocorticoids in normal individuals, as
well as the consequences of disease states affecting this system.
Several genetic mutations that result in varying levels of
transcortin deficiency, which either decrease the binding af-
finity for cortisol or result in complete loss of transcortin pro-
duction, have been discovered in humans. Clinically, these
individuals presented with depression, fatigue (especially
when stressed), hypotension, and other symptoms of adrenal
insufficiency, despite tests demonstrating that cortisol levels in
the bloodstream were low-normal or normal. These genetic
anomalies could be a possible factor in poorly understood
conditions, such as chronic fatigue syndrome and some idio-
pathic chronic pain disorders, and the importance of maintain-
ing normal free glucocorticoid levels should be emphasized in
clinical settings.
Normal GR functionality is vital for an effective and timely
response to glucocorticoid release. In humans, mutations in
the gene coding for the receptor can selectively alter the bind-
ing properties of the GR, leading to glucocorticoid insensitivity
in target tissues that express mutated GRs. This condition leads
to excess activation of the HPA axis, increased levels of ACTH
release, adrenal hyperplasia, and excessive production of all
cortical steroids. In some of the milder cases, the mutation
results in lower GR-binding efficacy and the effects are benefi-
cial to the individual: increased glucocorticoid resistance.
enhanced insulin sensitivity, lower fasting blood sugar, lower
cholesterol levels, and a lower risk of dementia, cardiovascular
disease, and type II diabetes. In addition, male carriers of these
types of mutation tend to be taller and more muscular while
females tend to have smaller waist sizes and weigh less. On the
other hand, mutations that lead to enhanced glucocorticoid
sensitivity tend to be less beneficial, producing a metabolic
phenotype that increases the risk of cardiovascular disease
and is associated with higher BMI, elevated cholesterol levels,
and lower bone density. Individual variation in GR sensitivity
may have a profound effect on overall health, and further
clinical examinations could reveal important targets for treat-
ment in affected individuals.
Glucocorticoid effects on metabolic activity Although the
behavioral response to glucocorticoids is difficult to isolate
from the effects of catecholamines in humans, animal studies
can examine the impact of exogenous glucocorticoids in adre-
nalectomized or normal subjects. Such studies have revealed
important physiological and behavioral aspects of the gluco-
corticoid system. Under normal conditions, glucocorticoids
influence metabolic processes and liver enzyme production.
Depending on the target tissue, these activities maybe anabolic
(requiring energy) or catabolic (producing energy). In the liver,
glucocorticoids increase the production of enzymes necessary
for gluconeogenesis (an anabolic process) while the response
in muscle and adipose tissue increases the breakdown of
carbohydrates, proteins, and lipids for energy in the form
of adenosine triphosphate (ATP) (a catabolic process). Gluco-
corticoids also inhibit the activities of insulin, preventing glu-
cose from entering cells and enhancing catabolic processes as
cellular glucose levels fall. The amino acids and/or fatty acids
that are by-products of catabolic breakdown exit the cell and
can be utilized in the liver for gluconeogenesis. Chronically
elevated glucocorticoid levels (under chronic stress conditions
or glucocorticoid resistance) lead to persistent hyperglycemia
and increase vulnerability to insulin resistance and type II
diabetes.
Glucocorticoids are necessary to maintain homeostasis and
they also play an important role in recovery in the aftermath of
injury or disease. Glucocorticoids are involved in catechol-
amine synthesis and reuptake in sympathetic nerve terminals,
which regulate the level of vascular smooth muscle tension in
response to changes in blood pressure and enable blood flow
to reach active tissues (such as skeletal muscle) during acute
stress. Glucocorticoids also demonstrate permissive effects on
the sympathoadrenal system during acute stress, allowing nor-
epinephrine (NE) and epinephrine (E) to exert full effects on
the liver, blood vessels, and other tissues. The processes that
occur during the initial alarm phase of acute stress require an
adequate store of liver glycogen for breakdown; free fatty acids
may be used to perform gluconeogenesis, but persistent activa-
tion of the HPA axis can lead to loss of muscle tissue, hypergly-
cemia (leading to type II diabetes), and deterioration of
immune and vascular tissues.
Endocrine functions are also influenced by social factors
in human and animal populations. In societies where a strict
social hierarchy exists, lower-ranking members demonstrate
adrenal hyperplasia, deterioration of immune tissues, and
increased vulnerability to disease. In addition, excess release
Adrenal Glands 33
of glucocorticoids impairs the functioning of the reproductive
system. In females, high levels of cortisol are associated with
increased risk of miscarriage and infertility, which influences
individual reproductive success.
The impact of glucocorticoids on reproductive be-
havior involves both physiological and psychological effects.
Glucocorticoids can act directly on the hypothalamus and
pituitary gland to inhibit the release of gonadotrophins
or gonadotrophin-releasing hormone or interact with gonadal
tissues to alter the response to luteinizing or follicle-
stimulating hormones. The administration of exogenous
glucocorticoids is associated with decreased gonadotrophin
release, an effect which also appears in response to chronic
stress but is less consistent during acute stress. Cortisol has also
been shown to enhance negative feedback from estrogens and
reduce its enhancement of gonadotrophin-releasing hormone
receptor expression.
In females, inhibition of luteinizing hormone (LH) activity
leads to inconsistencies in the reproductive cycle, including an
extended follicular phase and impaired uterine maturation
during the luteal phase. These effects reduce the probability
of egg fertilization and implantation. Higher levels of cortisol
are also associated with an increased risk of miscarriage during
early pregnancy. Increased glucocorticoid release also has a
marked effect on sexual receptivity and proactive behaviors
that are likely due, in part, to increased vigilance and anxiety
during stress as well as decreased estrogen levels in the blood-
stream. Normally, estrogen acts to facilitate genital sensitivity
to tactile stimulation, which enhances sexual receptivity.
For pregnancies that are carried to term, maternal stress has
been associated with shorter gestation periods and smaller
babies. Infants born to mothers with high cortisol levels during
the third trimester tend to show more signs of negative affect
(negative facial expressions, crying/ fussiness) and tend to be
rated as more difficult by their mothers. Maternal behaviors are
also altered, with several species demonstrating less affection
and infant-directed behaviors in mothers with high glucocorti-
coid levels. The combination of prenatal stress and distant
parenting also influences the stress reactivity of the infants,
causing enhanced emotionality and anxiety during infancy
and increased susceptibility to mood disorders and substance
abuse in adulthood.
In males, the effect of glucocorticoids on the gonadal hor-
mones is similar, with a significant decrease in testosterone as
a result. Although deficits in testosterone are not sufficient to
impair fertility, reproduction is inhibited by the disruption of
erectile function. Inhibition of the parasympathetic system dur-
ing stress also prevents the development of an erection, and
excess sympathetic activity can result in premature ejaculation if
an erection forms. These effects, in addition to decreased libido,
have a severe impact on reproductive capacity in males.
Mood The HPA axis influences cognition and mood through
GRs and MRs located in the prefrontal cortex, hippocampus,
amygdala, and limbic system. While the administration of
exogenous glucocorticoids or GR agonists has been shown to
enhance acute cognitive performance in some laboratory tests,
assessments following chronic administration produced evi-
dence of impaired working memory, verbal memory, learning
(as assessed by repeated administration of a verbal memory
task), and memory retrieval. The effects of acute administration
may be a result of temporarily enhanced vigilance and atten-
tion that is associated with the acute stress response, while
repeated administration mimics the detrimental effects that
result from chronically elevated glucocorticoid levels. Gluco-
corticoids disrupt normal metabolic activity in the hippocam-
pus, a process that increases neuronal loss following stroke or
seizure and increases the rate of loss as an individual ages.
Glucocorticoids have also been implicated in increased vulner-
ability to neurotoxins and glutamate excitotoxicity.
GRs in the limbic system and prefrontal cortex enable HPA
axis activity to influence mood and other psychological states.
Disturbances in the HPA axis and glucocorticoid system are
associated with mood disorders, with multiple studies demon-
strating elevated cortisol levels and impaired GR negative feed-
back in patients diagnosed with mood disorders. Although
acute administration of glucocorticoids sensitizes serotonin
(5-HT)-lA autoreceptors, chronically high levels of glucocorti-
coids lead to desensitization and impaired functioning of
these receptors has also been identified in patients with
mood and anxiety disorders. Elevated cortisol appears to be a
sign of poor prognosis, and these patients seem to be particu-
larly resistant to treatment and more vulnerable to relapse. The
relationship between glucocorticoids and depression is of par-
ticular interest due to the pattern of HPA axis dysregulation
seen in patients with major depression. While older depression
models focus on serotonin and NE, more recent studies have
presented clear evidence of the relationship between the HPA
axis and depression, as well as many potential targets for novel
treatment. Individuals with major depression demonstrate ele-
vated blood cortisol levels as well as glucocorticoid resistance,
a condition that interferes with normal negative feedback path-
ways. Experimental treatment with substances that alter HPA
axis activity has demonstrated positive results, but further stud-
ies are needed to identify the best possible targets and to
evaluate any potentially negative effects of long-term use. In
addition to HPA dysregulation, major depression is also asso-
ciated with progressive degeneration of the hippocampus,
amydgdala, and prefrontal cortex, seen in patients with major
depression. These effects are likely mediated by a combination
of excess glutamate activity induced by glucocorticoid activity
and a loss of gamma-aminobutyric acid (GABA)ergic and sero-
tonergic activity, which has been shown to protect against
hippocampal neuronal loss in experimental subjects.
Aldosterone
Aldosterone is classified as a mineralocorticoid, and though it
is produced in the adrenal cortex, its regulation is not through
the HPA axis, but rather through angiotensin II and serum
potassium levels. The purpose of aldosterone is to maintain
fluid and electrolyte balances within the body. Aldosterone is
produced in the zona glomerulosa of the adrenal cortex. Syn-
thesis begins with the conversion of cholesterol to pregneno-
lone. Pregnenolone is converted to progesterone, which is then
converted to II-deoxycorticosterone. II-Deoxycorticosterone is
converted into corticosterone and then to aldosterone.
Aldosterone release is regulated by the renin-angiotensin
system of the kidney. This system is activated in response to
activation of the SNS. This process begins when NE is released
from the SNS and acts on the juxtaglomerular (JG) cells of
34 Adrenal Glands
the vascular arteriole of the renal glomerulus. This causes the
release of renin, which is then converted into angiotensin II.
Angiotensin II travels to the zona glomerulosa of the adrenal
cortex and stimulates the release of aldosterone. When aldoste-
rone is released, it causes an increase in sodium levels within the
blood, causing an increase in blood pressure. This increase
in blood pressure signals the JG cells of the kidney to decrease
production of renin, which in turn reduces aldosterone secretion.
The mineralocorticoid receptors (MRs) are not well under-
stood in terms of how they are activated or even their makeup.
Due to the high affinity nature of glucocorticoids to these
receptors, it is unclear how aldosterone binds to these receptors
in competition with other hormones. In addition, the location
of these receptors has been somewhat paradoxical. The studies
that have been undertaken to understand MRs have found a
greater number of these receptors in the hippocampus as
opposed to the epithelial tissues that mineralocorticoid hor-
mones would target. That is not to say that these receptors do
not exist in the epithelial tissue; however, their number pales in
comparison to those that have been found in the hippocam-
pus. Additional research is needed to understand how aldoste-
rone binds to MRs, given the lower circulating levels and lower
affinity to these receptors compared to glucocorticoids.
Androgens
The innermost layer of the adrenal cortex releases androgens -
steroid hormones that impact the development of male sex
organs and secondary sexual characteristics. The primary an-
drogen produced by the adrenal cortex is dehydroepiandros-
terone (DHEA) (and the sulfate metabolite, DHEA-S), which is
a precursor to sex hormones such as estradiol and testosterone.
Isolating the role of DHEA has been difficult due to the signifi-
cant impact of these metabolites and the fact that no specific
nuclear receptor has been identified for DHEA. Originally,
DHEA was thought to be biologically inactive, serving only as
an alternative source of sex hormones, but research in the last
few decades has revealed important systemic and behavior
effects that result from the release of this hormone.
Early research indicated a possible role for DHEA as an
antiglucocorticoid hormone when it was shown that the rela-
tive balance of DHEA to cortisol is modified by stress (in favor
of cortisol) and that coadministration of DHEA and dexameth-
asone produces a decrease in the effect of glucocorticoids on
cytokine release (IL-4 and IL-6) and lymphocyte suppression.
Further, the administration of DHEA impairs fear conditioning
in rats through activation of sigma- 1 chaperone protein recep-
tors. The role of DHEA in glucocorticoid antagonism appears
to be a vital part of the body's antistress system, despite the
absence of clear nuclear receptor activities. Though its role in
sex hormone production is also essential for normal function-
ality, DHEA is by no means an inactive precursor.
The synthesis of DHEA in the brain appears to occur inde-
pendently of the adrenal gland, but its activities may also
impact global functioning and thus bear mentioning. In
humans, levels of DHEA in the bloodstream peak in the sec-
ond decade of life, declining thereafter throughout an indivi-
dual's lifetime. Evidence of DHEA-S's neuroprotective activities
in the hippocampus suggests that loss of this hormone through
age or chronic stress may be related to increased risk of neuro-
nal loss. In a small study of middle-aged and older adult
patients with depression and/or low blood DHEA levels, the
administration of exogenous DHEA led to temporary improve-
ments in depression scores and enhanced some cognitive abil-
ities during the treatment period. Unfortunately, a number of
double-blind, placebo-controlled studies employing exoge-
nous DHEA in normal, healthy, older adult patients failed to
detect cognitive improvements, suggesting that not all older
individuals experience behavioral or cognitive declines as a
result of age-associated decreases in DHEA. Further, indivi-
duals born with a mutation in the gene for the enzyme
required for the conversion of pregnenolone to DHEA demon-
strate sexual infantilism, hypertension, and hyperkalemia but
do not suffer from cognitive deficiencies, indicating that expo-
sure to maternal DHEA during early development is sufficient
for normal brain development; however, older adults with this
condition have not been closely studied. In rodents, DHEA was
found to bind to the N-terminus of microtubule-associated
protein 2 (MAP2) cytoskeletal proteins in the brain, which
are involved in the extension of neuronal axons and dendrites.
The involvement of DHEA in neural development and plastic-
ity may be required during early life but becomes less impor-
tant after childhood, when puberty-induced changes lead to
increased conversion of DHEA to sex hormones involved in the
expression of secondary sex characteristics.
In addition to MAP2 binding, DHEA and DHEA-S also act
as noncompetitive antagonists at GABA subtype A receptors
and enhance N-Methyl-D-aspartate (NMDA)-mediated release
of NE in hippocampal neurons through activation of the chap-
erone protein sigma- 1. The enhancement of NMDA activities
may explain some of the improvements in memory function-
ing seen in depressed and elderly patients, but others have
suggested that cognitive changes may be due to reduced anxiety
from DHEA-mediated antiglucocorticoid effects. Indeed, it was
shown that the administration of DHEA in ovarictomized rats
reduced behavioral signs of anxiety and was associated with
enhanced learning in several maze paradigms. There is also
evidence of DHEA and DHEA-S neuroprotective effects follow-
ing stroke or spinal cord injury. Mice and rats treated with
DHEA demonstrated significantly less white matter loss in
ischemic stroke models while mice treated with DHEA in
conjunction with spinal cord injury showed significant
improvements in motor functionality relative to vehicle con-
trols. Moreover, DHEA has been shown to protect against
neurotoxicity induced by glutamate or amyloid (5 protein,
which may explain why significant declines in DHEA levels in
the elderly and patients with depression or posttraumatic stress
disorder (PTSD) are associated with increased cognitive decline
and/or exacerbated symptomatology. Decreases in DHEA
levels normally associated with aging have less of a direct
impact, but may increase vulnerability to neuronal loss in the
presence of cardiovascular disease and/or chronic stress. Inter-
ventions to reduce chronic stress have showed that cognitive -
behavioral techniques prevented HIV-associated declines in
DHEA and led to increases in available DHEA in normal,
healthy adults, suggesting that tipping the balance of DHEA
to cortisol in favor of the former may diminish some of the
negative impacts of stress. More research is needed to fully
understand the mechanisms leading to age-related DHEA
declines, and why most healthy, older adults do not exhibit
neurological impairment as they age.
Adrenal Glands 35
Adrenal Medulla
The adrenal medulla is responsible for the storage and release
of the catecholamines, NE, and E. This article focuses on the
behavioral effects of NE and E. The adrenal medulla is made up
of chromaffin cells that contain granules composed of catecho-
lamines, adenine nucleotides such as ATP, proteins, and lipids.
In humans, there are two types of these cells: E-storing cells and
NE-storing cells. There are subtle differences between these two
types of cells. For example, E-storing cells contain more glyco-
protein compared to NE-storing cells. The ratio of E-storing
cells to NE-storing cells differs among species.
The synthesis of catecholamines in the sympathoadrenal
system occurs in the adrenal medulla. Synthesis begins when
tyrosine is converted to dihydroxyphenylalanine (DOPA) that
is then converted to dopamine. Dopamine is converted to NE
by the enzyme dopamine (1-hydroxylase, and NE is converted
to E by the enzyme phenylethanolamine N-methyltransferase
(PNMT). PNMT is an important enzyme in the conversion of
NE to E, and its synthesis is controlled by cortisol. Although
80% of the catecholamines released from the adrenal medulla
are E (and 20% NE), only about 10% of circulating catechola-
mines are E from the adrenal medulla, indicating that extra-
adrenal sources of catecholamines, especially of NE, contribute
more to circulating levels.
The adrenergic receptors are G protein-coupled receptors and
contain two classes of adrenergic receptors: a and p. Within the a
class, there are two main types of receptors: and a 2 ; and within
the p class, there are three main types of receptors: p 1( p 2 , and p 3 .
The a class of receptors share dual roles, the most notable being
that of causing vasoconstriction on arteries and veins, more
specifically the arteries to the heart. The receptors are mainly
responsible for contraction of smooth muscle and are located
on the smooth muscle in the gastrointestinal system, blood
vessels, and the bronchioles located in the lungs, to name a
few. These receptors also aid in sodium reabsorption by the
kidneys in connection with the renin-angiotensin system. The
a 2 receptors are responsible for preventing release of insulin and
glucagon from the pancreas. The receptors are mostly found
in the heart, where they work to increase heart rate and contrac-
tility. They are also found in the kidneys, where they work to
promote the release of renin to stimulate aldosterone release.
The p 2 receptors are found on smooth muscle and work to de-
crease gastrointestinal motility, increase vasodilatation to the
skeletal and cardiac muscles, and increase bronchodilation.
The p 3 receptors are found in adipose tissues and work to
increase fatty acids in the blood.
Connection to the SNS
The adrenal medulla, the part of the adrenal gland responsible
for catecholamine release, is directly innervated by the SNS.
However, during a physiological stress response, the adrenal
medulla is not the only source of catecholamines. The suprasp-
inal structures responsible for the activation and release
of catecholamines from the adrenal gland begin with activation
of the PVN of the hypothalamus. The PVN has catecholaminer-
gic and noncatecholaminergic projections to the locus coeru-
leus (LC), which then has catecholaminergic projections to the
medulla oblongata. The medulla makes projections to pregan-
glionic fibers of the dorsal and intermediolateral gray area of
the spinal cord. There are some projections to the ventral horn
of the spinal cord from the medulla; however, this is somewhat
minimal. The preganglionic fibers from the medulla have cho-
linergic projections to the postganglionic fibers of the sympa-
thetic chain. These postganglionic fibers directly innervate the
adrenal medulla, causing the synthesis and release of the cate-
cholamines E and NE. Additional sources of NE are released
following activation of the LC and the postganglionic fibers of
the SNS to the target organs.
Catecholamines and behavior
The release of E and NE from the adrenal medulla and sympa-
thetic ganglia rapidly elicits a number of measurable changes
in an individual. Specific effects within a tissue tend to be
associated with the type of adrenergic receptor that is expressed
by the cells. While a receptors are associated with vasoconstric-
tion in the circulatory system (increasing blood pressure), P
receptors are associated with vasodilation of the bronchi of the
lungs and increased heart rate and strength of ventricular ejec-
tion. Overall, sympathoadrenal activity leads to dilation of the
pupils, piloerection, increased sweating, metabolic activity,
and strengthening of muscles, as well as decreased activity in
the gastrointestinal tract. Some cognitive functions are actually
enhanced, including attention and memory formation. The
interaction of the HPA axis, amygdala, and hippocampus
during stressful situations or traumatic events allows for the
formation of fear memories, which can be triggered upon
exposure to a stimulus associated with the event. From an
evolutionary perspective, these are necessary for animals to
learn to avoid dangerous situations in the future, such as the
location of a predator's den. Unfortunately, these memories
can be quite resistant to extinction and under certain condi-
tions may lead to posttraumatic symptomatology. In humans,
for example, the continual immediate recall of these fear or
traumatic memories can lead to the development of PTSD.
Patients with this disorder experience symptoms associated
with recalling these events, such as recurrent or intrusive
thoughts and dreams about the event. Additionally, there
may be environmental or psychological cues that give the
individual a feeling that the event is recurring. Disturbances
in sleep, appetite, and mood are not uncommon for indivi-
duals with PTSD. It is unclear how catecholamines interact
with the HPA axis, hippocampus, and the amygdala, and
what role they may play in the development and maintenance
of PTSD.
Disorders of the Adrenal Cortex
Cushing’s syndrome
The pathological condition of hyperadrenalism is known as
Cushing's syndrome. Patients demonstrate adrenal hyperplasia,
primarily in response to excess ACTH release from the anterior
pituitary gland, as well as hypercortisolism and excess androgen
production. Excess ACTH may be caused by pituitary adenomas,
hypersecretion of CRH from the hypothalamus, or ectopic
tumors that release ACTH (such as an abdominal carcinoma),
while hypercortisolism can result from adrenal adenomas or
chronic glucocorticoid treatment. Regardless of the source
of the syndrome, Cushing's patients generally present with
36 Adrenal Glands
abdominal and upper back (buffalo hump) fat deposits and a
rounded, swollen appearance in the face (moon face) as well as
problems with acne and hirsutism (excess facial hair). The
majority of patients also demonstrate hypertension, elevated
blood sugar, and protein deficiency in tissues other than the
liver due to the metabolic effects of excess cortisol. Protein loss
in the skeletal muscles causes severe weakness, and loss of lym-
phoid proteins severely compromises the immune system and
increases vulnerability to infection. Many patients also develop
collagen weakness and skin abnormalities, as well as
osteoporosis.
Addison’s disease
Hypoadrenalism, also known as Addison's disease, results from
insufficient levels of adrenocortical hormones secondary to
adrenal atrophy. The majority of cases are caused by autoimmu-
nity against the adrenocortical tissue, although tuberculosis
and some cancers may also destroy the adrenal tissue. Patients
present with severely diminished extracellular fluid volume,
hyponatremia, hyperkalemia, and slight acidosis as a result of
aldosterone deficiency. Complete absence of mineralocorticoid
secretion can lead to shock and death within days. The loss of
cortisol inhibits the ability to maintain a normal resting glucose
level between meals and slows metabolic functions throughout
the body. Without cortisol, the body cannot respond normally
to stress, and common infections may be lethal in Addison's
patients. These patients may also present with characteristic
pigmentation of the skin and mucous membranes in response
to enhanced release of ACTH and melanocyte-stimulating hor-
mone in the absence of normal negative feedback provided by
the adrenocortical hormones. These patients can lead relatively
normal lives if supplemental mineralocorticoids and glucocor-
ticoids are provided on a daily basis.
Summary
The adrenal glands play a vital role in homeostatic mechan-
isms that are necessary for survival and recovery when the body
is subjected to internal and external stressors. Although these
organs exist outside the central nervous system, hormones
released from the adrenal cortex and medulla have a significant
impact on neurological processes and overt behavior, while
also acting peripherally to alter metabolic and immunological
processes. The significance of normal adrenal and HPA axis
functionality is readily observed when dysregulation or disor-
ders are present. Chronic, uncontrollable stress can lead to
insulin resistance, blunted negative glucocorticoid feedback
in the limbic system, elevated basal glucocorticoid levels, and
enhanced stress reactivity. Long-term exposure to elevated glu-
cocorticoid levels is also associated with atrophy of the hippo-
campus and progressive cognitive dysfunction with aging.
Pathological adrenal insufficiency is characterized by fatigue,
slowed metabolic function, and abnormal electrolyte levels
and blood pressure, and can lead to death if not treated. On
the other hand, excess adrenal secretion leads to hypertension,
elevated blood glucose, and protein insufficiency, which weak-
ens skeletal muscles and collagen in the skin and increases
vulnerability to osteoporosis. Continued research into the fac-
tors that influence adrenal gland functions and disease states is
essential for the creation of novel therapies that could prevent
dysfunction and the deleterious consequences of chronic stress
and adrenal disorders.
See also: Aging and the Brain; Anxiety and Fear; Anxiety Disorders;
Catecholamines and Behavior; Crisis Management; Depression;
Perceived Control; Posttraumatic Stress Disorder; Separation Anxiety;
Sleep, Biological Rhythms, and Performance; Social Anxiety Disorder;
Stress and Illness; Suicide.
Further Reading
Funder JW (2000) Aldosterone. In: Fink G (ed.) Encyclopedia of Stress, vol. 1 ,
pp. 141-145. London: Academic Press.
Gagliardi L, Ho JT, and Torpy DJ (2010) Corticosteroid-binding globulin: The clinical
significance of altered levels and heritable mutations. Molecular and Cellular
Endocrinology 316: 24-34.
Guyton AC and Hall JE (2006a) Adrenocortical hormones. In: Textbook of Medical
Physiology, 11th edn., pp. 944-960. Philadelphia, PA: Elsevier.
Guyton AC and Hall JE (2006b) The autonomic nervous system and the adrenal
medulla. In: Textbook of Medical Physiology, 11th edn., pp. 748-760. Philadelphia,
PA: Elsevier.
Hadley ME and Levine JE (2007a) Catecholamines and the sympathoadrenal system.
In: Endocrinology, 6th edn., pp. 315-336. Upper Saddle River, NJ: Pearson
Princeton Hall.
Hadley ME and Levine JE (2007b) Adrenal steroid hormones. In: Endocrinology,
6th edn., pp. 337-365. Upper Saddle River, NJ: Pearson Princeton Hall.
Johnson E0, Kamilaris TC, Chrousos GP, and Gold PW (1992) Mechanisms of stress:
A dynamic overview of hormonal and behavioral homeostasis. Neuroscience and
Biobehavioral Reviews 16: 115-130.
Maninger N, Wolkowitz 0M, Reus VI, Epel ES, and Mellon SH (2009) Neurobiological
and neuropsychiatric effects of dehydroeplandrosterone (DHEA) and DHEA sulfate
(DHEAS). Frontiers in Neuroendocrinology 30: 65-91 .
McEwen BS and Mendelson S (1993) Effects of stress on the neurochemistry and
morphology of the brain: Counterregulation versus damage. In: Goldberger L and
Breznitz S (eds.) Handbook ot Stress: Theoretical and Clinical Aspects, 2nd edn.,
pp. 101-126. New York: Free Press.
Nicolaides NC, Galata Z, Kino T, Chrousos GP, and Charmandari E (2010) The human
glucocorticoid receptor: Molecular basis of biologic function. Steroids 75: 1-12.
Parker KL and Rainey WE (2004) The adrenal glands. In: Griffin JE and Ojeda SR (eds.)
Textbook ot Endocrine Physiology, 5th edn., pp. 319-348. New York: Oxford
University Press.
Aggression
B J Bushman, Ohio State University, Columbus, OH, USA; VU University, Amsterdam, The Netherlands
E H O’Brien, University of Michigan, Ann Arbor, Ml, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary Antisocial behavior Any behavior that either
Aggression Any behavior that is intended to harm another damages interpersonal relationships or is culturally
person who does not want to be harmed. undesirable.
Aggression had existed for a long time before people became
interested in studying it. Dinosaurs, birds, insects, Nean-
derthals, indigenous tribes, and almost our entire evolutionary
lineage adopted aggression as a useful survival tool. Killing,
violence, and intimidation proved quite useful in obtaining
food, sex, territory, and other resources, among all the species.
When humans evolved as hunters and gatherers and settled into
smaller living communities, aggression became a particularly
constructive asset. Males could use aggression to win mates,
provide protection, and hunt other animals. Females could
use aggression to defend offspring and guard domestic capital.
In these ways, our most aggressive ancestors were also the most
successful in passing on their genes to future generations, thus
providing us with a highly effective blueprint for survival.
However, aggression became less and less adaptive as we
continued to evolve. Small tribes transformed into complex
societies. Individual effort was unable to compete with group
success. Over millions of years, humans became (and con-
tinue to be) highly social creatures as they learned to bank
on cooperation and interdependence for survival. In turn,
aggression toward others seemed more likely to hinder
group progress than promote it. Men and women were forced
to resort to more accommodating behaviors in order to live
longer, healthier lives, and ultimately increase their chances
of making a splash in the gene pool.
It is both peculiar and astounding, then, that aggression has
come to characterize so much of our subsequent history. The
rapid rise of interpersonal violence within early societies, the
widespread occurrence of murder throughout the Middle Ages,
and the countless casualties from any one of the hundreds of
wars, all demonstrate our consistent proclivity toward aggres-
sion and violence. In fact, we experienced only 26 days of
world peace (defined as the absence of international wars) in
all of the 40 years following the Second World War. Although
numerous sources suggest that overall aggression has been on a
steady generational decline, its occurrence and effects are con-
siderably prevalent even today.
Why has something seemingly so maladaptive and detri-
mental to our survival not been left behind by evolution? How
have we developed into such a socially dependent species
while failing to learn from our most destructive social mis-
takes? What can we do to change our behavior before it
becomes too late? Where can we find the answers?
Such is the pursuit of the psychological study of human
aggression. Though the act itself is certainly not a recent phe-
nomenon, modern scientific research on aggression has
revealed a great deal of insight into its many characteristics,
causes, and means of reduction.
What Is Aggression?
In social psychology, aggression is typically defined as any
behavior that is intended to harm another person who does
not want to be harmed. It is distinct from violence, which is
usually defined as extreme physical harm with injury or death
as its goal. Violence is a specific form of aggression. All violence
is aggressive; not all aggression is violent.
Importantly, this definition of aggression highlights three
critical features. First, aggression must be an actual behavior. It
is not an emotion, thought, or memory, but a real-life observable
behavior, such as a punch or a verbal curse. Second, aggression
must be purposeful and intentional. The person who uses
aggression very much intends to harm the other person; it is
not an accident or by-product. Third, aggression always involves
undesired harm to the other party; the victim must not want to
be harmed. Thus, autonomous behaviors such as masochism
and suicide do not qualify as aggression.
Aggression takes a variety of forms and can vary in function.
While its general definition remains constant, it is important to
distinguish between these individual components.
Forms of Aggression
Aggression is often expressed in distinct forms, typically falling
within each of three sets of criteria. First, aggression can be
physical, verbal, or relational. Physical aggression refers to
using physical behaviors to harm others, and includes beha-
viors such as hitting and shooting. It is the form of aggression
most clearly linked with violence. Verbal aggression refers
to using words to harm others, and includes behaviors such
as screaming and cursing. Relational aggression, also known as
social aggression, refers to harming a person's relationships,
feelings of acceptance, or social inclusion, and includes beha-
viors such as talking about others behind their backs or inten-
tionally excluding a person from one's own group of friends.
The social pain involved in relational aggression can be more
lasting than physical pain.
Second, aggression can be direct, indirect, or displaced.
Direct aggression refers to an aggressive act against a person
who is physically present while it occurs. This would include
aggressive behaviors such as kicking another person or
37
38 Aggression
yelling at a person to his or her face. In contrast, indirect
aggression is expressed when the victim is absent. This would
include aggressive behaviors such as destroying property or
spreading rumors.
Sometimes people displace their aggression against an
innocent substitute target. In the classic Freudian example of
displaced aggression, a man is angered by his boss at work and
kicks his dog when he gets home from work, rather than
retaliating directly against the boss. In addition, the victim
does not always have to be completely innocent. In triggered
displaced aggression, the target commits some kind of minor
offense, which in turn causes a person to aggress. For example,
the man who was berated by his boss might come home to
find that the dog has spilled some of its food on the floor.
Displaced aggression is used primarily when the primary
target is not available (e.g., when the man's boss leaves work
early) or when retaliation against the target may result in
undesirable consequences (e.g., the boss might respond by
firing the man).
Third, aggression can be active or passive. Active aggression
is expressed when the aggressor resorts to explicitly harmful
behavior toward the victim, such as slapping or yelling. On the
contrary, passive aggression refers to failure on the aggressor's
part to act in a helpful way, or withholding some helpful
behavior, such as 'forgetting' to inform someone of important
information or deliberately showing up late for an important
group meeting.
Direct and active forms of aggression can be quite risky,
leading to injury or even death. Thus, most people prefer to use
indirect and passive forms of aggression instead.
Functions of Aggression
Aggressive acts may also differ in their function. Consider two
examples. In the first, a husband finds his wife and her lover
together in bed. He takes his rifle from the closet, and shoots
and kills both individuals. In the second, a 'hitman' uses a rifle
to kill another person for money. The form of aggression is the
same in both examples (i.e., physical aggression caused by
shooting and killing victims with a rifle). However, the motives
appear quite different. In the first example, the husband
appears to be motivated by anger. He is enraged when he
finds his wife making love to another man, so he shoots
them both. In the second example, the 'hitman' appears to be
motivated by money. The 'hitman' probably does not hate his
victim. He might not even know his victim, but he kills the
person anyway for the money. To capture different functions
or motives for aggression, psychologists make a distinction
between reactive aggression (also called hostile, affective, angry,
impulsive, or retaliatory aggression) and proactive aggression
(also called instrumental aggression). Reactive aggression is
'hot,' impulsive, angry behavior that is motivated by a desire
to harm someone. Proactive aggression is 'cold,' premeditated,
calculated behavior that is motivated by some other goal
(obtaining money, restoring one's image, restoring justice).
Although the function of aggressive behavior is important
to consider, recent research suggests that it is almost impossible
to distinguish between reactive and proactive aggression. Real-
life scenarios are typically not as distinct and unambiguous as
the husband and hitman examples.
Measurement of Aggression
Aggression is scientifically measured inside and outside the
laboratory. Field research conducted outside the lab looks at
statistics to establish a correlation between aggression and
predicting variables. Instances of violence and aggression
such as school shootings or violent crimes are analyzed after
they occur, and factors such as gender, poverty, and age are
examined to determine what conditions might have led the
aggressors to act in such ways.
Aggression is also studied behaviorally in the lab. Given the
practical and ethical restraints of using human participants,
researchers have developed a number of creative methods to
address the issue. They cannot give the participants in their
studies guns, knives, or even boxing gloves. Today, there are
three primary ways in which laboratory aggression is measured.
First, the 'hot sauce paradigm' measures aggression by requiring
participants to indicate the amount of hot sauce an ostensible
partner, who very much dislikes spicy food, should consume.
Second, in the 'bungled procedure' paradigm, participants are
asked to shoot a human target with a type of pellet gun, but
before they are able to do so, the equipment experiences a
'malfunction' and they never actually use the gun. Aggression
is measured by the type of gun and number of bullets chosen
by the participant.
The third method is perhaps the most common. Participants
are asked to participate in a competitive reaction time task, in
which they press a button as quickly as possible in a series of trials
with an ostensible partner. The 'loser' of each round is shocked by
the opponent, who sets the degree of stimulation. The newer
model uses blasts of noise through headphones, adapted from
the original paradigm that used physical shocks. Noise levels are
randomized throughout the task, and aggression is measured by
analyzing the responses of the target participant when given the
opportunity to 'blast' his or her partner. In other variations of the
paradigm, participants earn money for participating in the study,
and they can subtract money from their ostensible partner.
Why Do We Use Aggression?
It is indisputable that people use aggression in a wide variety
of ways. In turn, many theories have been developed in an
attempt to explain why we so often aggress against others. The
main theories are summarized in the following sections.
Instinct Theories
Instinctive theories of aggression state that aggressive behavior
is simply an innate survival tendency, inherent in many spe-
cies. According to these theories, first advanced by Charles
Darwin, violence and aggression are unmotivated, automatic
forces that occur naturally and have been passed down as an
adaptive evolutionary trait; we are wired to be aggressive in
order to increase our chances of survival. We do not learn to be
aggressive or become aggressive after certain life experiences.
Rather, as political philosopher Thomas Hobbes stated, "we
are naturally violent creatures."
Sigmund Freud believed that all humans are born with both
a 'life instinct,' which creates, sustains, and promotes life,
and a 'death instinct,' which destroys life. According to this
Aggression 39
psychoanalytic view, we stop short of destroying ourselves
because the life instinct overpowers the death instinct, and
diverts our harmful urges outward toward others. Therefore,
the basis of aggression lies within the internal battle to preserve
our own life.
Building on Freud's ideas, ethologist Konrad Lorenz pro-
posed that aggressive urges can build up, in much the same
way that hydraulic pressure causes fluids to build up within a
restricted space, and that these urges must be released in some
way. Subsequently, we engage in aggressive behavior to release
our natural pent-up levels of aggressive instincts.
Research refuting instinctive theories reveals that aggression
does not seem to be completely unmotivated. There is even
evidence to suggest that humans actually derive pleasure from
harming others who have provoked them.
Frustration-Aggression Theory
Frustration-aggression theory, proposed in 1939 by a group of
researchers from Yale University, was summarized in two bold
statements: (1) "the occurrence of aggressive behavior always
presupposes the existence of frustration" and (2) "the existence
of frustration always leads to some form of aggression. " Whereas
instinct theories focused on internal factors that increase
aggression, the frustration-aggression theory focused on external
factors. However, it readily became apparent that not all aggres-
sion results from frustration, and not all frustrations lead to
aggression. Besides an inclination to aggress, frustrations actu-
ally stimulate a number of different inclinations such as an
inclination to escape or to find a way around the obstacle to
the goal. The revised theory proposed that the inclination that
eventually dominates is the one that is most successful in
reducing frustration. In other words, people learn through
experience what actions are effective in reducing aggression.
This idea opened the door for learning theory explanations of
aggression (see next section).
The frustration-aggression theory provides a good explana-
tion of why poverty is a consistent predictor of aggression and
violence. It is frustrating not to be able to obtain basic needs
such as food and shelter.
Fifty years later, Leonard Berkowitz revised the frustration-
aggression theory by proposing that all unpleasant events - not
only frustration - deserve to be recognized as important causes
of aggression. The idea is that unpleasant events (including
frustrations) automatically produce primitive fight or flight
reactions. When we experience an unpleasant event, we want
to stop it or leave. The occurrence of aggression depends on
how the unpleasant event is interpreted and on the presence
of aggressive cues. For example, if a person has just seen a
violent movie and is pushed from behind while exiting the
theater, he or she may very well act in an aggressive manner.
Learning Theories
The frustration-aggression theory was seminal to the develop-
ment and popularity of other situational explanations of
aggression. These explanations went against the assumption
that people are aggressive because they are born that way.
Learning theories of aggression are based on both operant
and classical conditioning. Operant conditioning, the defining
feature of behaviorism, proposes that behavior is motivated by
pursuit of pleasure and avoidance of pain. Classical condition-
ing, as famously demonstrated by Ivan Pavlov's experiments
involving dogs, proposes that behaviors are determined by the
learned pairing of two or more stimuli.
Other learning theories of aggression go beyond simple
conditioning. In his social learning theory (also called obser-
vational learning theory), Albert Bandura theorized that obser-
vation and imitation are at the root of aggressive behavior. In
Bandura's studies, children who observed an actor hitting a
'bobo' doll (a large, inflatable toy clown) imitated the actor's
aggressive behavior and also hit the 'bobo' doll. Bandura pro-
posed that aggression is not simply a mindless mimicry of
another's behavior, but also involves cognitive inferences, gen-
eralizations, and interpretations. For example, if a child sees
one parent hit the other across the face following an alterca-
tion, the child may not only imitate the action, but may also
conclude that it is acceptable to hit someone who provokes
you. Recent research has revealed that this process of learning-
from-imitation even has a neurological basis. 'Mirror neurons'
identical to those that are fired when one is performing a
behavior are fired also in response to observing someone else
perform the behavior.
Social-Cognitive Theories
As learning theories increased in complexity, they gave rise to
theories that focused on higher cognition and information
processing. These social-cognitive theories focus on how
thoughts influence behaviors. One social-cognitive theory,
developed by Rowell Huesmann and colleagues, focuses on
scripts. The term script is borrowed from theater. In a play
or movie, a script tells the actor what to say and do. The
fundamental element in a script is the vignette, defined as 'an
encoding of an event of short duration,' consisting of both a
perceptual image and a 'conceptual representation' of the event.
A simple vignette might consist, for example, of an image of one
person hitting another (image) in anger over something the
other person has done (a conceptual representation). A script
consists of a sequence of vignettes. Such scripts define situa-
tions and guide behavior: the person first selects a script to
represent the situation and then assumes a role in the script.
Once a script has been learned, it may be retrieved at some later
time as a guide for behavior.
A second social-cognitive theory, developed by Kenneth
Dodge and colleagues, focuses on attributions, which are the
explanations people make about why others behave the way
they do. For example, if a person trips you, a hostile attribution
would be that the person did it on purpose to hurt you.
Aggressive people tend to make hostile attributions.
A third social-cognitive theory, developed by Craig Anderson
and colleagues, is called the General Aggression Model. According
to this model, aggression is the result of situational and per-
sonality factors that influence internal states such as thoughts,
feelings, and physiological arousal (e.g., heart rate).
What Causes Aggression?
While much is still unknown, research has shown that there are
three major causes of aggression: nature, nurture, and the
40 Aggression
environment. The factors are not mutually exclusive or even
independent; often they act together.
Nature
Gender differences
Gender differences in aggression are apparent from a very early
age, and appear to remain relatively consistent throughout life.
By preschool, boys display higher levels of physical aggression
than girls. This divide becomes more pronounced through child-
hood and adolescence, and is clearly demonstrated by the drastic
gender difference in murder rates around the world. Research
further shows that males are more likely than females to express
aggression in order to ensure the sexual fidelity of their partners.
For example, men often report higher levels of anxiety and
dissatisfaction toward sexual infidelity than toward emotional
infidelity (i.e., being in love with another person), whereas
females show the opposite pattern. Females show higher levels
of relational and indirect aggression than do males.
These gender differences highlight our evolutionary ties
with aggression. A man can most effectively pass on his genes
by making sure that a partner is carrying his child, and so he
is able to use strength and power to suppress his mate's
partners. A woman is most likely to pass on her genes by
making sure that the man she mated with remains com-
mitted to her child, and so she uses indirect aggression to
intimidate rival mates.
Age differences
Across cultures, aggressive behaviors appear very early in child-
hood. Infants display angry facial expressions by 4 months,
and interpersonal aggressive behaviors develop quickly there-
after. The existence of aggressive behavior at such a young age
suggests that there must be at least some inborn aggressive
tendencies.
Research shows that toddlers 1-3 years old are the most
aggressive group of humans on the planet. In daycare settings,
about 25% of interactions among toddlers involve some kind
of physical aggression (e.g., one child pushing another child
down). No other group, not even violent youth gangs or hard-
ened criminals, resorts to physical aggression 25% of the time.
Fortunately, children cannot do much damage at that age!
After age 3, most people become less aggressive over time.
However, a subset of people become more aggressive over
time. The most dangerous years for this subset of individuals
(and for society) are between ages 15 and 30, when violent
crime rates are the highest.
Although these generalizations summarize the data accu-
rately, exact developmental trends in general aggression are
difficult to measure because aggressiveness manifests itself in
different ways at different ages - for example, in taking things
at age 4, fighting at age 8, telling lies about others at age 12,
vandalism at age 16, and murder at age 27.
Personality traits
We all know that some people lash out aggressively at others
at the slightest provocation, while other people hardly
ever lash out. There is a 'dark triad' of personality that is
linked to aggression, violence, and other antisocial behaviors:
(1) psychopathy, (2) narcissism, and (3) Machiavellianism.
Psychopathy is a personality disorder marked by callous
and unemotional affect and low empathy for others. Psycho-
paths often engage in antisocial behavior to gratify their
own desires, and show little for their actions. They focus on
immediate rewards rather than long-term consequences, and
have difficulty learning from their past mistakes. The term
narcissism comes from the mythical Greek character who fell
in love with his own image reflected in water. Narcissists have
a grandiose, inflated view of themselves, and a sense of enti-
tlement. Narcissists lash out aggressively at others who criti-
cize them. The term Machiavellianism comes from the Italian
Renaissance diplomat Niccolo Machiavelli, who wrote The
Prince, a guide for the prince to maintain his power and
authority over subordinates by any means, including cunning-
ness and force.
Biological factors
There are a variety of biological factors that appear to influence
aggression. While most of them are insufficient to cause aggres-
sion and require some sort of situational trigger, they nonethe-
less increase the likelihood of aggression.
People with lower levels of arousal, such as blood pressure
and heart rate, tend to exhibit more aggression than those with
average (or higher) arousal ratings, perhaps because they seek
to increase their arousal levels by engaging in antisocial behav-
ior. In one experiment, men who responded very little physio-
logically to violent movie clips acted more aggressively than
those who had a higher response.
Research has also established a strong link between low
levels of the neurotransmitter serotonin and aggressive behav-
ior. Serotonin is referred to as the 'feel good' chemical in our
brains, and so when we lack a sufficient amount of it, we
may be more inclined to behave aggressively. This relationship
has been demonstrated causally using both animal and human
participants. Moreover, the male sex hormone testosterone has
been linked with increased aggression, but less consistently.
Higher levels of plasma testosterone probably increase aggres-
sion slightly, but the outcome of winning and dominating
affects testosterone levels just as much.
Overall executive functioning also has a significant impact
on the likelihood of aggression. This type of higher-order
cognitive functioning occurs toward the front of the brain
just behind the forehead, called the prefrontal cortex. Damage
to the prefrontal cortex has been shown to directly increase
aggression levels. Prefrontal cortex damage can also create
several problems that increase the likelihood of aggression,
such as low IQ and the development of attention deficit hyper-
activity disorder (ADHD).
Genetics
Although there is yet to be a conclusive demonstration of the
hereditability of aggressive inclinations in humans, research
suggests that certain people have a genetic predisposition
to behave aggressively. In one study, people with a particular
variation of the gene that determines monoamine oxidase
activity (MAO A) were at a significantly higher risk of becoming
aggressive adults, but only if they had been abused in childhood.
Abused children with lower MAOA became more aggressive
adults than both nonabused children with MAOA depletion
and abused children with normal MAOA. Another important
Aggression 41
study showed that individuals who had a gene variation that
lowered serotonin and dopamine activity were at a higher risk
for antisocial behavior in adolescence.
Nurture
There is substantial support for the argument that aggression
is, at least partly, determined by situational factors. For instance,
although the sex differences discussed earlier may suggest that
aggressive tendencies are innate, sex differences in aggression
disappear under high levels of provocation. In fact, statistics
show that female partners are actually more likely than their
male counterparts to exhibit domestic physical aggression in
heterosexual relationships (although males are likely to cause
more physical injury or even death). Clearly, 'nature' cannot
fully explain aggression.
Physical situation
Subtle unpleasant changes in our immediate surroundings
often influence our levels of aggression. One salient correlate
of aggression is heat. Research has consistently demonstrated a
link between hot temperatures and aggression. Other nonso-
cial, unpleasant sensations such as noxious odors (e.g., smoke,
pollution), and loud (particularly uncontrollable) noises have
also been linked to aggression. Certain visual stimulation,
including aggressive cues (e.g., weapons), can also trigger
aggressive behavior.
It is also important to consider how the immediate situa-
tion influences aggression. For example, rates of aggression
may increase among certain populations such as prisoners,
not because those individuals are naturally more aggressive,
but perhaps because the enclosed prison environment makes
them more prone to aggressive behavior.
Social situation
Our social environment as well as our physical environment
can contain unpleasant situational cues. Crowding, defined as
the perception that there are too many people in the vicinity,
can increase aggression.
When we are rejected by others in a social situation, we are
also more prone to use aggression. For example, an analysis of
school shooters found that 13 out of the 15 shooters had been
rejected by a romantic partner prior to the shooting spree. The
greater the degree of social rejection, such as in extreme cases
like ostracism (being directly isolated or rejected by others),
the more prone a person is to using aggression.
Aggressive arousal
Many factors that increase aggression (e.g., heat, violence in
media) also increase physiological arousal. Thus, it is not
surprising that arousal is linked to aggression. There are at
least four reasons why arousal might increase aggression.
First, high arousal might be interpreted as an aversive stimulus,
which would trigger aggression in the same ways that other
aversive stimuli have been shown to do. Second, arousal nar-
rows attention. If aggressive cues are salient in the situation,
then people will focus most of their attention on the aggressive
cues. Third, arousal increases the dominant response, which is
defined as the most common response in that situation. Thus,
whatever people are normally inclined to do (including
behaving aggressively), they will be even more strongly
inclined to do when they are physiologically aroused. Fourth,
arousal from any source (e.g., exercise, coffee, an exciting
movie) may be misinterpreted as feelings of hostility or aggres-
sion if people are later provoked, a process called excitation
transfer. For example, a woman is accidentally bumped while
walking to her car after a long workout at the gym. Her
increased heart rate and perspiration from exercise might be
mislabeled as anger at being bumped, if she yells at the person
who bumped her.
Disinhibiting factors
All people have aggressive impulses, but most are able to
inhibit these impulses. However, some situational factors can
reduce aggressive inhibitions. One such factor is alcohol. To
use an analogy, alcohol increases aggression by paralyzing the
brakes, not by stepping on the gas. In murder cases, at least half
of the perpetrators were intoxicated when they murdered the
victims (often the victims were also intoxicated). Laboratory
experiments have shown that alcohol seems to increase aggres-
sion in combination with other factors. Factors that normally
increase aggression (e.g., frustration, provocation) have a
stronger effect on intoxicated people than on sober people. If
someone is insulted, his or her response will be more violent if
he or she is drunk than when he or she is sober. When there is
no provocation, however, the effect of alcohol on aggression
may be negligible.
Similarly, anonymity has been consistently linked to
increased aggression. When people are anonymous, they
seem to lose their inhibitions against behaving aggressively.
Perhaps this is why violent crimes are much higher at night
than during the day time, and why bank robbers and members
of the Ku Klux I<lan wear masks when they commit violent
crimes. Laboratory experiments show that aggressive behavior
increases when people cannot be identified by name or
appearance.
Environment
Apart from personality and situational influences, a person's
overall environment can have quite a significant impact on
aggression.
Family
Children from neglected or broken family backgrounds are
more likely to become aggressive adults than children from
stable families. Moreover, children are more likely to become
aggressive adults if they witness a significant amount of domes-
tic violence while growing up. Those who are physically abused
at a young age are more likely to become physically abusive
parents and romantic partners themselves.
Peers
As children grow, peer groups wield a greater influence on
their behavior. Children who are rejected by their peers are
more likely to behave aggressively later in life, and vice versa
(aggressive children are likely to be rejected by their peers).
Having antisocial peers is also a risk factor for aggression and
violence.
42 Aggression
Community
Beyond family and peers, a person's collective culture also
influences aggression. Societal norms and expectations differ
widely across cultures, and often impact on aggressive behavior.
Violence and aggression are more acceptable in some cultures
(e.g., cultures that place a large emphasis on honor) than in
others. There are great regional differences in aggression in the
United States, with violence levels being much higher in
the South than in the North. Psychologists Richard Nisbett
and Dov Cohen theorize that this discrepancy is a result of
a southern culture of honor. Southern settlers typically came
from herding families that relied on livestock to make a living,
whereas Northern settlers relied on agriculture. Aggression is
theoretically more practical and useful in protecting animals
rather than crops, and so the idea of using aggression to
preserve the honor and success of their family business
became ingrained in individuals from the South. The overall
condition of a particular community also affects aggression.
Impoverished neighborhoods are often more conducive to
crime and violent behavior. Similarly, communities that are
characterized by high levels of racism, prejudice, or intoler-
ance provide a breeding ground for aggression.
Media
Violence in the mass media can increase aggression and fear
of victimization, and make people numb to the pain and
suffering of others. Experimental studies have shown that vio-
lence in the mass media causes aggression, and longitudinal
studies have shown that the effects can persist for decades.
In fact, in 1972, the US Surgeon General issued a warning
about the harmful effects of violence in TV programs and
films. Violent video games, which were developed after the
Surgeon General had issued his warning, may be even more
harmful than violence in TV programs or films, for at least
three reasons. First, video game play is active, whereas watch-
ing TV is passive. People learn better when they are actively
involved. Suppose you wanted to teach a person how to fly an
airplane. What would be the best method to use: read a book,
watch a TV program, or practice on a video game flight
simulator?
Second, players of violent video games are more likely to
identify with the violent characters therein. If the game is a
first-person shooter, players have the same visual perspective as
the killer. If the game is third person, the player controls the
actions of the violent character from a more distant visual
perspective. In either case, the player is linked to a violent
character. In a TV program, viewers might or might not identify
with a violent character. Third, violent games directly reward
violent behavior by awarding points or allowing players to
advance to the next game level. In some games, players are
also rewarded through verbal praise, such as the words 'Nice
shot!' or 'Impressive!' after killing an enemy. It is well known
that rewarding behavior increases its frequency. In TV pro-
grams, reward is not directly tied to the viewer's behavior.
A recent Dutch study by Hanneke Polman and her colleagues
provided the first evidence that playing violent games produces
stronger effects than simply observing violence. In this study,
some participants played violent games, whereas other partici-
pants watched the games being played, and the effects of
aggression were stronger on the players than the watchers.
What Reduces Aggression?
Although aggressive behavior may never be eliminated from
society, various treatments have been developed to reduce
aggression. In all cases, any method of reduction is most effective
when applied in early childhood before aggressive behavior
becomes crystallized. In general, two primary approaches have
been used to reduce aggression: behavior modification and
cognitive-behavioral therapy.
Behavior Modification
The main goal of behavior modification is to reduce aggression
by reinforcing the learning of nonaggressive behaviors. This
method is most effective when applied to instances of proac-
tive aggression. People often aggress in subtle, indirect ways,
such as talking behind the back of a coworker or inviting all
friends but one to a party. Proactive aggression involves a
certain degree of planning, patience, and cunning in order to
achieve one's goals, for example, trying to malign the reputa-
tion of a detested peer. In order to reduce this type of aggres-
sion, behavior modification principles use a technique called
the differential reinforcement of alternative behavior. Rather
than simply rewarding someone for not using aggression, this
approach seeks to actively reinforce prosocial behaviors. In this
way, aggression is decreased because prosocial behaviors are
increased. For example, an aggressive child could be given
a candy bar every time he helps his sister. Furthermore, it is
especially helpful to reinforce positive social skills such as
engaging in 'small talk' so that the child learns effective ways
to achieve a goal without using aggression (e.g., accommodat-
ing, negotiating). The overall effectiveness of these methods is
increased by the presence of prosocial role models, such as
parents or celebrities, who reinforce and model prosocial,
nonaggressive behaviors.
Cognitive-Behavioral Therapy
Cognitive-behavioral therapy is commonly applied to treat
reactive, hostile aggression. Given that these cases are charac-
terized by feelings of anger and rage, cognitive-behavioral
techniques attempt to reduce emotional responses to provoca-
tion. Techniques typically involve meditation, relaxation, and
concentration on peaceful imagery. In addition, people are
trained to mentally prepare for future provocation by learning
to interpret provocative events in a more benign manner.
For example, in self-instructional training, people memorize
nonaggressive thoughts like 'I know how to control my anger,'
'I will not lose control,' and 'Just continue to relax.' Individuals
are trained to manage their aggressive thoughts by imagining
a potential conflict and visualizing a calm, peaceful solution.
When behavioral techniques like meditation and relaxation
are combined with the rehearsal of nonaggressive cognition,
reactive aggression is most effectively reduced.
Conclusion
Aggression is a complex and multifaceted behavior, and has
been used by living species for millions of years to attain
Aggression 43
desirable goals and resources. Aggression researchers have
attempted to delineate the origins, causes, and defining char-
acteristics of aggression. Over time, aggression and violence
levels have declined. We hope that aggression research will
help this trend to continue.
See also: Anger; Group Dynamics; Media Influence on Behavior;
Personality Development; Social Development (Attachment,
Imprinting).
Further Reading
Anderson CA and Bushman BJ (2002) Human aggression. Annual Review of
Psychology 53: 27-51 .
Bandura A, Ross D, and Ross SA (1961) Transmission of aggression through imitation
of aggressive models. Journal of Abnormal and Social Psychology 63: 575-582.
Berkowitz L (1993) Aggression: Its Causes, Consequences, and Control. New York:
McGraw-Hill.
Bushman BJ and Baumeister RF (1998) Threatened egotism, narcissism, self-esteem,
and direct and displaced aggression: Does self-love or self-hate lead to violence?
Journal of Personality and Social Psychology 75: 219-229.
Bushman BJ and Huesmann LR (2006) Short-term and long-term effects of violent
media on aggression in children and adults. Archives of Pediatrics & Adolescent
Medicine 160: 348-352.
Buss AH (1961) The Psychology of Aggression. New York: Wiley.
Carlson M, Marcus-Newhall A, and Miller N (1990) Effects of situation aggression cues:
A quantitative review. Journal of Personality and Social Psychology 58: 622-633.
Dodge KA (1980) Social cognition and children’s aggressive behavior. Child
Development 51: 620-635.
Dollard J, Doob LW, Miller NE, Mower OH, and Sears RR (1939) Frustration and
Aggression. New Haven, CT: Yale University Press.
Huesmann LR and Eron LD (1984) Cognitive processes and the persistence of
aggressive behavior. Aggressive Behavior 10: 243-251 .
Loeber R and Hay D (1997) Key issues in the development of aggression from
childhood to early adulthood. Annual Review of Psychology 48: 371-410.
Lorenz K (1966) On Aggression, Wilson MK (trans.). New York: Harcourt, Brace.
Miczek KA, Mirsky AF, Carey R, DeBold J, and Raine A (1994) An overview of biological
influences on violent behavior. In: Reiss AJ Jr, Miczek KA, and Roth JA (eds.)
Understanding and Preventing Violence: Biobehavioral Influences, vol. 2, pp. 1-20.
Washington, DC: National Academy Press.
Nisbett RE and Cohen D (1996) Culture of Honor: The Psychology of Violence in the
South. Boulder, CO: Westview Press.
Warburton WA, Williams KD, and Cairns DR (2006) When ostracism leads to
aggression: The moderating effects of control deprivation. Journal of Experimental
Social Psychology 42(2): 213-220.
Relevant Websites
http://www.rcgd.isr.umich.edu/aggr/ - Aggression Research Program.
http://www.colorado.edu/cspv/ - Center for the Study and Prevention of Violence.
http://www.mediafamily.org/index.shtml - National Institute on Media and the Family.
http://www.psychology.iastate.edu/faculty/caa/csv/index.htm - The Center for the Study
of Violence.
http://socialpsychology.org/ - The Social Psychology Network.
Aging and Cognition
S Horning and H P Davis, University of Colorado, Colorado Springs, CO, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Classical conditioning A form of learning that pairs a
neutral stimulus to a conditioned stimulus in order to
invoke a similar behavioral response when the neutral
stimulus is present that is invoked by the conditioned
stimulus.
Cognition A psychological function referring to the act of
processing and manipulating information, including such
processes as awareness, perception, reasoning, and learning.
Declarative memory Memory for facts or information that
can be consciously stated or declared.
Executive function A broad term describing the higher
order cognitive processes which are responsible for the most
complex human functions. Within the brain, the frontal
lobe is the brain region associated with executive functions.
Inhibition The cognitive ability that enables an individual
to refrain from expressing a behavioral response.
Nondeclarative memory Memory for skills or procedures
that does not necessarily require conscious awareness.
Priming The phenomenon that occurs when the
presentation of a stimulus influences the response to a
subsequently presented stimulus, usually occurring outside
of an individual's immediate awareness.
Processing speed A measure of the time required to
respond to and/or process information in one's
environment.
Risky decision-making The decisional cognitive processes
involved in the choice to engage in unsafe or risky behaviors.
Set-shifting A cognitive process that refers to the ability to
shift one's attention back and forth between two tasks or two
lines of thought.
Working memory The ability to temporarily store
and manipulate information in one's mind for
immediate use.
Introduction
Literally, thousands of studies now exist showing that various
aspects of cognition change with age. Age-related declines in
cognitive abilities such as learning, memory, attention, execu-
tive function, decision-making, and problem-solving have
been detected using both cross-sectional and longitudinal
designs. One controversy that has arisen when comparing
results from these two types of designs is when cognition first
starts to decline. Longitudinal studies generally place the
beginning of decline later than cross-sectional studies. Earlier
views tended to place the beginning of cognitive decline in the
sixties or later with only a few studies suggesting that decline
might begin as early as in the fifties. There is now compelling
evidence that many aspects of cognition begin to decline in
middle age and that a significant portion of individuals over
70 years of age living independently can be classified as having
a significant cognitive impairment. In fact, 10-30% of indivi-
duals over 60 years of age have been found to have a diagnos-
able cognitive impairment as a result of various pathologies
associated with aging. Similarly, ~22% of individuals over
71 years of age in the United States have been estimated as
having a significant cognitive impairment without dementia.
The prevalence of cognitive impairment without dementia is
greater in men, older individuals, and individuals with a lower
educational level.
In this article, changes in cognition will be examined from
a lifespan perspective reviewing the available data on cognitive
changes from early adulthood to old age. Until very recently,
few studies examined cognition in middle-aged individuals.
It is now recognized that it is important to know when change
in cognition begins because if we wait until old age, precipitat-
ing or mediating causes may have disappeared or changed.
We will begin with the area of cognition that has received
the most attention, memory.
Declarative Memory
Memory has been divided into the domains of declarative and
nondeclarative memory systems, also frequently referred to as
explicit and implicit memory, respectively. Declarative mem-
ories are facts that can be consciously stated, or declared. For
example, the recollection that I drove to work yesterday would
be a declarative memory. The skills and procedures required
for the drive would be classified as nondeclarative memories.
Many daily tasks, such as riding a bike, engaging in games or
sports, or typing a paper at the computer, fall within the
domain of nondeclarative memory. Performance on these
procedural tasks, in which past experiences serve as an aid,
do not necessarily require conscious awareness. In contrast,
declarative memory occurs with conscious awareness. In the
laboratory, declines in performance on both declarative and
nondeclarative tasks have been observed in older individuals.
Tests of verbal and visuospatial material have been used to
assess declarative memory across the lifespan. In an investiga-
tion of memory performance, children and older adults were
found to exhibit poor recall on multitrial tests of both verbal
and spatial information. For the verbal test, participants were
administered five immediate trials of the Rey Auditory Verbal
Learning Test (RAVLT). In this test, participants hear 15 con-
crete nouns on each trial, but in a different order. After each
presentation, participants are instructed to recall as many
words as they can remember. The number of words recalled
on each trial is the measure of performance. For the spatial test,
participants were administered a version of the game of
44
Aging and Cognition 45
Memory Cards (also referred to as Concentration). Twenty-
four cards are in a 4 x 6 matrix on the computer screen and
participants are instructed to turn over two cards at a time
for matches and attempt to get as few mismatches as possible.
Five immediate trials are administered with the spatial location
of the 12 matching pairs always in the same location. The
percentage of matches out of all attempts is the measure of
performance.
Latent growth curve analyses were used to examine per-
formance across trials on both the RAVLT (N = 2124) and
Memory Cards (N=1579) tests for five age groups (5-12,
13-18, 19-39, 40-59, 60-92). In this analysis, the intercept
(initial start of the learning curve), slope (rate of learning), and
shape (change in rate of learning) were examined. Significant
differences were detected for both tasks on the first trial (inter-
cept), but not on the rate of learning on subsequent trials
(slope). The spatial and verbal learning parameters were not
significantly correlated. Thus, we propose that there is not a
single general learning ability underlying these different task
modalities.
An earlier version of this data set was used to show changes
on the first trial of these two tasks for participants 15-89 years
of age (age groups: 15-19, 20s, 30s, 40s, 50s, 60s, 70s, 80s).
Figure 1 shows age changes for performance on these two tasks
with a sample size that has virtually doubled in size since the
original report. An analysis of performance on the first trial of
the RAVLT detected significantly poorer recall in individuals
50 years of age and older, while the 80-year-old indivi-
duals showed significantly poorer verbal recall than all youn-
ger age groups. A similar significant decline occurred for the
Memory Cards task. That is, individuals 50 years of age and older
demonstrated poorer performance than individuals 15-29 years
of age, and individuals in their 80s showed poorer spatial
memory than all younger age groups. The decline of both
spatial and verbal declarative memory by 50 years of age is
consistent with other reports of significant declarative memory
decline in middle-aged humans for both verbal and spatial
information. Similar declines in midlife have been reported
for declarative memory in rodents and nonhuman primates.
Nondeclarative Memory
Findings from classical conditioning studies of the eye-blink
response and word-stem priming task will be described as both
paradigms have been used to examine nondeclarative memory
changes across the adult lifespan. Briefly, to exemplify the
eye-blink procedure used in research studies, a conditioning
stimulus tone is activated for 500 ms with onset of a mild puff
of air to the eye as the unconditioned stimulus at 400 ms.
A conditioned response is scored if the eye-blink occurs prior
to delivery of the puff of air. That is, a conditional response
occurs on a trial if there is an eye-blink between 29 and 400 ms
after tone onset. A deficit in acquisition of the classical
conditioned eye-blink response has been detectable in the
40-50-year-old subjects in previous research investigations.
Similarly, declines in classical conditioning of the eye-blink
response have been reported in rabbits and cats at ages equiva-
lent to middle age in humans.
Word-stem priming has received the most attention in
studies examining nondeclarative memory in older indivi-
duals. Priming is said to occur when the presentation of a
Figure 1 Standard Z-scores for performance on the first trails of the RAVLT and Memory Card task for participants from 15 to 89 years of age
(RAVLT N = 2266; Memory Cards N= 1518). Adapted from Foster SM, Cornwell RE, Kisley MA, and Davis HP (2008) Cognitive changes across
the life span. In: Qualls SH and Smyer MA (eds.) Changes in Decision Making Capacity in Older Adults: Assessment and Intervention, pp. 25-60.
Hoboken NJ: Wiley.
46 Aging and Cognition
stimulus influences the response to a subsequently presented
stimulus. For example, when a person provides pleasantness
ratings to a list of words that includes the word motel, and is
later asked to complete the word stem mot, the probability that
the subject answers motel is higher than for word stems that
were not primed by the pleasantness rating (e.g., completing
the baseline control stem acc with the nonpresented word
accord). In one study, word-stem priming in each decade of
life from the 20s through the 80s was assessed and individuals
in their 70s and 80s scored significantly lower priming scores.
No decline in priming was detected for younger age groups.
However, priming scores for individuals in their 40s, 50s, and
60s tended to be intermediate between the younger and older
age groups. The failure to detect a decline in priming by
middle-aged individuals might have been due to the small
group sizes. More recently and with larger age groups, impaired
word-stem priming in a middle-age group (mid-30s to early
50s) as compared to a young adult group (late teens to
mid-30s) was reported. Thus, significant nondeclarative mem-
ory deficits, like declarative memory deficits, are detectable in
middle-age individuals with further decline occurring in the
elderly (individuals in their 70s and 80s).
Executive Function
Executive function is a broad umbrella term that describes the
higher-order cognitive processes which are responsible for the
most complex of human functions and behaviors. Although
there is much debate regarding the specific cognitive processes
that should be defined as part of executive functioning, the
literature seems to agree on the fact that executive control is
necessary for goal-directed behavior, such as the formulation
of a goal, devising strategies and plans to achieve the goal, and
the execution and implementation of such plans. The underly-
ing cognitive processes identified in the literature as executive
functions include, but are not limited to, the following:
focused and divided attention, inhibition, self-regulation, ini-
tiation of behaviors, working memory, planning, organization,
problem-solving, information processing, and cognitive set-
shifting.
Within the brain, the frontal lobes, specifically the pre-
frontal cortex, is the brain region associated with executive
functioning. Executive functions are usually quite vulnerable
to cognitive decline, as the frontal lobes are one of the initial
brain regions to experience atrophy, or neuronal cell death, in
response to natural age-related and pathological processes,
including Alzheimer's disease and frontotemporal dementia.
As a result, executive dysfunction can manifest itself in a myriad
of cognitive and behavioral problems, including impulsivity,
poor judgment and decision-making, perseveration, and an
inability to switch back and forth between two lines of thought.
Consequently, impairments in executive functioning can det-
rimentally affect older adults and their ability to live indepen-
dently. Due to the multiple processes that underlie executive
functioning, and the lack of agreement among researchers
regarding the processes that should be defined as part of execu-
tive control, this section will focus on understanding age-related
declines in executive functions for the following areas: cognitive
set-shifting, inhibition, working memory, and problem-solving.
Cognitive set-shifting, also referred to as task-switching
or mental flexibility, is an individual's ability to shift their
attention between two tasks or lines of thoughts. Set-shifting
is often assessed by two popular neuropsychological measures:
the Trail Making Test (TMT) and the Wisconsin Card Sorting
Test (WCST). The TMT has two parts: Part A and Part B. Part A
requires an individual to connect 25 numbers in sequence as
quickly as possible. Part B requires an individual to shift back
and forth between connecting number and letters in alphabet-
ical and numerical order as quickly as possible. Part B of the
TMT, therefore, can detect problems in an individual's ability
to shift set. In a study with a sample size of 680 participants
ranging in ages from 18 to 89, age-related changes in perfor-
mance on the TMT were found. As the participants' age
increased, their performance on Part A and Part B declined.
Furthermore, the age-related effects were more prominent for
Trial B than A; the trial that required shifting. The results of
this study suggest greater difficulty for individuals in their
ability to set-shift with increasing age.
The WCST, another prominent test of executive function-
ing, assesses individuals' ability to set-shift, as well as their
inhibitory control and planning abilities. The test requires an
individual to sort a target card by placing it under one of four
stimulus cards, which differ by shape, number, and color.
An individual must determine the rule for sorting the cards
(i.e., sorting them by shape, number, or color), and is given
feedback regarding the correctness of each of their choices.
After ten trials, the rule is changed, and the individuals must
alter their decisions accordingly. This test will be used to exem-
plify changes in both set-shifting and inhibitory control as it
requires an individual to shift cognitive sets and inhibit old
responses when adjusting to the new rule.
In an investigation to study the age-related changes on
WCST performance, 48 younger adults (18-38years) and 49
older adults (60-86 years) completed a computerized version
of the WCST. The level of education and health status of
participants were statistically controlled to avoid potential con-
founding variables. Overall, the older adult participants per-
formed more poorly on the task as compared to the younger
adults. Additionally, the older adults were more likely to make
perseverative errors than the younger adults. Perseveration
refers to the inability to cognitively shift focus away and inhibit
a prior way of responding. In the WCST, perseverative errors
refer to the mistake of sorting cords by an old rule instead of
the new ride. The results of this study further exemplify that
individuals show deficits in their ability to set-shift, as well as
in inhibition, as they age.
Inhibitory control is another cognitive process part of exec-
utive functioning that has been demonstrated to decline with
age. Inhibition is often measured by the well-known Stroop
task. The Stroop task requires participants to look at a card with
four color names printed on it in conflicting colored ink.
Instead of reading the words, individuals are asked to quickly
name the color of the ink each word is printed in and their
response is timed. Therefore, the test requires individuals to
inhibit an overlearned response (i.e., reading the words) to
respond in a novel way (i.e., stating the color of ink).
To investigate age-related changes in inhibition, a research
investigation was conducted using a version of the Stroop task
in a sample of 80 adult males of various ages across the
Aging and Cognition 47
lifespan. Participants were given four trials. In trial 1, partici-
pants simply read a card with four color words on it as quickly
as possible. In trial 2, participants saw four colors and were
asked to name the color of the ink. In trial 3, participants read a
card with four color words on it, but printed in conflicting
colored ink. In trial 4, the inhibition condition, participants
were asked to name the color of the ink the words were printed
in and were told not to read the words. The results demon-
strated that the older adult participants did not differ from the
younger group on the simple reading task, yet were signifi-
cantly slower during the inhibition condition, suggesting a
deficit in inhibitory control. Other investigators have reported
similar findings on the age-related changes on the Stroop test,
but argue that this deficit is related to a general slowing of
cognitive processing. However, differences among ages on the
inhibition condition, but not the reading conditions, reveal
that performance is affected by another process other than
slowed reaction time. Regardless of the underlying cause, sug-
gested as due to slowed information processing or natural age-
related changes in the brain, the cognitive and behavioral
manifestations of age-related changes produce deficits in the
ability to shift cognitive sets and apply inhibitory control.
Furthermore, as in the realm of memory, deficits in executive
function are first detectable in middle age.
The Tower of Hanoi (TOH) and Tower of London (TOL)
are two popular problem-solving tasks that have been used in
hundreds of studies with young adult college students, various
patient populations, and age groups varying from childhood to
old age. The TOH has attracted the interest of cognitive psy-
chologists because of its application to the development of
solution strategies in problem-solving. It is a transformation
puzzle that requires a repetitive strategy for the achievement
of an optimal solution. The task consists of three pegs with a
number of rings varying in size. Typically, at the start, the rings
are stacked on the left peg with the largest disk on the bottom
and the smallest on the top. The research subject is instructed
to achieve the same disk arrangement on the right peg by
moving one disk at a time and never placing a larger disk on
a smaller disk. The minimal moves for an optimal solution are
2"— 1 where n represents the number of disks.
To explore problem-solving, individuals between the ages
of 15-89 were tested on both a 3- and 4-ring puzzle (N = 755).
Figure 2 shows age changes for performance on the 3- and
4-ring TOH. On the 3-ring TOH puzzle, a significant increase
in the number of moves to solve the puzzle was detectable in
the 60s. On the 4-ring TOH, a significant decline in performance
of individuals in their fifties was found. Thus, like memory,
there is a decline in problem-solving by at least the 50s.
The TOL was developed to assess planning in individuals
who sustained damage to the frontal lobes. The TOL is similar
to the TOH in that one must move a set of objects (balls of
differing colors) from a start arrangement on three or more
pegs to a goal position in as few moves as possible with the
constraints that balls cannot be stacked above the peg and only
one ball at a time can be moved. Figure 2 shows age changes
for performance from 15 to 89 years of age on the TOL.
A significant impairment was detected in individuals in their
60s and older. In addition, TOL performance in 300-plus
individuals between 21 and 79 years of age was examined
on the CANTAB test battery. Within this study, individuals in
the oldest age group (74-79 years) required more moves to
successfully complete TOL problems.
Working memory refers to a cognitive process that involves
the active manipulation of information while maintaining it in
consciousness. For example, multiplying 16 x 18 in your head
requires the active manipulation and holding of information
in awareness. To assess this ability, performance on both a
verbal and spatial n-back task for ~400 individuals was inves-
tigated. In the n-back task, participants are shown a series of
stimuli and asked to indicate for each stimulus if it is identical
to the stimulus n-back in the series (e.g., Is the present stimulus
identical to the stimulus presented two presentations back?).
Changes across the lifespan on a 3-back visuospatial and verbal
task are present in Figure 3. Individuals in their 60s and older
demonstrated significantly poorer working memory perfor-
mance than younger individuals. Similar results have been
reported on several working memory tasks with significant
deficits detectable in the 40s.
Decision-Making
As we age, we are faced with making decisions that can dramat-
ically impact our quality of life. For example, 'which Medicare
supplement will provide the most benefit?', or 'what form of
care will be best for my ailing spouse?' are typical challenging
decisions made by older individuals. A laboratory analog of
risky decision-making is the Iowa Gambling Task. For this task,
research participants are presented four decks of cards on a
computer screen and asked to select from any deck. The goal of
the game is to select cards that result in winning money and
avoid losing money. Participants are told some decks are better
than other decks and they can avoid losing if they stay away
from the worst decks. The two bad decks have high rewards,
but even higher loss. The two good decks have low rewards, but
even lower loss. Participants play 100 trials. The results of this
test for the last 25 trials of the Iowa Gambling task for 872
individuals between 15 and 89 years of age are displayed in
Figure 3. The individuals in their 80s made significantly poorer
choices than young adults, which demonstrates a decline in
decision-making abilities with age.
Processing Speed
Processing speed is a measure of the time required to respond
to and/or process information in one's environment. With
increasing age, processing speed has been shown to slow dra-
matically. Thus, older individuals require more time to com-
plete a cognitive task when compared to younger adults.
Additionally, a speed of processing theory has been formulated
within the cognitive aging literature that accounts for some of
the age-related changes that occur in cognition. A large number
of studies demonstrate that along with a common factor of age-
related decline, processing speed is at least a partial mediator of
age-related performance deficits in a variety of cognitive tasks.
Thus, the slowing of individuals' ability to process information
inevitably impacts their memory, decision-making, and other
cognitive functions.
This reduced, or slowed, speed of processing has been
found to account for poorer performance by older adults on
48 Aging and Cognition
Tower of Hanoi and London problem-solving
15-19 30s 50s 70s 15-19 30s 50s 70s 15-19 30s 50s 70s
20s 40s 60s 80s 20s 40s 60s 80s 20s 40s 60s 80s
TOH 3-ring TOH 4-ring TOL
Figure 2 Standard Z-scores for performance on the average of five trials for the 3- and 4-ring Tower of Hanoi and moves on 21 trials on the
Tower of London for participants between 20 and 89 years (Tower of Hanoi N = 1293; Tower of London /V = 2036). Adapted from Foster SM,
Cornwell RE, Kisley MA, and Davis HP (2008) Cognitive changes across the life span. In: Qualls SH and Smyer MA (eds.) Changes in Decision
Making Capacity in Older Adults: Assessment and Intervention, pp. 25-60. Hoboken NJ: Wiley.
Executive function/decision-making
Figure 3 Standard Z-scores for number of category shifts in the Wisconsin Card Sorting Test (WCST), the percent correct responses on a verbal
and visuospatial /7-back task, and the number of choices from good decks minus bad decks in the Iowa Gambling Task for participants from 1 5 to
89 years of age (WCST N = 1 874; /7-back verbal and spatial N = 745; Iowa Gambling Task - Risky Decision N = 872). Adapted from Foster SM,
Cornwell RE, Kisley MA, and Davis HP (2008) Cognitive changes across the life span. In: Qualls SH and Smyer MA (eds.) Changes in Decision
Making Capacity in Older Adults: Assessment and Intervention, pp. 25-60. Hoboken NJ: Wiley.
a complex decision-making task. For example, an investigation
compared young and elderly adults on three risky decision-
making tasks (Iowa Gambling Task, Cambridge Gambling
Task, and Balloon Analogue Risk Task). One goal of this
study was to determine if a participant's age directly influenced
their performance, or if the effect was indirect via age-related
changes in processing speed and/or memory. A path analysis
found that speed of processing fully mediated performance on
Aging and Cognition 49
the Cambridge Gambling Task and the Balloon Analogue Risk
Task, and that memory acted as a full mediator of performance
on the Iowa Gambling Task. These findings were the first
demonstration that age effects on decision-making can be
fully mediated by cognitive factors such as processing speed
or memory. This path analysis approach provides a promising
avenue for examining the possibility that age-related deficits on
other tasks will be either fully mediated (age effect disappears)
or partially mediated (age effect significantly reduced) by pro-
cessing speed or other cognitive variables.
The Relationship Among Cognitive Functions
The examination of cognition across the adult lifespan reveals
normal age-related declines in almost all areas, including mem-
ory, executive functioning, decision-making, and processing
speed. Although each of these cognitive functions are discussed
as separate entities, declines in one area of cognition are highly
correlated to declines across other cognitive domains as a factor
of natural aging processes. In other words, the various areas of
cognition are thought to be related to each other. This suggests
that changes in several areas of cognitive functioning may be
accounted for by changes in one area of cognition, such as speed
of processing or fluid intelligence. Although most areas of cog-
nition do decline with age, an individual's ability to retain
general knowledge or information acquired over time remains
intact. The amount of acquired knowledge, known as crystal-
lized intelligence, has been found to not only remain stable over
time, but also may continue to increase with age.
Cognitive Rehabilitation: Behavioral Interventions
and Their Impact on Cognition
Natural aging processes can negatively impact our cognitive
functioning as the literature suggests. As a result, these changes
in cognition manifest themselves in the ability of older adults
to function in everyday life, which can not only influence their
quality of life but also their ability to live independently. Aside
from natural aging, older adults are at a high risk for the
development of dementia, including Alzheimer's disease and
vascular dementia. Despite the high rate of individuals devel-
oping dementia-related cognitive impairment annually, little
is known about the cause of these fatal neurodegenerative
diseases. However, in recent years, much research has been
focused on finding preventative measures to reduce the risk
of cognitive decline among the older adult population.
Certain behaviors, such as solving daily crossword puzzles
or playing strategic board games, have long been suggested as
having a positive effect on cognition and may thwart age-related
cognitive decline. Empirical research has begun to explore the
impact of behavioral interventions on cognitive functioning,
which has been referred to as cognitive enrichment. According
to the cognitive-enrichment hypothesis, engaging in certain
behaviors, such as intellectually stimulating activities, physical
exercise, and socialization, can advantageously affect cognitive
functioning in older adulthood. Despite the mainstream popu-
larity of the 'use it or lose it' mentality, which is encompassed
by the cognitive-enrichment hypothesis, much controversy and
debate surrounds this theory. Therefore, the empirical evidence
for and against this hypothesis will be reviewed.
Training the brain through mental exercise has been com-
pared to training the physical body through physical exercise.
Conceptualizing the brain and mental activity in this manner
has produced multiple lines of research aimed at determining
if mental training can improve mental functioning and de-
crease the rate of cognitive decline. Due to the plasticity of
the human brain, or its ability to alter and change functionally
and physically, researchers speculate that the brain is capable
of learning and changing over one's entire lifespan. Based on
this idea, a research study investigated the effects of a brain
training program on normal age-related memory loss in older
adults. For this study, a large sample of older adults was
recruited and randomly assigned to one of three experimental
conditions: an experimental training group, an age-matched
control group, and a no-contact control group. Participants in
the experimental training condition used a cognitively demand-
ing computer program for 1 h a day over a period of 8- 1 0 weeks,
which required them to engage in intense auditory and language
exercises. At the end of the study period, all participants were
compared on a battery of neuropsychological tests to determine
if the brain training generalized to other broad measures of
memory. The results showed that the participants who received
the daily training exercises had made significant improvements
in their global auditory memory score as compared to partici-
pants in the control groups. After a 3-month follow-up, the
participants in the experimental condition continued to outper-
form the control group participants on a digit span forward test.
Upon completion of this study, the researchers concluded that a
plasticity-based training program can improve memory func-
tioning among seniors.
Other research studies have reported similar findings regard-
ing the benefit of training interventions on cognitive function-
ing. For instance, over 2500 older adult participants were
recruited to determine if a cognitive training procedure would
improve their overall cognition. Participants were assigned to
either an experimental condition in which they received cogni-
tive training in memory, reasoning, or visual search, or they were
assigned to a control group. The results demonstrated that
participants who received training in a particular domain (i.e.,
memory training) also demonstrated improvement in a 2-year
follow-up in the same domain they received training as com-
pared to the control group. Overall, the study demonstrated
that this type of intervention does have benefits to the specific
domain targeted by training, but may not generalize to other
cognitive domains. A similar, large-scale study which investigated
the generalizability of the effects of brain training on a sample
of over 1 1 000 adults found that although the training resulted
in improvement on the specific cognitive tasks, the effects did
not generalize to tasks that were not used during training.
Based on the available literature, cognitive training has
been found to benefit an older adult's cognitive functioning
albeit, in most cases, limited to the cognitive domain receiving
the training. In addition to cognitive training, the effects of
engaging in mentally stimulating activity on cognition in older
adulthood have also been studied. According to the cognitive-
enrichment hypothesis, individuals who have greater rates of
engagement in cognitively demanding activities should have a
reduced risk for cognitive decline compared to those who do
50 Aging and Cognition
not actively engage in these types of activities. In order to
investigate this phenomenon, the methodologies within these
studies rely on the use of self-report measures of cognitive
activity. The researchers conducting these studies also vary in
their interpretation of stimulating or intellectually demanding
cognitive activity, such as reading a book or playing chess.
Regardless, the empirical evidence within the literature strongly
supports the cognitive-enrichment hypothesis; higher rates of
engaging in stimulating cognitive activity among older adults
has been associated with a reduced risk of cognitive decline.
Although there is strong supporting evidence for the
cognitive-enrichment hypothesis, others question the reliabil-
ity and the generalization of these findings. Two major limita-
tions have been reported within the cognitive training studies
that have tested the enrichment hypothesis. First, the effects of
the training are narrow and often do not generalize to other
areas of cognition. For instance, memory training exercises
may not benefit reasoning or decision-making, they benefit
only memory. Second, the majority of cognitive training stud-
ies do not monitor or investigate long-term effects. Therefore,
it is difficult to determine the long-lasting effects of these
interventions or whether or not the improvements continue
at a steady rate over a period of time. Third, the research
methodology investigating the effect of engagement in sti-
mulating mental activity on cognition has been questioned.
Self-report measures, for example, can be unreliable due to
inaccurate estimations of one's own mental activity level.
Also, a desire to maintain a positive self-image, also known as
a desirability bias, can distort an individual's portrayal and
report of their actual level of activity. An additional limitation
within this body of research is the variability among researchers
regarding the conceptualization of mental activity and mental
exercise. Researchers have defined it as anything from attending
lectures to watching television, while making arbitrary infer-
ences to the level of cognitive demand each activity requires.
Other major criticisms regarding the research finding a
relationship between level of mental activity and cognitive
functioning is that it is correlational, making it is impossible
to determine the direction of causality within these studies.
An opposing hypothesis to the conclusions drawn by
the cognitive-enrichment research is that a reduction in the
level of stimulating mental activity in older adults can be an
early indication of a dementing process. In other words, the
disease process is driving the level of mental activity, not that
the mental activity is driving the lack of or presence of
a disease process. Additionally, another rival explanation to
the findings of the cognitive-enrichment studies explains that
older adults who report high levels of mental activity may have
always engaged in intellectually demanding activity and may,
therefore, have had a higher socioeconomic status than their
counterpart. Thus, having the advantage of a higher economic
status may have afforded them a higher quality health care,
better nutrition, better educational opportunities, or other
unknown variables that could have influenced their cognitive
functioning in old age.
The influence of educational attainment and other early
environmental factors on late life cognition has been referred
to as the cognitive reserve hypothesis. Models of cognitive
reserve have been discussed as both passive and active pro-
cesses; active processes refer to the brain's attempt to repair or
cope with damage, whereas passive models refer to the brain's
capacity to withstand damage or atrophy. A common passive
model of cognitive reserve is the threshold model which indi-
cates that all individuals experience cognitive decline as they
age; however, individuals who have higher levels of cognitive
functioning in their young adulthood will decline, but their
cognitive performance may never drop to clinically significant
degrees of impairment. For example, in a longitudinal study
of cognitive performance over time, individuals of varying
degrees of education and knowledge demonstrated decline in
cognitive performance at similar rates. Therefore, this finding is
more conducive to understanding cognitive reserve through a
threshold paradigm. In other words, individuals with greater
educational attainment and higher cognitive abilities in their
youth will experience similar rates of cognitive decline as those
who have less education and cognitive abilities. However, due
to the fact that their premorbid level of functioning was higher,
the time it takes for these individuals to reach functional levels
of impairment may be longer than the average individual.
Thus, they are typically able to function independently without
problems despite some degree of cognitive decline.
Besides the literature on cognitive fitness, the cognitive-
enrichment hypothesis also extends to the use of other behav-
ioral interventions that may reduce the risk of cognitive decline,
including exercise. A substantial body of research has been
dedicated toward exploring the effects of physical exercise on
cognitive functioning. One particular study aimed to determine
the effects of an aerobic fitness program on older adults' cogni-
tion. A total of 124 older adults were placed into two groups:
an aerobic exercise group and a nonaerobic exercise group.
Before completing the exercise training, all participants were
tested on a variety of cognitive measures to obtain baseline
measures. The participants in the aerobic group walked for a
period of 1 h three times per week over 6 months, whereas the
nonaerobic group simply engaged in stretching and toning
exercises over the same period. At the end of the testing period,
the participants in the aerobic fitness group had demonstrated
specific improvements on several cognitive measures compared
to the participants in the nonaerobic group.
Likewise, a meta-analysis reviewing the relevant research
that investigated the implementation of an aerobic exercise
program on older adults' cognition was conducted. After ana-
lyzing 18 relevant research articles, the authors of the meta-
analysis reported a large effect size of 0.48. The data was
consistent with the view that an aerobic exercise training pro-
gram can have an advantageous effect on an older adult's
cognitive functioning. Overall, the research in this area sug-
gests that engagement in aerobic exercise may thwart cognitive
decline or reduce the risk of dementia among older adults.
The research testing the cognitive-enrichment hypothesis
suggests that certain behaviors and interventions, specifically
cognitive training fitness programs, mental activity, and aero-
bic exercise, can help an older adult's cognitive functioning.
In addition to the behaviors examined in this brief review, the
cognitive-enrichment hypothesis also extends to and explores
other behaviors and their relation to cognitive functioning
among older adults, including socialization and mental health
practices. Despite the findings supporting cognitive enrich-
ment, much debate exists on the validity and the interpretation
of this body of research. Although this area of research remains
Aging and Cognition 51
new, the preliminary findings are encouraging regarding the
possibility of delaying and reducing normal age-related cogni-
tive decline in older adulthood. Interpretations of the cognitive-
enrichment hypothesis, however, should be done with caution
due to the fact that these behaviors may only delay normal
age-related decline and may not influence the outcome of
a pathological dementing process. However, future studies
should continue to explore and examine the cognitive-
enrichment hypothesis in an effort to validate and understand
its implications.
See also: Aging and the Brain; Clinical Psychology: An Information
Processing Approach; Cognition and Personality; Decision Making
(Individuals); Memory; Personality Development; Problem Solving;
Social Cognition.
Further Reading
Anderson PJ (2008) Towards the developmental model of executive function. In:
Anderson V, Jacobs R, and Anderson PJ (eds.) Executive Functions and the Frontal
Lobes: A Lifespan Perspective, pp. 3-21. New York, NY: Taylor Francis Group.
Bal SK (2004) Leisure activity and risk of dementia: Scrabble, anyone? Canadian Family
Physician 50(1): 51-53.
Ball K, Berch DB, Helmer KF, et al. (2002) Effects of cognitive training interventions with
older adults: A randomized controlled trial. Journal of the American Medical
Association 288: 2271-2281 .
Bechara A, Damasio AR, Damasio H, and Anderson SW (1994) Insensitivity to future
consequences following damage to human prefrontal cortex. Cognition 50: 7-15.
Berg WK and Byrd DL (2002) The Tower of London spatial problem-solving task:
Enhancing clinical and research implementation. Journal of Clinical and
Experimental Neuropsychology 24(5): 586-604.
Bialystok E and Craik FIM (eds.) (2006) Lifespan Cognition: Mechanisms of Change.
New York, NY: Oxford University Press, Inc.
Bizon JL, LaSarge CL, Montgomery KS, McDermott AN, Setlow B, and Griffith WH
(2009) Spatial reference and working memory across the lifespan in male Fischer
344 rats. Neurobiology of Aging 30(4): 645-655.
Cohn NB, Dustman RE, and Bradford DC (1984) Age-related decrements in Stroop color
test performance. Journal of Clinical Psychology 40(5): 1 244-1 250.
Colcombe S and Kramer AF (2003) Fitness effects on the cognitive function of older
adults: A meta-analytic study. Psychological Science 14(2): 125-130.
Craik FIM and Salthouse TA (eds.) (2000) The Handbook of Aging and Cognition,
2nd edn. Mahwah, NJ: Lawrence Erlbaum Associates.
Craik FIM and Salthouse TA (eds.) (2008) The Handbook of Aging and Cognition,
3rd edn. Philadelphia, PA: Psychology Press.
Davis HP, Cohen A, Gandy M, et al. (1990) Lexical priming deficits as a function of age.
Behavioral Neuroscience 104: 288-297.
De Luca C and Leventer RJ (2008) Developmental trajectories of executive functions
across the lifespan. In: Anderson V, Jacobs R, and Anderson PJ (eds.) Executive
Functions and the Frontal Lobes: A Lifespan Perspective, pp. 23-56. New York, NY:
Taylor Francis Group.
Fisk JE and Sharp CA (2004) Age-related impairment in executive functioning:
Updating, inhibition, shifting, and access .Journal of Clinical and Experimental
Neuropsychology 26(7): 874-890.
Foster SM, Cornwell RE, Kisley MA, and Davis HP (2008) Cognitive changes across the
life span. In: Qualls SH and Smyer MA (eds.) Changes in Decision Making Capacity
in Older Adults: Assessment and Intervention, pp. 25-60. Hoboken, NJ: Wiley.
Fristoe NM, Salthouse T, and Woodard JL (1 997) Examination of age-related deficits on
the Wisconsin Card Sorting Test. Neuropsychology 11(3): 428-436.
Harrison J and Buschwald J (1983) Eyeblink conditioning deficits in the old cat.
Neurobiology of Aging 4: 45-51 .
Hebert LE, Beckett LA, Scherr PA, and Evans DA (2001) Annual incidence of Alzheimer
disease in the United States projected to the years 2000 to 2050. Alzheimer Disease
and Associated Disorders 15(4): 169-173.
Henninger DE, Madden DJ, and Huettel SA (2010) Processing speed and memory
mediate age-related differences in decision making. Psychology and Aging
25(2): 262-270.
Herndon JG, Moss MB, Rosene DL, and Killiany RJ (1997) Patterns of cognitive
decline in aged rhesus monkeys. Behavioural Brain Research 87: 25-34.
Hertzog C, Kramer AF, Wilson RS, and Lindenberger U (2009) Enrichment
effects on adult cognitive development: Can the functional capacity of older
adults be preserved? Psychological Science in the Public Interest
9(1): 1-65.
Kramer AF, Colcombe SJ, McAuley E, Scalf PE, and Erickson Kl (2005) Fitness, aging
and neurocognitive function. Neurobiology of Aging 26S: S124-S127.
Kramer AF, Erickson Kl, and Colcombe SJ (2006) Exercise, cognition, and the aging
brain. Journal of Applied Physiology 1 01 : 1 237-1242.
Kramer AF, Hahn S, Cohen NJ, et al. (1999) Ageing, fitness, and neurocognitive
function. Nature 400: 418-419.
Lezak M, Howieson DB, and Loring DW (2004) Neuropsychological Assessment,
4th edn. New York: Oxford University Press.
Mahncke HW, Connor BB, Appelman J, et al. (2006) Memory enhancement in healthy
older adults using a brain plasticity-based training program: A randomized,
controlled study. Proceedings of the National Academy of Sciences 103(33):
12523-12528.
Oft A, Breteler MM, Van Harskamp F, et al. (1995) Prevalence of Alzhiemer's disease
and vascular dementia: Association with education. The Rotterdam study. British
Medical Journal 310: 970-973.
Owen AM, Hampshire A, Grahn JA, et al. (2010) Putting brain training to the test. Nature
465: 775-778.
Park DC, Lautensch lager G, Hedden T, Smith AD, Smith PK, and Davidson NS (2002)
Models of visuospatial and verbal memory across the adult life span. Psychology
and Aging 17(2): 299-320.
Phillips LH and Henry JD (2008) Adult aging and executive functioning. In: Anderson V,
Jacobs R, and Anderson PJ (eds.) Executive Functions and the Frontal Lobes:
A Lifespan Perspective, pp. 23-56. New York, NY: Taylor Francis Group.
Plassman BL, Langa KM, Fisher GG, et al. (2008) Prevalence of cognitive impairment
without dementia in the United States. Annals of Internal Medicine 148(6):
427-434.
Robbins TW, James M, Owen AM, et al. (1998) A study of performance on tests from
the CANTAB battery sensitive to frontal lobe dysfunction in a large sample of
normal volunteers: Implications for theories of executive functioning and
cognitive aging. Journal of the International Neuropsychological Society
4: 474-490.
Ryan L, Ostergaard A, Norton L, and Johnson J (2001) Search and selection processes
in implicit and explicit word-stem completion performance in young, middle-aged,
and older adults. Memory and Cognition 29(5): 678-690.
Salthouse TA (1 996) The processing-speed theory of adult age differences in cognition.
Psychological Reviews. 403-428.
Salthouse TA (2000) Aging and measures of processing speed. Biological Psychology
54: 35-54.
Salthouse TA (2006) Mental exercise and mental aging: Evaluating the validity of
the ‘use it or lose if hypothesis. Perspectives on Psychological Science
1(1): 68-87.
Salthouse TA (2009) When does age-related cognitive decline begin? Neurobiology of
Aging 30: 507-514.
Salthouse lb (20W) Major Issues in Cognitive Aging. NewYork, NY: Oxford University
Press, Inc.
Salthouse TA and Davis HP (2006) Organization of cognitive abilities and
neuropsychological variables across the lifespan. Developmental Review
26: 31-54.
Salthouse TA and Ferrer-Caja E (2003) What needs to be explained to account for
age-related effects on multiple cognitive variables? Psychology and Aging
18(1): 91-110.
Salthouse TA, Pink JE, and Tucker-Drob EM (2008) Contextual analysis of fluid
intelligence. Intelligence 36: 464-486.
Shallice T (1982) Specific impairments of planning. Philosophical Transactions of the
Royal Society of London, Biology 298: 1 99-209.
Snowden JS, Neary D, and Mann DMA (2002) Frontotemporal dementia. British Journal
of Psychiatry 1 80: 1 40-1 43.
Stern Y (2009) Cognitive reserve. Neuropsychotogia 47: 2015-2028.
Stroop JR (1935) Studies of interference in serial verbal reactions. Journal of
Experimental Psychology 1 8: 643-661 .
Stuss DT and Benson DF (1984) Neuropsychological studies of the frontal lobes.
Psychological Bulletin 95(1): 3-28.
Tombaugh TN (2004) Trail making test A and B: Normative data stratified by age and
education. Archives of Clinical Neuropsychology 19: 203-21 4.
Tucker-Drob EM, Johnson KE, and Jones RN (2009) The cognitive reserve hypothesis:
A longitudinal examination of age-associated decline in reasoning and
processing speed. Developmental Psychology 45(2): 431-446.
52 Aging and Cognition
Uttl B and Graf P (1997) Color-word Stroop test performance across the adult life span.
Journal of Clinical and Experimental Neuropsychology 1 9(3): 405-420.
Wang JJ and Kaufman AS (1993) Changes in fluid and crystallized intelligence across
the 20- to 90-year age range on the K-BIT. Journal of Psychoeducational
Assessment 1 1 : 29-37.
Woodruff-Pak DS (1988) Aging and classical conditioning: Parallel studies in rabbits
and humans. Neurobiology of Aging 9(5/6): 511-522.
Woodruff-Pak DS and Thompson RF (1988) Classical conditioning of the eyeblink
response in the delay paradigm in adults aged 18-83 years. Psychology and Aging
3(3): 219-229.
Zhang Z, Davis HP, Salthouse TA, and Tucker-Drob EM (2007) Correlates of individual,
and age-related, differences in short-term learning. Learning and Individual
Differences 17(3): 231-240.
Aging and the Brain
P J Colombo, M E Crawley, B S East Jr., and A R Hill, Tulane University, New Orleans, LA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Calcium hypothesis of aging The hypothesis that cognitive
and memory deficits associated with aging are due to
dysfunction of mechanisms for regulating levels of calcium
in neurons.
Cholinergic hypothesis of aging Pathological aspects of
aging, such as Alzheimer's disease and cognitive
impairments, are due to the selective loss of neurons in the
basal forebrain that produce the neurotransmitter
acetylcholine.
Dementia with lewy bodies (DLB) The second leading
cause of dementia behind Alzheimer's disease. The primary
characteristic of DLB is the presence of Lewy bodies which
are small clumps of protein that are found within the
intracellular fluid of cells in the brain.
Nonsteroidal anti-inflammatory drugs (NSAIDS)
NSAIDs, such as ibuprofen, aspirin, and naproxen, are
nonnarcotic analgesics that have anti-inflammatory effects
that may be beneficial in coping with harmful effects of
aging.
Oxidative stress Cellular damage resulting from the
accumulation of reactive oxygen species - molecules
produced by the mitochondria in cells as a byproduct of
energy metabolism.
For much of the history of modern brain science, it has been
widely held that the mammalian brain undergoes significant
developmental changes early in life, followed by a steady,
and occasionally precipitous, decline across the life span.
At the heart of this thinking is the assumption that plasticity -
the ability to change structure and function - is a feature of the
developing brain but not of the adult brain. This and other
long-held assumptions have been challenged in recent dec-
ades, and we are coming to a new understanding of the brain's
amazing ability to remain plastic and to change across the life
span. For example, the common belief that we lose significant
numbers of neurons as we age has been reconsidered. Indeed,
it is now well established that the adult brain retains the
ability to produce new neurons. In this article, we first examine
prominent theories of brain aging at the cellular level of analy-
sis. These include the cholinergic hypothesis, oxidative stress,
the calcium hypothesis, and genetic models of aging. Then,
we discuss some of the age-associated changes in the vascular
system and neuron numbers in the various systems of the
brain. Next, common age-associated neuropathological condi-
tions and therapeutic interventions are explained (excluding
Alzheimer's disease which is addressed in a separate article).
We conclude with some rather encouraging evidence that
it may be possible to maintain a healthy brain well into
advanced age.
Major Neuronal Theories of Brain Aging
Cholinergic Hypothesis of Aging
The cholinergic hypothesis of aging states that loss of func-
tion among cholinergic neurons is a primary contributor to
Alzheimer's disease and loss of cognitive function as we age.
Neurons of the basal forebrain are the primary source of
the neurotransmitter acetylcholine, and they extend processes
throughout the brain to regions that include the limbic system
and the neocortex. The cholinergic hypothesis is supported by
reports of relationships between the integrity of basal forebrain
cholinergic neurons and the degree of age-related impairment
in both animal and human subjects. These correlational results,
however, do not establish the contribution of age-related
changes in the basal forebrain cholinergic system to cognitive
impairments. Evidence that supports a functional link comes
from two main sources: treatments targeted at improving cho-
linergic function relieve impairment in aged subjects, and sensi-
tivity to anticholinergic agents increases with aging. Evidence
that is not consistent with the cholinergic hypothesis includes
reports that selective removal of cholinergic neurons in the
basal forebrain of a rodent model of aging fails to reproduce
impairments commonly found in aged subjects. Moreover,
therapeutic interventions that target the cholinergic system,
which included until recently all drugs approved by the
Food and Drug Administration for use for the treatment of
Alzheimer's disease, produce modest effects, at best. Thus,
while the cholinergic hypothesis has generated considerable
research and revealed targets for therapeutic intervention, it is
now widely held that other types of neurons, in addition to
cholinergic neurons, are likely to contribute to age-associated
deficits. In the next section, we consider some of the theories
that describe why and how other types of neurons might
become dysfunctional or die off as we age.
Oxidative Stress
In this section, we discuss a series of hypotheses on the causes
of brain aging that are all centered on the theme of oxidative
stress. Oxidative stress is the gradual accumulation of reactive
oxygen species (ROS) within the cells of the body as we age
(ROS is a subset of 'free radicals'). ROS are biological com-
pounds such as superoxide anion radical (0 2 ), hydrogen per-
oxide (H 2 0 2 ), and the hydroxyl ions (OTP). These molecules
are produced in mitochondria, which are energy-producing
components found inside the cells of the body. If not kept in
check, ROS can lead to the oxidation of lipids, proteins, and
DNA that can cause their destruction, malfunction, or muta-
tion. Accordingly, the body has a host of systems designed to
53
54 Aging and the Brain
watch over ROS. These usually take the form of enzymes (such
as superoxide dismutases) and metabolites (such as vitamin C).
The body has two main pathways with regard to ROS: one that
produces ROS and one that breaks ROS down. The oxidative
stress theory proposes that some process or combination of
processes causes an imbalance in these two pathways, which
leads to an increase in oxidative damage and what we know as
aging. ROS are utilized by various leukocytes, also known as
white blood cells, in the body to mount an immune response
when faced with pathogens. The production of ROS by mito-
chondria and the antagonistic response by the immune system
represent two distinct pathways, and also two distinct aspects
of the oxidative stress theory of aging that we discuss next.
Inflammation is a natural part of the body's defenses
against invading pathogens and injury. The brain contains a
collection of cell types, including microglia, that are ready to
respond to potential threats. When a pathogen is detected,
microglia are activated and proinflammatory cytokines -
proteins secreted by cells of the immune system to carry signals
between cells - such as tumor necrosis factor a, interleukin- 1,
and interleukin-6 are released. These in turn activate more
microglia, which attack invading cells, destroy debris, and
call in cellular reinforcements. Once the threat is eliminated,
they then return to an inactive state until they are called upon
again. There is mounting evidence that, as we age, microglia tend
to remain in the active state and do not return to their inactive
state. This may lead to a prolonged or increased immune
response and consequently, an increase in oxidative stress.
The respiratory burst system is an important immune response
in the brain in which activated microglia use an enzyme known
as NAD PH oxidase to rapidly produce large amounts of the ROS
superoxide, which is used to kill the intruding cells via oxidative
damage. In addition to an increase in ROS production, activated
microglia can also damage the body in the form of autoimmune
attacks. Activated microglia have been implicated in the neuro-
nal death associated with Alzheimer's and Parkinson's diseases
among others. Given the link between inflammation and degen-
erative disease, it is easy to see why researchers would want to
investigate the use of anti-inflammatory medication to treat
these symptoms as we discuss later in this section.
The mitochondrial DNA theory of aging proposes that an
accumulation of mutations and deletions in mitochondrial
DNA contributes to aging processes. Mitochondrial DNA is
the DNA found in the mitochondria - a structure in cells that
generates adenosine triphosphate (ATP), which provides a
source of chemical energy for cells. Mitochondrial DNA is
involved in functions such as cellular metabolism, calcium
storage, and oxidative phosphorylation - a metabolic process
that uses energy released by the oxidation of nutrients to pro-
duce ATP. Oxidative phosphorylation results in reactive oxygen
species (ROS), which can be damaging to cells as described
previously. Because of their role in oxidative phosphorylation,
mitochondrial DNA uses oxygen at a very high rate and thus
produces the majority of reactive oxygen species in a cell.
The accumulation of ROS can have harmful effects on the
brain's neurons, leading to oxidative damage and cell death.
There is evidence that aged animals have higher levels of ROS
in comparison with younger animals, suggesting again that
aging is a consequence of the damage done by the accumula-
tion of free radicals in cells.
In addition to promoting free radical production, mito-
chondrial DNA is more susceptible to mutations than DNA
found in the nucleus of cells, and has lower DNA repair.
Mitochondrial DNA is not fully protected by histones - the
proteins that package DNA - and this makes the DNA more
susceptible to free radicals and mutations. One current theory
suggests a cyclic mechanism for mitochondrial DNA mutations
and cell death due to ROS. Because mitochondrial DNA is
highly susceptible to mutations, the accumulated mutations
are thought to impair cellular oxidative phosphorylation, lead-
ing to increased levels of ROS. Increased ROS, in turn, leads to
further mutations and the cycle continues. Mutations in mito-
chondrial DNA have been shown to increase with age as well as
in the presence of neurodegenerative disorders. In addition,
mice with point mutations in mitochondrial DNA as well as
full deletions showed reduced life span and early onset of age-
related characteristics. Also, human studies show that mito-
chondrial DNA deletions increase with age and some research
indicates that these deletions occur in cells with higher levels of
oxidative damage. ROS are not the only molecules implicated
in the aging process. In the next section, we see what role
calcium is believed to play in the aging process.
Calcium Hypothesis
While it is well known that calcium helps promote healthy
bone structure, it also plays an important role within neurons.
Calcium molecules often serve as 'second messengers' within
neurons by helping to pass along messages received from other
neurons via receptors. One way in which they pass messages
is through a type of channel in the neuron's membrane that
opens in response to signals sent from neighboring neurons.
This channel allows calcium ions to enter the cell. The calcium
ions that enter the cell are highly regulated as only a minute
amount will activate a host of cellular processes. As one might
expect, if the cell is particularly sensitive to calcium, then dysre-
gulation of calcium ions may be responsible for what we call
aging. This line of investigation is referred to as 'the calcium
hypothesis.' As we get older, there is an increase in the concen-
tration of L-type voltage-gated calcium channels and more cal-
cium tends to enter the cell each time the calcium channels
are opened. The age-associated increase in calcium channels is
correlated with decreased cognitive function in aged animals
whereas blocking these channels improves animals' memory.
In addition to the dysregulation of extracellular sources of
calcium, aging may also be due to dysfunction of intracellular
calcium release. Each cell contains an organelle that is called
the endoplasmic reticulum, and one of its many functions is to
store calcium reserves until needed. When called upon, the
endoplasmic reticulum can release the calcium by way of either
ryanodine receptors or IP 3 receptors. Chronic elevation of
intracellular calcium is related to excitotoxicity - a form
of cell death - in tissue cultures. The ryanodine receptors medi-
ate a process known as calcium-induced calcium release in
which calcium from the endoplasmic reticulum is released
when the receptor detects calcium. When ryanodine is adminis-
tered to the ryanodine receptors, it binds in the place of calcium
and this has been shown to help maintain calcium levels.
Understanding the roles of cellular and molecular mechan-
isms of aging may eventually prove useful in combating their
Aging and the Brain 55
negative effects. In order to investigate how specific molecular
changes influence development across the life span, scientists
have come up with several different kinds of genetic models;
we now look at those models that focus on aging.
Genetic Models
Certain strains of mice are often inbred to create a population
of animals with similar characteristics so that a specific phe-
nomenon or characteristic can be studied in a controlled man-
ner. The senescence accelerated mouse (SAM) model of aging
provides a rodent model of aging and aging-related disorders
and is comprised of two main strains: accelerated senescence-
prone (SAM-P) and accelerated senescence-resistant (SAM-R)
mice. They were developed in 1973 when Dr. Toshio Takeda
and his research group noticed that, while maintaining a
colony of a specific strain of mice, some litters would show
accelerated characteristics of aging and had a shorter life
span. When compared to their SAM-R counterparts, the SAM-P
mice have shorter life span, hearing impairments, brain atro-
phy, learning and memory impairments, cataracts, osteoporo-
sis, and immune system impairments. SAM-P mice also have
higher levels of ROS. Some researchers attribute this fact to
functional impairments in mitochondrial DNA, resulting in
increased output of ROS. As stated previously, increased levels
of ROS are one potential mechanism by which age-related
neurodegeneration and cell death can occur. SAM-P mice also
show accelerated age-related decline in some important
cellular functions such as cell survival and proliferation. This
impairment of cellular function can lead to degeneration of
tissues.
SAM models provide a useful research tool for aging stud-
ies. One strain of SAM-P (SAMP-8) shows learning and mem-
ory deficits from early in development. This particular strain
shows higher levels of ROS early on and is of particular interest
to researchers in the dementia field because SAMP-8 mice show
early onset of development of (3-amyloid plaques, which are
implicated in Alzheimer's disease and other forms of dementia.
SAM-P strains can be useful in the development of drug thera-
pies and treatments for age-related disorders because of their
varying diversity. For example, while one strain may show a
deficit in learning and memory, another strain may show
osteoporosis, while yet another strain would demonstrate
frontal lobe atrophy. Given the variety of age-related disorders
present in strains of the SAM-P mice, they are a useful and
effective tool for aging research. Now that we have covered
some prevailing ideas about aging on a cellular-molecular
level, we discuss aging-related changes at a broader neuroana-
tomical view in the paragraphs that follow.
Systems-Level Anatomical Changes in the Brain
with Age
Vascular Changes
The brain needs a constant supply of blood to carry oxygen and
essential nutrients to the neurons in the brain in order for them
to function properly. The carotid arteries are the two main
arteries that carry blood to the brain, while other arteries
within the brain carry the blood to specific regions. If blood
supply to a region of the brain is cut off, then surrounding
neurons will begin to die and the area will become damaged
in what is commonly known as a stroke. Stroke can cause
permanent damage to brain regions, leading to serious disabil-
ities or death. Given that the incidence of stroke increases with
age, it is important to be well informed about the associated
risk factors for a stroke.
The vast majority of strokes are classified as ischemic
strokes. These occur when blood flow in the brain becomes
blocked. This blockage is usually made up of a fatty deposit
known as an atheroma or a plaque. The plaque builds up
in arteries, causing the blood vessel walls to harden and shrink
in size, which is a condition known as atherosclerosis.
The buildup becomes larger as more blood platelets aggregate
together in the area, contributing to the formation of a blood
clot. An ischemic stroke can be further classified as either a
thrombotic or an embolic stroke, depending on where the clot
initially forms. If the clot forms inside the blood vessels of the
brain, then it is classified as a thrombosis; if it forms elsewhere
in the body before traveling to the brain, then it is considered
an embolism. An intracerebral hemorrhagic stroke is a less
common type of stroke in which a weak spot on a blood
vessel known as an aneurysm enlarges and bursts. Blood spills
out into the brain when the aneurysm ruptures, causing an
increase in pressure and extensive damage to surrounding
areas. A subarachnoid hemorrhage is similar to an intracere-
bral hemorrhage, but the blood vessel bursts near the brain's
surface and the blood spills out into the region between the
skull and the brain known as the subarachnoid space.
Several risk factors can increase the likelihood of having a
stroke, one of which is age. Even though a stroke can occur at
any age, the majority of strokes happen after middle age and
the incidence increases the older one gets. High blood pressure
puts a person at greater risk of having a stroke because it causes
stress on the heart and blood vessels that can lead to athero-
sclerosis or rupturing of an aneurysm over time. High blood
pressure and high cholesterol are very unhealthy for the car-
diovascular system. Maintaining a healthy cardiovascular and
circulatory system can greatly reduce the chance of having a
stroke. High blood pressure and high cholesterol can be treated
through exercise, weight loss, regulation of diet, and with
medication if necessary. Having a condition known as atrial
fibrillation increases the risk of stroke. An atrial fibrillation
occurs when the top two chambers of the heart (the atria) beat
irregularly, causing blood to pool in the heart and clots to form.
This can result in an embolic stroke if one of the clots moves to
the brain. Diabetics are at greater risk for having a stroke because
they tend to have high blood pressure; furthermore, the damage
to the brain during a stroke is greater when blood sugar is very
high. Smoking dramatically increases the chance of a stroke,
since smoking reduces oxygen levels in the blood and causes
the heart to work harder. A transient ischemic attack (TIA) or
'mini-stroke' is caused by temporary blockage of the blood
vessels. It does not cause permanent damage but should be
recognized as a warning sign of a future stroke.
Diagnostic tests need to be run before administering treat-
ment to a person who has had a stroke. Blood thinners
are an effective treatment for ischemic stroke but if the stroke
is hemorrhagic, then administering blood thinners would
increase bleeding and damage. Most physicians use X-ray and
56 Aging and the Brain
brain imaging techniques such as CT (computed tomography)
and MRI (magnetic resonance imaging) to find the blockage or
area of bleeding. An angiogram is an X-ray technique in which
dye is injected into arteries to more accurately locate obstruc-
tions and aneurysms. An angiogram can be used as a screening
tool to look for abnormal blood vessels. Once a diagnosis is
made, treatment can begin. For ischemic strokes, the primary
medical treatments are administering a clot-dissolving tissue
plasminogen activator (tPA), anticoagulant drugs, antiplatelet
drugs, and low dose aspirin. The main surgical treatments used
for ischemic strokes are a carotid endarterectomy (removing
the blockage from the artery), performing an angioplasty to
reduce plaque buildup, and setting up a stent to improve blood
flow in the vessel. The primary treatment for hemorrhagic
strokes is through surgical interventions to stop the bleeding
and repair the damaged blood vessels.
Neuron Loss
As reviewed in the previous sections, most theories of brain
aging are built around the notion that the most likely cause of
age-associated cognitive decline is the widespread loss of neu-
rons. This view is changing rapidly due to development of new
techniques for assessing neuron numbers and better methods
for early diagnosis of pathological conditions. Two regions of
the brain have come under close scrutiny - the neocortex, which
is responsible for higher order cognition, and the hippocampus,
which is implicated in long-term memory. It has been known
since the 1950s that the hippocampus is important for memory,
and memory impairments associated with aging are similar to
those caused by damage to the hippocampus. However, it is
now known that there is no significant age-associated loss of
neurons in the hippocampus, even among aged subjects with
severe memory impairments. This fact was supported by a
rodent model of aging and a newly developed technique for
unbiased counts of neuron numbers. These findings have since
been confirmed among both nonhuman and human primates.
The assumption that neurons are lost in the cerebral cortex
as we age is also coming under increasing challenge. When
pathological conditions such as Alzheimer's disease are con-
trolled for, it appears that there is little loss of cortical neurons
in a variety of regions including visual and motor cortex, and the
prefrontal cortex, which is an area important for using experi-
ence to guide choices about current situations. This finding is
encouraging as it suggests that there may be a useful distinction
between 'successful' and 'pathological' aging. In addition,
instead of focusing exclusively on age-associated neuron loss,
investigators are beginning to test hypotheses about how
changes in connectivity between neurons may account for def-
icits associated with aging, particularly in the prefrontal cortex,
an area strongly implicated in age-associated cognitive decline.
In specific, recent evidence suggests that there is a decrease in
both excitatory and inhibitory synapses, the points of commu-
nication among neurons, in certain regions of the prefrontal
cortex. In addition, there is evidence that the vesicles that con-
tain neurotransmitters at certain synapses may change with age
and this may be related to cognitive status among the elderly.
These and other functional changes in communication among
neurons will continue to be a promising focus of investigation.
Age-Associated Neurological Disorders
Parkinson’s Disease
Parkinson's disease is a chronic, progressive, degenerative dis-
order of the central nervous system. Common symptoms of the
disorder include muscle rigidity, tremor, bradykinesia (slowing
of movement), postural instability, and cognitive problems.
The primary symptoms of Parkinson's disease are a result of
decreased stimulation of the motor cortex by the basal ganglia,
a group of nuclei in the forebrain associated with motor con-
trol as well as motor and cognitive habits. The decreased signal
from the basal ganglia is due to degeneration of neurons in the
substantia nigra, a nucleus of cells in the midbrain that release
the neurotransmitter dopamine. Dopamine is a catecholamine
neurotransmitter that plays an important role in sleep, cogni-
tion, motivation, learning, attention, and movement. Within
the basal ganglia, there is a direct pathway that provides an
excitatory signal to the cortex and facilitates voluntary move-
ment and an indirect pathway that provides an inhibitory
signal to the cortex and inhibits movement. The ultimate result
of neuronal damage in Parkinson's disease is inhibition of the
direct pathway and excitation of the indirect pathway, both
leading to the classic symptoms of tremor, muscle rigidity,
slow movement, and postural instability.
The cause of cell loss in the substantia nigra is unknown,
and Parkinson's disease is considered to be idiopathic (having
no known cause). However, in recent years, theories have
emerged as to the etiology of neuronal degeneration. One
proposed mechanism of cell death in Parkinson's results
from the accumulation of a protein common to neural tissues,
a-synuclein, as a complex which forms what are known as
Lewy bodies (described in the next section). Another proposed
mechanism of cell death involves the oxidative stress model of
aging which we described previously. In this case, cell death is a
result of iron accumulation in the substantia nigra. Cells in the
substantia nigra have a large amount of melanin, resulting in
the dark color that gives the structure its name ('black sub-
stance'). The iron binds to neuromelanin and ultimately results
in accumulation of reactive oxygen species (ROS) which are
toxic to cells and can result in cell death.
There is currently no cure for Parkinson's disease, although
treatments can be effective at managing symptoms. The most
widely used treatment is levodopa (L-Dopa), which is the
biological precursor to dopamine and increases the levels of
dopamine in the brain. L-Dopa is often given in conjunction
with dopa-decarboxylase inhibitors. Dopa-decarboxylase is the
mechanism by which L-Dopa is metabolized into dopamine,
and the inhibitors help to prevent it from being metabolized
elsewhere in the body before it reaches the brain. Even
with dopa-decarboxylase inhibitors as a part of treatment,
only about 1-5% of L-Dopa reaches dopaminergic neurons.
Although L-Dopa is the most commonly used line of treatment,
other therapies are available such as agents that act by stimu-
lating dopamine receptors. In addition, there are drugs avail-
able that inhibit the breakdown of dopamine, such as
monoamine oxidase-B inhibitors. These dmgs interfere with
the naturally occurring enzyme that breaks down excess dopa-
mine in the brain, effectively increasing levels of dopamine
in the brain.
Aging and the Brain 57
In addition to therapies that pharmacologically enhance
the dopaminergic system, there are also surgical treatments.
The most widely used surgical treatment of Parkinson's disease
is deep brain stimulation, or DBS. DBS involves the surgical
implantation of what has been likened to a 'brain pacemaker'
that sends electrical pulses to specifically targeted areas of the
brain including the internal capsule and the subthalamic
nucleus. The direct effect of DBS on neurotransmitter release
and overall brain physiology is not fully known or understood,
but it can provide relief from major Parkinson's symptoms.
However, DBS is only used in the most extreme cases, typically
after other lines of therapy have failed. Other surgical therapies
that are not widely used but have shown some promise as
potential treatments are lesions of the subthalamic nucleus
and the globus pallidus. Though other options exist, DBS
remains the most commonly used surgical therapy for Parkin-
son's disease.
Lewy-Body Dementia
Dementia with Lewy bodies (DLB), often referred to as Lewy
body dementia, is the second leading cause of dementia
behind Alzheimer's disease. The primary characteristic of DLB
is the presence of Lewy bodies in the brain, named after
Dr. Friederich Lewy in 1912. Lewy bodies are small clumps of
protein that are found within the intracellular fluid of cells in
the brain. The primary protein that makes up these Lewy
bodies is known as a-synuclein, which is a naturally occurring
protein believed to be used in the formation of vesicles, which
help store neurotransmitters used for cellular communication.
It is not currently known exactly what causes the a-synuclein
protein to go rogue, but one theory is that the body can start
to overproduce a-synuclein. Another is that some malfunc-
tion eventually makes the protein insoluble and more able
to clump together. One theory even proposes that Lewy
bodies might actually be 'good,' and are made by the body to
inactivate rogue a-synucleins that have not been properly
eliminated.
In addition to the production of Lewy bodies, DLB also
leads to a decrease in certain neurotransmitter systems such as
acetylcholine and dopamine. Similar to Parkinson's disease, a
dysfunction of these chemical messengers can lead to symp-
toms such as slow movement, rigidity, and a shuffling gait.
Along with Parkinson-like symptoms, people with DLB typi-
cally exhibit a fluctuating level of cognitive impairment. One
day the patient may seem alert and active but the next day will
be sluggish and easily confused. Another common symptom of
those with DLB is frequent hallucinations, which are thought
to be related to the formation of Lewy bodies in the temporal
lobe. Unfortunately, there is currently no known cure for DLB.
One commonly prescribed treatment is a group of drugs
known as cholinesterase inhibitors, which stop the breakdown
of acetylcholine, allowing it to remain active longer. Caution
must be used when medicating DLB patients with neuroleptics
(antipsychotics) because they have been shown to be sensi-
tized to these treatments and may develop the aptly named
neuroleptic malignant syndrome. This syndrome is character-
ized by various homeostatic imbalances such as muscle rigid-
ity, fever, and autonomic dysfunction.
Interventions
One potential intervention for age-related functional decline is
caloric restriction. Caloric restriction involves reducing the
amount of calories consumed, either by reducing overall food
intake or by reducing the caloric content of the food without
altering overall consumption. Reducing a nutritious diet by
20-50% can result in beneficial effects such as: increased life
span, increased protection from harmful effects of stress, and
slowed onset of many age-related diseases. Additionally,
reduced calorie diets show lowered risk for disorders such as
Parkinson's disease and Alzheimer's disease. Caloric restriction
has been shown in animal models to be an effective strategy
for combating age-related cognitive decline. Mouse models of
Alzheimer's disease show slowed progression of the disease
when the caloric intake is reduced 30-40%. Also, squirrel
monkeys put on a reduced calorie diet have fewer (5-amyloid
plaques, a congregation of misfolded proteins and amino acids
widely thought to be a main contributor to Alzheimer's dis-
ease. Additionally, long-term studies of rhesus monkeys show
that animals kept on a restricted diet chronically (about 30% of
control animals) show better health overall when compared
on measures such as rate of aging and incidence of disease.
Though caloric restriction has been shown to be a
promising potential intervention for age-related health decline,
in practice, it proves slightly more difficult. For example, issues
with implementation may come about as it can be difficult to
maintain a chronically restricted diet. Additionally, quality of
life can be affected by such a change in one's dietary intake.
To combat these issues, research is now focused on areas of
caloric restriction mimetics. This area of research is aimed at
development of compounds, such as resveratrol, that can
mimic caloric restriction by targeting both stress and metabolic
pathways.
Caloric Restriction Mimetics
Resveratrol is a polyphenolic phytoalexin produced in plants
and is primarily found in the skin of grapes, which is a constit-
uent of red wine. It is also found in blueberries, peanuts,
pistachios, and groundnuts in lower concentrations. It has
been investigated extensively for its cardio-protective, cancer
preventive, estrogenic, anti-inflammatory, and antioxidant
effects. Resveratrol has been implicated as a possible antiaging
drug because of its similarities with caloric restriction in
improving overall health. Resveratrol circulating in the blood
stream is rapidly metabolized by the body and has a short half-
life of decay. The chemical products of resveratrol's me-
tabolism have a much longer half-life than resveratrol in its
unconjugated form. Accordingly, it is difficult to tell how much
resveratrol is actually accumulating in the tissues of the body
before being broken down. Resveratrol has been shown to
lower the risk of cardiovascular disease by increasing vasodila-
tation and reducing the aggregation of platelets in blood ves-
sels. Since resveratrol selectively inhibits COX1 (see section on
the role of coxl), it decreases the likelihood of restricted blood
flow due to blockage. Resveratrol also promotes vasodilation
by improving nitric oxide signaling, which is an important
mechanism for protecting against ischemic damage. Since res-
veratrol can cross the blood brain barrier, it promotes vascular
58 Aging and the Brain
protection in the brain from cerebral ischemias, which are a
major cause of stroke. Research with rodents has demonstrated
that resveratrol can decrease motor damage and the amount of
tissue death following a stroke.
In addition to regulating vascular homeostasis, COX
enzymes play an important role in the body's inflammatory
and immune response. Resveratrol has been shown to decrease
inflammation by selective COX inhibition and by decreasing
superoxide formation in rats. Resveratrol has also been shown
to reduce reactive oxygen species (ROS) damage as well
as promote antioxidant effects and reduce inflammation.
The exact mechanisms by which resveratrol reduces oxidative
stress are not fully understood and are still under investigation.
In aging research, resveratrol has been described as a possi-
ble mimetic for caloric restriction. Early research yielded excit-
ing results indicating that the life span of lower organisms
could be extended by manipulations of the gene for silent
information regulator 2 (Sir2). However, these increased
longevity results were not replicated in rodent models. Sir2
is part of a class of proteins called sirtuins that are histone
deacetylases. Extra copies of Sir2 have been correlated with
increased life span in yeast, worms, and flies. Mammals have
seven sirtuins and S1RT1 is the mammalian homolog of Sir2.
Resveratrol has been shown to activate SIRT1 in mammals, but
this has not been shown to extend life span in rats. Resveratrol
has numerous protective qualities that are beneficial to main-
tain good health and mimic metabolic changes associated with
caloric restriction.
Nonsteroidal Anti-Inflammatory Drugs
In addition to caloric restriction, nonsteroidal anti-inflammatory
drugs (NSAIDs), such as ibuprofen, aspirin, and naproxen, can
be beneficial in coping with harmful effects of aging. These
drugs are nonnarcotic analgesics that have anti-inflammatory
effects and fever-reducing abilities. NSAIDs block the synthesis
of prostaglandins, which cause inflammation at the site of
an injury to protect from infection and cause increased pain
sensitivity. Nonselective NSAIDs prevent the release of prosta-
glandins by directly inhibiting the enzymes Cyclooxygenase 1
(COX1) and Cyclooxygenase 2 (COX2). These enzymes play
an important role in the body's inflammatory response as well
as in regulating vascular homeostasis. COX1 synthesizes
thromboxane A2, which promotes clot formation through
increasing platelet aggregation. In addition, thromboxane hi
constricts blood vessels in order to raise blood pressure and
retain heat. Prostacyclines are synthesized by COX2 to prevent
platelet aggregation and dilate the blood vessels. Together, the
inhibition of these compounds reduces inflammation and
associated pain. In high doses, nonselective NSAIDs have neg-
ative side effects on the upper and lower gastrointestinal (GI)
tract that can result in painful ulcers. The inhibition of COX1
with NSAIDs diminishes the protection of the GI tract's
mucosa lining. Selective NSAIDs called COX2 inhibitors were
created in an attempt to reduce inflammation and pain
without damaging the mucosa lining of the GI tract. Early
clinical trials demonstrated that COX2 inhibitors significantly
decreased gastrointestinal problems associated with NSAIDs.
However, the results of these clinical trials also demonstrated
an increased risk of cardiovascular problems. The strongest
COX2 inhibitors, such as rofecoxib, were significantly corre-
lated with cardiovascular complications when compared with
placebos and nonselective NSAIDs. Currently, celecoxib is the
only COX2 inhibitor available in the United States and FDA
changed its label to include warning about possible cardiovas-
cular problems.
The majority of individuals taking high doses of NSAIDs are
those suffering from arthritis, which causes inflammation in
the joints and considerable pain. The incidence of osteoarthri-
tis is correlated with age, so it is important to understand the
risks and benefits of using high doses of NSAIDs. Low dose
aspirin has been shown to possess cardio-protective effects and
it is highly recommended that the elderly and those indivi-
duals at high risk for heart attack and stroke take low dose
aspirin. Aspirin irreversibly inhibits the ability of individual
platelets to produce thromboxane hi, and therefore decreases
the likelihood of dangerous clots forming in the blood vessels.
Ibuprofen has been shown to interfere with the irreversible
binding of aspirin to platelets, resulting in reduced platelet
aggregation. Therefore, it is important to be knowledgeable
about possible drug interactions when taking low dose aspirin.
Another type of inflammation, called neuroinflammation,
is a pathological characteristic of Alzheimer's disease (AD) in
affected brain regions where damage is sustained. Misfolded
proteins called fl-amyloid plaques and neurofibrillary tangles
develop in the brain of individuals with AD. These plaques and
tangles prevent the communication between neurons and
eventually lead to neuronal cell death. Neuroinflammation is
concentrated around these (5-amyloid plaques and neurofibril-
lary fibers. This finding supported the hypothesis that anti-
inflammatory dmgs may aid in the prevention or treatment
of AD. Epidemiological evidence demonstrated a lower inci-
dence of AD in individuals with arthritis who were currently
taking high doses of NSAIDs. Subsequent clinical trials, how-
ever, did not support this hypothesis and some trials were
suspended due to health concerns over using high doses of
NSAIDs. It is too early to recommend the use of NSAIDs
for AD prevention based on current research, but scientists
are still investigating the relationship between neuroinflamma-
tion and AD.
Maintaining a Healthy Brain
Successful Aging
There are many things one can do to help prevent or prolong
the onset of age-related deficits. One of the most effective
methods observed thus far is readily accessible and inexpen-
sive: exercise. Cardiovascular health can be a strong predictor
of mental health as we age, and several studies have shown
a link between increased blood pressure in mid-life and
diminished cognitive functions in later life. Studies have also
linked antihypertensive treatment with reduced cognitive
decline in people with hypertension compared to those
who remain untreated. Another major cardiovascular concern
linked to cognitive decline is dyslipidemia, an abnormality in
the amount of lipids (such as cholesterol) in the blood. Studies
have shown a relationship between cognitive decline and high
levels of cholesterol for both high-density and low-density
cholesterol. There is not yet strong evidence to support the
Aging and the Brain 59
use of statins to treat cognitive decline related to dyslipidemia.
Type 2 diabetes has been implicated in a more rapid cognitive
decline when compared to nondiabetics. The duration of
diabetes has also been shown to correlate with increased risk
of cognitive decline. There has been modest success thus far
with the use of antidiabetic medication to ameliorate the
symptoms. Finally, a link has also been shown between hyper-
insulinemia, or excessive levels of insulin in the blood, and
cognitive decline.
Exercise does more than just improve the brain through
cardiovascular health. Exercise has also been shown to increase
the levels of a protein in the brain known as brain-derived
neurotrophic factor, or BDNF. BDNF is one of many chemicals
known as neurotrophins that encourage the growth and sur-
vival of neurons within the brain. In rats, lack of exercise can
decrease the expression of BDNF whereas as little as a week's
exposure to exercise has been shown to greatly increase BDNF
levels. Not only does exercise increase BDNF but it also
increases the production of new neurons in the hippocampus,
a brain structure associated with learning and memory. Exer-
cise has also been shown to facilitate the induction and main-
tenance of long-term potentiation, a phenomenon in which
two cells are better able to communicate with one another as
the result of activation. On a more practical level, exercise has
also been shown to improve rats' performances on the Morris
water maze in which they are tasked with finding a platform
that is hidden below the surface of the water. The improvement
seen in the Morris water maze and other such memory tasks
can be abolished if the animal has its BDNF signaling blocked,
implicating BDNF as a likely component of this pathway.
Many studies have shown that our ability with language
and verbal manipulation is also implicated in aging. One
study, sometimes referred to as the 'Nun Study,' included
several members of the convent of the School Sisters of Notre
Dame in a long-term study of their cognitive abilities, and the
nuns agreed to donate their brains for examination upon their
deaths. One benefit of conducting an aging study among a
convent of nuns is that from the time they entered (with the
average age being 22) each nun has lived a relatively similar
life and thus from then on, most nonbiological differences
between them were kept to a minimum. Before taking their
vows, each nun was required to write an autobiography of their
life up to that point. When the brains were analyzed, there was
a strong correlation between those nuns that aged successfully
and idea density scores (the ability to express a large amount of
ideas in a small amount of words) and grammatical complexity
in their autobiographies written several decades earlier. One
evaluation of the data even suggests that those nuns that had a
more positive outlook on life were healthier and lived longer.
In this article, we have discussed many of the important
mechanisms of aging and age-related decline in brain function,
as well as potential therapies for combating these negative
aspects of aging. There is still much debate about whether what
we consider aging is a pathological condition or if there are
simply various degrees of severity to the declines and deficits
associated with the aging process. While aging is often viewed
with emphasis on the degenerative diseases (Alzheimer's dis-
ease, Parkinson's disease, dementia with Lewy bodies), several
of the theories presented here involve a more gradual accumula-
tion of damage as part of a general aging process. At the cellular
level, aging is currently thought to be largely due to accumula-
tion of ROS and their harmful effects on cells. In addition to
ROS, calcium dysregulation may play a significant role in the
age-associated cognitive deficits. Research in the field of aging
has led to the development of major theories of brain aging
centered on the cholinergic system, oxidative stress, mitochon-
drial DNA, and calcium dysregulation. To combat the conse-
quences of aging, researchers have studied genetic models like
the senescence accelerated mouse (SAM) model of aging as well
as possible interventions that include caloric restriction, resvera-
trol, and NSAIDs. Current research has provided us with many
ways to intervene and potentially stave off the aging process, but
there is still much to be explored. As the average life span and the
number of aged individuals increases (especially with a looming,
aging 'baby boomer' generation), our efforts to understand the
mechanisms of brain aging are increasingly coming to the fore-
front of scientific research.
See also: Aging and Cognition; Alzheimer's Disease; The Brain; Brain
Chemicals: Global Projections of Ancient Aromatic Neurotransmitters;
Brain and Behavior Relationships; Memory; Memory, Neural
Substrates.
Further Reading
Burke SN and Barnes CA (2006) Neural plasticity in the ageing brain. Nature Reviews
Neuroscience 7: 30-40.
Lin MT and Beal MF (2006) Mitochondrial dysfunction and oxidative stress in
neurodegenerative diseases. Nature 443: 787-795.
Wickelgren I (1996) For the cortex, neuron loss may be less than thought. Science
273: 48-50.
Relevant Websites
http://www.nia.nih.gov/- U.S. National Institutes of Health, National Institute on Aging.
Agnosia (including Prosopagnosia and Anosognosia)
A Berti and M Neppi-Modona University of Torino, Torino, Italy
© 2012 Elsevier Inc. All rights reserved.
Glossary
Focal deficit This expression indicates a symptom that
depends on a circumscribed (opposite to diffuse) brain damage.
Modularity In neuropsychology, the term modularity refers
to the idea that cognition and consciousness have a
composed structure made up of relatively independent and
domain-specific processing components.
Neuropsychology It is a multidisciplinary scientific
approach to the study of cognitive brain functions in normal
and brain-damaged subjects through paradigms derived
from Psychology and Neurology.
Presemantic memory It is the collection of visual-structural
knowledge about stimuli in the absence of knowledge about
their identity and function.
Processing stage It is the result of a transformation of the
information entering a biological or nonbiological system
from one status to another.
Semantic memory It is the collection of knowledge,
meaning, and concepts that have been acquired over the
course of the life span.
Vestibular reflex The vestibuloocular reflex is elicited by
irrigating cold or warm water into the external auditory
canal. This maneuver evokes reflex eye movements, and it is
commonly used to evaluate the function in the peripheral
vestibular system or, together with other tests, to evaluate
brain stem death. More recently, it has been employed for
temporally reactivating brain circuits partially affected by a
brain damage.
Visual Agnosia: Overview
Visual agnosia refers to the general impairment of stimulus
recognition in the visual modality in the absence of percep-
tual deficits, memory problems, and general intellectual
impairment. Individuals with visual agnosia demonstrate
normal recognition of objects through modalities other
than vision (touch, audition, and verbal description of
objects function). Agnosia does not necessarily impair the
recognition of all visual stimuli, but can selectively affect
certain categories of percepts (objects, faces, colors, written
words, body parts, environmental scenes), leaving others
intact. Patients are well aware of their predicament. In
1889, Freund, a German neurologist of Breslau, described a
case of visual recognition deficit that he named optic aphasia
and he interpreted as a consequence of the disconnection of
visual from language areas. A year later (1890), Lissauer, a
colleague of Freund, published a paper where he clearly
distinguished two different forms of agnosia: apperceptive
agnosia, or inability to construct a good perceptual represen-
tation from the visual input, and associative agnosia, or
inability to access the stored knowledge related to the per-
cept. Despite these two authors being the first to report
documented cases of agnosia, the term agnosia was coined
by Freud in 1891 to describe object recognition problems in
some individuals. Although it is a relatively rare neurological
symptom, with some 100 cases published between 1890 and
1990, its study has greatly contributed to our understanding
of how the process of visual recognition is organized in the
human brain. As already mentioned, the most common form
of agnosia is agnosia for objects, but forms of agnosia spe-
cific for a category of objects are also possible (e.g., agnosia
for faces (prosopagnosia), agnosia for colors (color agnosia),
agnosia for words (pure alexia or agnosic alexia), agnosia for
scenes). Here, we briefly review these different forms of
agnosia.
Object Agnosia
Although there is no standard taxonomy of object agnosias,
most neuropsychologists agree on Lissauer's original dis-
tinction between apperceptive and associative agnosia,
depending on the processing stage of visual information
affected by the brain lesion. Because this account has
continued to be employed in the neuropsychological litera-
ture to the present day, we will use it as a general framework
for our present purpose. As we will see, however, this classi-
cal dichotomy captures well the cases at one end and the
other of the agnosia spectrum, but not the variants falling
in between.
Apperceptive Agnosia
This kind of agnosia is evident when patients are unable to
recognize objects because they cannot see them properly, in the
absence of elementary visual deficits. This type of agnosia is
thought to arise from a breakdown at relatively early stages of
visual processing, where the elementary features of the stimu-
lus are processed. Object recognition through verbal descrip-
tion by the examiner is preserved, instead. These patients
have relatively good visual fields, visual acuity, brightness dis-
crimination, color vision, depth, and motion perception.
Despite this, shape perception is abnormal in such a way that
patients cannot recognize or copy pictures, letters, and even
simple geometric shapes. Some authors consider a prerequisite
for differentiating these patients from those showing deficits
in the early stages of perception that they are normal on figure-
ground discrimination tests but fail on Efron's test (judge
whether two pairs of simple geometrical figures are same
or different). If tested, identification of object orientation in
space may become pathological, despite the preserved visuo-
motor ability to mimic object orientation based on the same
visual information (this finding has been taken as evidence
60
Agnosia (including Prosopagnosia and Anosognosia) 61
supporting a dual stream model of visual perception, which
attributes to the occipitotemporal ventral pathway ('what'
pathway) object identification and to the occipitofrontal path-
way ('how' pathway) object-directed action control). Recogni-
tion of real objects appears to be better than recognition of
geometric shapes, possibly because of the availability of cues
such as size and surface properties such as color, texture, and
specularity rather than object shape. In most cases of appercep-
tive agnosia, the brain damage is diffuse, often caused by
carbon monoxide poisoning. In the rare cases with circum-
scribed brain lesions, the damage primarily affected the ventral
occipitotemporal cortex bilaterally. According to a widely
accepted interpretation, apperceptive agnosia can be consid-
ered a deficit of shape perception resulting from defective
perceptual grouping of object local features into a global per-
cept. However, clinical findings show that apperceptive agno-
sia covers a wide spectrum of disorders, some of which fall in
between apperceptive and associative agnosia. Hence, two neu-
ropsychologists, Riddoch and Humphreys, proposed to differ-
entiate between distinct subtypes of apperceptive agnosia, each
corresponding to a defective processing stage along the hier-
archically organized stream of visual information processing
leading to conscious object perception (according to Marr's
computational model of vision): Shape agnosia and integrative
agnosia, which are 'closer' to the apperceptive type; transforma-
tional agnosia and agnosia due to impairment of internal object
representation, which are 'closer' to the associative type. Shape
agnosia results from a deficit of the initial processing stage of
visual recognition and consists in the inability to organize the
sensory input into a unified percept: these patients complain of
blurred or unclear vision and are unable to discriminate stim-
ulus boundaries from the background or other contiguous or
overlapping shapes, as well as its orientation and size. Integra-
tive agnosia consists of the inability to integrate single object
features into a global shape, in the presence of the ability
to identify single object details. This deficit is more severe
when object shape is defined by high frequency details
or when overlapping figures must be identified, but is reduced
when silhouettes of objects with reduced internal details are
used for discrimination. Transformational agnosia is a deficit of
perceptual categorization (first described in 1982 by E. War-
rington), which occurs when patients can recognize objects
presented in a canonical view, but fail when they are presented
in noncanonical views. What is loss is the ability to manipulate
the mental representation of the object to match it with its
perceptual image. Agnosia due to impairment of internal represen-
tations of objects occurs when a structural description of the
object is formed normally, but its internal representation
stored in presemantic memory cannot be accessed through a
given route to match on-line descriptions encoded by the
visual system.
Associative Agnosia
This type of agnosia occurs when patients can form a structural
description of the visual object (object copy is preserved), yet
being unable to recognize it. Associative agnosic patients can-
not identify objects even by nonverbal means (e.g., by panto-
miming their use or grouping together dissimilar objects from
the same semantic category); however, recognition is preserved
in the tactile modality (by touching the object) or from a
spoken definition. It is still debated whether cases of relatively
pure associative agnosia are possible, that is, without a con-
comitant apperceptive deficit. Intrahemispheric location of the
lesion is generally occipitotemporal, unilateral (with the prev-
alence of left hemispheric lesions), or bilateral. Following a
left-sided lesion, associative agnosia can be present in both
visual hemifields, probably because the semantic role played
by this hemisphere is greater than that of the right hemisphere.
Associative agnosia has been explained as a deficit of arousal of
the semantic associations related to the visual percept: patients,
despite being able to form a normal visual representation of
the stimulus, are unable to grasp its meaning and therefore
cannot name it. In visual naming tasks, errors tend to be
semantic (e.g., 'knife' for 'fork') or totally unrelated to the
stimulus (e.g., 'horse' for 'chair') or consist of the identification
of the superordinate ('flower' for 'daisy'). Miming the use of a
visually presented object is also impaired, whereas it is not if it
follows a verbal description of the object. Patients also fail in
semantic categorization and association tasks, as well as in the
description of the semantic attributes of an object. According
to one view, associative agnosia is the consequence of a dis-
connection between visual areas and other brain centers
responsible for language or memory. This hypothesis, how-
ever, does not account for the inability of agnosic patients to
convey information nonverbally and to access old knowledge
through vision. A different interpretation, perhaps the most
widely accepted, assumes that stored visual memory represen-
tations have been damaged in such a way that the newly
formed visual percepts cannot be matched against them and
therefore recognition is impossible. Associative agnosia must
be differentiated from optic aphasia and semantic amnesia. In
the former, a visual stimulus is recognized but cannot elicit
naming, whereas in the latter, semantic knowledge and nam-
ing of the stimulus are inaccessible, irrespective of the sensory
modality which conveys the information.
Face Agnosia (Prosopagnosia)
The term prosopagnosia (from the Greek prosopon, meaning face,
and agnosia, meaning ignorance) refers to the inability to recog-
nize familiar faces. The symptom was first reported by the
Italian ophthalmologist Quaglino in 1867, but the term proso-
pagnosia was adopted only in 1 947 by the German neurologist
Joachim Bodamer. The deficit is confined to the identification
of physiognomic traits, as shown by the fact that identification
is preserved through nonphysiognomic cues such as voice, a
particular item of clothing, a scar, and gait. In the most severe
cases, patients cannot recognize their own faces at the mirror or
that of closest relatives. In the milder forms, the deficit may be
limited to friends, acquaintances, and to famous people. Per-
ceptual categorization of the stimuli is preserved (patients
know that a face is a face) as well as the ability to differentiate
faces by sex, race, age, and emotional expressions. What is lost is
the ability to identify a face as a known face. Psychophysical and
neuropsychological studies on face recognition abilities in nor-
mal and brain-damaged patients have revealed that familiar and
unfamiliar face processing follow dedicated routes in the left
and right hemisphere, respectively. Nonrecognition of familiar
62 Agnosia (including Prosopagnosia and Anosognosia)
faces is more likely to occur after a left brain lesion, whereas
nonrecognition of unfamiliar faces is more likely to occur after
a right brain lesion. This distinction led the neuropsychologist
A.L. Benton (1980) to differentiate between two independent
face processing deficits produced by brain damage: apperceptive
prosopagnosia, which refers to a defective perceptual processing
of face information and is brought out by unfamiliar face tasks,
and associative prosopagnosia, which involves an additional
mnestic component and is elicited by familiar face tasks. Recent
fMRI data have located a specific area of the fusiform gyrus
in the ventral part of the inferior temporal cortex called the
fusiform face area (FFA), which is activated more strongly in
response to face stimuli than other stimulus categories. Lesion-
producing prosopagnosia often involve the FFA and are more
frequent in the right side of the brain or are bilateral. The degree
of domain specificity of FFA for face processing is currently
object of debate.
Apperceptive and Associative Prosopagnosia
According to the model of visual recognition proposed by
Bruce and Young (1986), the identification of a face is
the final stage of a sequence of operations made by distinct,
hierarchically organized, information processing modules dis-
tributed along the occipitotemporal ventral pathway of the
brain. At the earlier stages, perceptual face processing results
in the construction of an object-centered, tridimensional struc-
tural description of the face. If this processing level is damaged,
patients are unable to recognize familiar faces ( apperceptive
prosopagnosia) as well as other categories of familiar stimuli.
At later processing stages, the structural description of the face
activates an abstract representation of it stored in recognition
units responsible for the feeling of familiarity. Then, the infor-
mation gains access to the semantic memory sectors ( identity
nodes ) containing biographical, relational, and other informa-
tion about the familiar face. From the identity nodes, informa-
tion finally has access to the modules containing the person's
name. The anatomofunctional independence of the names
module is confirmed by the existence of anomia for proper
names following left brain damage and by the frequent inabil-
ity experienced by normal subjects to retrieve the name of an
otherwise well-known person. Associative prosopagnosia is nor-
mally consequent upon an impairment either at the level of
recognition units or of identity nodes, or both: patients pass
the tests involving face identification, but fail recognition of
familiar faces, though they show good semantic knowledge
of the persons to whom the faces refer. Prosopagnosia must
be distinguished from the impairment of retrieving from
semantic memory any type of information related to an indi-
vidual and/or an entity (loss of semantic knowledge for indi-
vidual entities).
Specificity of the Face Deficit
A recurrent issue of debate in the neuropsychological literature
is whether prosopagnosia is specific to faces or extends to other
categories of objects. This latter view is supported by documen-
ted cases of prosopagnosic patients who also had difficulties in
discriminating among the members of other stimulus cate-
gories, for example, a chair from an armchair, car makes, and
mammals of similar shapes. Moreover, there have been cases
of farmers who had found it difficult to differentiate among
their own cows, and an ornithologist who had lost the ability
to discriminate birds of different species. However, there have
been an equivalent number of reports of prosopagnosics with a
preserved ability of differentiating exemplars of the same cate-
gory when they are not faces. So, the question is whether faces
are special stimuli with special features requiring a dedicated
brain-processing module, or it is the nature of the processing
they undergo which differs from that applied to other classes of
stimuli (e.g., words and objects). Authors supporting this latter
view - for example, Farah - claim that faces require an holistic
processing mode to be identified for which would be responsi-
ble the right hemisphere, whereas words require an analytical
processing mode, for which would be responsible the left
hemisphere, and objects share both processing modalities.
So, depending on the severity of the deficit, the global mode,
the analytical mode, or both can be impaired, resulting in
different patterns of recognition deficit involving mostly faces
(mild deficit of global perception), words (mild deficit of ana-
lytical perception), and objects (more severe deficit of global
and/or local perception). This theory, however, has been
challenged by the reports of patients with agnosia for faces
and words but not for objects, and agnosia for objects without
impairment of either faces or words. Further support of the
theory that faces are a special class of stimuli comes from
neurophysiological recordings in the monkey and fMRI data
in man. The former revealed the existence in the inferior tem-
poral cortex of the monkey brain of neurons selectively acti-
vated by familiar faces or by face details; the latter has shown
the existence in the human fusiform gyrus of the inferior tem-
poral cortex of one area predominantly activated by face stimuli
(fusyform face area (FFA)), and a different area, located in the
parahippocampal region, predominantly activated by scenes
depicting places or buildings, called parahippocampal place
area (PPA), and unresponsive to faces. Lesions to this area can
give rise to a different form of agnosia for landmarks and the
environment called topographic agnosia (see section 'Agnosia for
Landmarks and Environment (Topographic Agnosia)').
Unconscious Face Recognition
This behavior consists in the fact that some prosopagnosic
patients react differently to famous faces that they deny recog-
nizing and to unfamiliar faces, thus showing some degree of
implicit knowledge of the famous faces. Implicit identification
of explicitly nonrecognized familiar faces can be evidenced by
psychophysiological and neurophysiological measures, such as
electrodermal skin conductance, pupillometry, event-related
brain potentials, and by psychological paradigms, such as
forced choice, reaction times, and priming (the improvement
in performance, which occurs in a decision task when a target
stimulus is preceded by a nontarget semantically related to the
target). When implicit recognition is probed by these means,
it is possible to see an increment in these indirect measures of
recognition when familiar faces stimuli are presented relative
to when unfamiliar ones are presented. Perhaps the most
convincing interpretation of this phenomenon is that con-
scious recognition requires a higher activation threshold from
the visual input than implicit recognition. According to this
Agnosia (including Prosopagnosia and Anosognosia) 63
view, if the lesion producing prosopagnosia completely
impairs the function of recognition units, both implicit and
explicit recognition will be impossible. If the impairment is
only partial, instead, the output from recognition units will be
sufficient for unconscious recognition but insufficient for overt
recognition.
Color Agnosia
Color agnosia describes a deficit in color naming and/or color-
object associations that cannot be accounted for by aphasia
or dyschromatopsia (impaired color discrimination). Such
patients perceive colors correctly, but are unable to name
them, to retrieve their name when given the name of an object
of a given color, to match a color with an object, or to sort
colors according to their hue. Some patients are agnosics for
colors only when they have to make a visuoverbal match
(naming a visually presented color and pointing to a color
named by the examiner); others are agnosics only when are
asked to retrieve colors from semantic memory (e.g., what
color is a banana?, coloring in a drawing of an apple). This
dissociation has been related to two different forms of color
agnosia: color anomia or aphasia and color amnesia, respectively.
Color anomia is considered a visuoverbal disconnection deficit.
Patients are unable to match verbal with visual color informa-
tion, but are capable to retrieve color information from seman-
tic memory. The interpretation of this behavior is framed in
anatomical terms: patients have a left occipital lesion which
causes right hemianopia and, hence, confines accessible visual
information in the left hemifield/right visual areas. Color nam-
ing requires callosal transfer of information to the left hemi-
sphere, which is interrupted by the lesion of the fibers
connecting the visual areas of the right hemisphere with the
language areas of the left hemisphere where the names of
colors are stored. Color amnesia is shown by patients who not
only are unable to name colors, but are also unable to retrieve
color information from semantic memory (e.g., when they
have to name the color of an object from memory (what
color is a panther?) or color in the drawing of the same object).
The disconnection hypothesis cannot account for this behav-
ior, which is more correctly described as a defective access to, or
degradation of, the semantic memory sector where associa-
tions between objects and colors are formed. The brain damage
responsible for color amnesia is in the territory of the tempor-
ooccipital branch of the left posterior cerebral artery, or is
bilateral.
Tactile Agnosia
Tactile agnosia is the inability to recognize objects through
palpation in the absence of elementary sensory deficits such
as ahylognosia (inability to identify substance features) and
amorphognosia (failure to recognize shape haptically). Recog-
nition through the visual modality is preserved. The locus of
the lesion involves the posterior-inferior portion of the parie-
tal lobe and can be unilateral or bilateral. When the lesion
involves the right hemisphere, only the contralateral hand is
agnosic, whereas both hands can be affected when the lesion
involves the left hemisphere. This latter condition parallels that
described in the visual modality for associative agnosia follow-
ing a left hemispheric lesion, and can be ascribed to the appar-
ently dominant semantic role of the left hemisphere. The
prevalence of tactile agnosia is much lower than the prevalence
of visual agnosia, but the difference may be more apparent
than real because of the features of cortical vascularization of
the parietal lobe. The available evidence does not permit to
differentiate between apperceptive and associative forms of
tactile agnosia. The deficit is interpreted as a failure of the
associative-semantic system to match the tactile features iden-
tifying an object with its meaning. Tactile agnosia must be
differentiated from tactile aphasia, which is shown by patients
who cannot name objects they are handling but can show that
they have recognized them either by miming their use or by
passing association or categorization tests. Tactile aphasia can
be considered a disconnection deficit between tactile informa-
tion about an object and its name, and is observed in patients
with callosal resection and in those with damage to the left
parietal lobe.
Agnosia for Words
Agnosia for words is also known as pure alexia, perceptual
word-form agnosia or pure word-form blindness and is usually
discussed in the context of language impairments. However,
given that these patients present a language impairment lim-
ited to visually presented words (e.g., reading), but not to
auditory presented stimuli, it can also be treated as an agnosic
symptom. These patients may be able to spell words out loud
and to write well, but fail in reading their own handwriting of
the same words. Their reading is typically extremely slow and
characterized by the processing of one letter at a time ('letter by
letter reading'). This deficit has also been called ventral simul-
tanagnosia, to indicate the failure in reading multiple letters
simultaneously. It remains controversial whether this form of
agnosia is limited to words or extends to other classes of visual
objects. The lesion site in these patients is typically in the
ventral occipitotemporal cortex.
Agnosia for Landmarks and Environment
(Topographic Agnosia)
This type of agnosia refers to the inability to recognize familiar
landmarks, scenes, and/or buildings, and can be differentiated
from other disorders affecting spatial orientation. These
patients, unsurprisingly, get easily lost and are unable to learn
new routes, despite being able to read and draw maps, describe
familiar routes, and provide directions. The lesions giving rise
to topographical agnosia usually follow bilateral or right poste-
rior artery infarction and may implicate the parahippocampal
place area in the parahippocampal gyrus.
Anosognosia
'Is there anything wrong with you?' 'No, I am fine.' 'So why are
you in the hospital?' 'I felt unwell.' 'Is there anything wrong
64 Agnosia (including Prosopagnosia and Anosognosia)
with your left arm or leg?' 'No, they are all right.' This might
seem a normal conversation between a doctor and a patient,
but it has, in fact, an extraordinary feature which affects the
significance of the patient's answer: despite her claim, the
patient was completely plegic on one side of the body. A stroke
caused a permanent damage to the right side of the brain and
the motor abilities reliant on the integrity of that hemisphere
were impaired. As a consequence, the patient was not able to
move the part of the body controlateral to the affected side of
the brain. The striking thing was that the patient claimed that
there was nothing wrong with her, and if she was asked more
stringent questions about the affected limbs, she firmly denied
her motor problem. This remarkable, well-known phenome-
non, first described in scientific papers by Austrian, German,
and Swiss neurologists at the turn of the nineteenth century,
was named anosognosia in 1914 by the French neurologist
Joseph Babinski. The term, derived from Greek, means 'lack of
knowledge for the illness,' and since then it is used to indicate
the behavior of patients who firmly deny their pathological
conditions. Anosognosia can be observed in association with
different brain diseases, ranging from denial of complex mental
and neurodegenerative disorders, such as schizophrenia and
Alzheimer disease, to the denial of more focal neurological
and neuropsychological deficits such as motor, sensory, and
cognitive disorders. Interestingly, patients affected by anosog-
nosia can show different degrees of knowledge of their neuro-
logical deficits, ranging from a resolute denial of the disease,
even when overtly faced with the negative consequences of their
symptoms, to attenuated forms of awareness where the
patients' behavior suggests that some 'dim knowledge' of their
status is present but not directly available for explicit report.
Another important aspect of anosognosia, both theoretically
and clinically, is its specificity. When a patient, following a
brain damage, has many co-occurrent neurological impair-
ments, the lack of knowledge is, in most cases, segregated to
only one pathological domain. For instance, if a right hemi-
sphere stroke has caused left hemiplegia, left hemianesthesia,
and cognitive impairments, the lack of knowledge is often
related to only one of these symptoms, with the patients deny-
ing a specific impairment, but admitting the others.
Selective-Specific Form of Anosognosia: The Case of
Anosognosia for Hemiplegia
One of the most studied forms of impaired knowledge for the
disease is anosognosia for hemiplegia (AHP) where, as indi-
cated earlier, patients deny their contralesional motor pro-
blems. This disturbance is frequent after damage to the right
hemisphere, with a prevalence ranging from 20% to 50% of
hemiplegic patients, depending on the studies, the differences
being related to the time of evaluation (acute vs. chronic phase
of the illness) and selection criteria. We take this form of
anosognosia as paradigmatic to indicate the selective and spe-
cific aspect of denial of illness. Other forms of anosognosia will
be briefly discussed after the description of AF1P. AHP patients,
if inquired about their potential capability of performing
actions either with the right or with the left hand, or even
bimanual actions, claim that they can perform any kind of
movement equally well. Patients' false belief of still being
able to move remains unchanged even when, having been
requested to actually perform different kinds of actions, sen-
sory and visual evidence from the affected motionless side
should suggest that no movement has been performed. To
better grasp the manifold aspects of AHP, a collage of a conver-
sation with a patient showing an intractable denial is reported
subsequently. The patient typically suffered from a right hemi-
sphere stroke and, as a consequence, she was left with contrale-
sional complete hemiplegia. The patient was evaluated
approximately 1 month after the brain infarct.
E: Where are we?
P: In the Hospital
E: Which hospital?
The patient answered correctly.
E: Why are you in the hospital?
P: Because I had a stroke.
E: What is a stroke?
P: I do not know.
E: How is your left arm?
P: Fine.
E: Can you move it?
P: Yes.
E: Would you be able to raise your left arm up in the air?
P: Yes, I would.
E: Would you be able to lift the telephone receiver with your left
hand?
P: Yes, I would.
E: Would you be able to open a bottle, using both hands?
P: Yes, I would.
E: Would you be able to brush your hair handling the hairbrush
with your left hand?
P: Yes, I would.
E: Would you be able to wash your face using both your hands?
P: Yes, I would.
(The patient was then asked to actually perform some
movements.)
E: Could you touch my hand with your right hand?
P: (The patient does it without any problem.)
E: Could you touch my hand with your left hand?
P: (Although she seems to try the movements, she cannot raise
the arm and reach the examiner's hand. Nonetheless, after a while
she says 'Done.')
E: Have you done it?
P: Yes, I think so.
E: Could you touch your left hand with your right hand?
P: (She does it without any problem, thus showing that she
knows what and where is the left plegic limb).
E: Could you open this bottle for me, please?
P: (The patient attempts to do it using only the right hand.)
E: Can you manage?
P: No.
E: Why?
P: Because it does not open.
E: How do you open a bottle?
P: With one hand I hold the bottle, with the other I unscrew the
cap.
E: Are you doing it?
P: Yes.
E: Can you put your left hand on your left shoulder?
P: Yes, I can.
E: Then please do it.
P: (the patient seems to try the movement: she also overtly looks
at the left motionless arm and at the shoulder. After that, she looks at
the examiner as if she had finished performing the requested
action).
E: Have you done it?
P: Yes, I think so.
(The patient, on the wheelchair, is taken to the bathroom and
placed in front of the basin.)
E: Could you wash your face using both hands, please?
Agnosia (including Prosopagnosia and Anosognosia) 65
P: (The patient takes the bottle of the liquid soap with the right
hand and attempts to soap her left hand as if the left hand was
actually over the basin, near the midline. However, the left hand was
not there because it was lying on her lap. After having soaped the
'ghostly' hand, she started to move the right arm/hand forward and
backward as if she was washing the two hands, one against the other.
Finally, she washed the face using the right hand.)
E: Are you washing both your hands?
P: Yes, I am.
E: Are you washing your face?
P: Yes, I am.
E: With both hands?
P: Yes, I think so.
(Then the patient was asked to brush her hair handling the
hairbrush with the left hand. The hairbrush was on the table and
the left hand was lying motionless on the table as well. She took the
hairbrush with the right hand and forced it below the left hand.
While 'holding' the hairbrush with the left hand, she moved the
head as if she was actually brushing the hair. After a while, she
looked at the examiner and seemed satisfied with her performance.)
E: Have you done it?
P: Yes, I have, but only on the left side (!).
Possible Interpretation
Although anosognosia was first reported by neurologists who
tried to interpret the impairment as a direct consequence of
the brain damage, in the second half of the last century, one
of the most accepted interpretations of anosognosia was based
on a motivational account according to which denial was
considered a defensive mechanism against the stress caused
by the illness. This caused a lack of studies investigating the
cognitive process affected in AHP, because the motivational
interpretation considered the disorder not directly due to the
brain lesion, and therefore as the direct effect of a damage to a
specific cognitive system, but as the consequence of the activa-
tion of a normal psychodynamic reaction. At the end of the last
century, however, this interpretation was questioned on the
basis of many clinical observations which were not predicted
by a pure motivational/psychodynamic account. As already
mentioned, anosognosia is often selective in patients with
multiple pathologies who may deny one deficit, but spontane-
ously report another. Also, the time course of AHP is at odds
with the idea of a psychodynamic reaction. Indeed, anosogno-
sia tends to ameliorate with time, and it is observed much
more frequently in the first period after the stroke than during
the chronic phase. A defense mechanism would, again, predict
the opposite temporal pattern, insofar as time is needed for
establishing a psychodynamic reaction. Interestingly, during
the amytal testing (a diagnostic tool that, through a pharmaco-
logic intervention - the injection in the brain circulation of a
barbituric - produces a temporary functional suppression of
the activity of one hemisphere, also causing a temporary hemi-
plegia contralateral to the injected hemisphere), anosognosia
appears immediately after the amytal injection and mainly
when it is the right hemisphere that is knocked down. Another
striking, opposite effect of a physiological manipulation is seen
when the injection of cold water into the left ear of the AHP
patient provokes an instantaneous, although temporary,
remission of the denial. This effect, triggered by a vestibular
reflex, lasts a few minutes during which a left side nystagmus
(left side reflexed ocular movement) is observed. In such cases,
the patients astonishingly admit their paralysis but again 'for-
get' it when the effect of the stimulation is over. Again, a
psychodynamic reaction should not be influenced by a physi-
ological manipulation. These and many other clinical and
experimental data suggested that AHP might be better
explained within a neuropsychological framework that would
consider the denial not as a normal functional reaction trig-
gered by the emotional strain, but as a specific cognitive deficit
caused by the brain damage. New approaches to the interpre-
tation of AHP explain the denial behavior by referring to a
cognitive model of motor production and motor control. In
these models, a comparator system has to match the congruity
between the intended movement and the sensory conse-
quences of the actually executed movement. When an intended
movement is not performed, the comparator should detect the
mismatch between the movement/no-movement conditions.
Within this context, two different proposals have been put
forward. According to some views, the comparator does not
work because the patient, as a consequence of the brain dam-
age, has lost the intention to move and therefore cannot dis-
cover that he/she is plegic. On the basis of anatomoclinical
correlations, other views consider anosognosia as a failure of
the comparator that, directly damaged by the brain lesion, does
not detect the mismatch between a desired action and the
actual status of the sensorimotor system (leading to altered
motor awareness), in the face of an intact capacity of program-
ming movements and forming sensorimotor predictions. This
hypothesis would imply that the brain activity leading to the
construction of a conscious intention of action is normal.
Moreover, it can also explain the neural base of the 'false belief
of being still able to move. In this view, the false belief would
not be a mere confabulation but would express the patient's
feeling of action related to the 'normal' or quasinormal activa-
tion of the intention programming system. Although more
studies are needed for verifying the 'motor' hypothesis of
AHP, this approach to the study and interpretation of AHP
was very innovative because, for the first time, anosognosia
was not considered as a secondary disorder due to the presence
of other neurological/neuropsychological symptoms, but
instead a specific disorder of motor consciousness. The study
of the anatomical localization of the lesion has shown that
patients with AHP, compared to patients with hemiplegia with-
out anosognosia, show damage in frontal premotor areas, sug-
gesting that motor production and motor control share
common anatomical substrates. These findings are directly rel-
evant for models of motor control and, more generally, for
accounts of consciousness. Indeed, the involvement of premo-
tor areas in self-monitoring of action implies that, at least for
motor functions, monitoring is neither the prerogative of some
kind of central executive system, hierarchically superimposed
on sensory-motor and cognitive functions, nor a function that
is physiologically and anatomically separated from the primary
process that has to be monitored. Instead, the anatomical cor-
relates of anosognosia show that monitoring can be implemen-
ted in the same neural network responsible for the process that
has to be controlled. It is worth noting that recently the idea that
AHP might be somehow related to a motivational reaction has
been again proposed on the evidence that anosognosic patients
may show some sort of implicit knowledge of being plegic. This
might imply that the anosognosic behavior is the result of a
66 Agnosia (including Prosopagnosia and Anosognosia)
repression mechanism. However, although the demonstration
of an implicit knowledge is necessary for admitting a psycho-
dynamic account of the disorders, its presence is not sufficient
for establishing a direct causal relation. For instance, it is plau-
sible that the impaired operation is not completely abolished
but instead damaged to a degree that allows an implicit but not
an explicit processing of the information that therefore would
not reach a sufficient strength for an overt recognition of the
disease. Although a motivated defense reaction may contribute
to some cases of denial of illness and may shape the severity and
the way in which anosognosia manifests itself, more research is
needed to better understand the involvement of defense
mechanisms in AHP and the possible interaction between
motivational repression and conscious control of action.
Other Specific Forms of Anosognosia
As already pointed out, selective anosognosia can be observed
for other neurological/ neuropsychological conditions that
further suggest its specific and selective nature. Interestingly,
many reports of selective forms of anosognosia also point to
the possibility that the disturbance can be considered a modu-
lar deficit of a monitoring/controlling system. Constructional
apraxia (CA), for example, is a disorder often associated with
right brain damage in which drawing, copying, and block
construction are severely impaired. This disorder can be con-
ceived as the motor expression of a more pervasive disorder in
combinatory and organizing activity, and it has recently been
associated with damage to the dorsolateral prefrontal cortex,
extending to subcortical structures like the putamen, the
insula, and the caudate nucleus. Interestingly, when the
patients are profoundly anosognosic (i.e., they are totally satis-
fied with their abnormal constructional performance), lesion
in Area 44 is additionally found. Area 44 is one of the premotor
areas which, as reported previously, is also affected in hemiple-
gic anosognosic patients. The anatomoclinical correlation
between anosognosia for CA and premotor lesion has been
taken as further evidence that in the motor domain, patients'
unawareness for their motor impasse can be the consequence of
inefficiency in comparing intended and anticipated action to
real action execution due to a lesion of a motor-monitoring
system implemented in the same area of motor programming.
Conceptually analogous findings have been reported for
unawareness of a different sensory disturbance. Cerebral acro-
matopsia is a rare disorder of color vision caused by bilateral
damage to the occipitotemporal cortex. Patients affected by this
disorder see the world in black and white and are usually aware
of their visual deficit. However, a fascinating single case of
acromatopsia showed that patients can be anosognosic even
for this very specific form of visual impairment; despite verbal
and nonverbal/perceptual testing clearly demonstrating the
presence of severe color blindness, the patient claimed that
his color vision was completely normal, denying any problems
even when faced with his errors in naming colors. Moreover,
color vision improved with time and a parallel improvement in
awareness was observed. The simultaneous occurrence of acro-
matopsia and anosognosia, their parallel recovery, and the
lesion site in visual areas suggest that both deficits were due to
dysfunction of the same brain region, implying that normal
perception and normal monitoring share common anatomical
substrates. In the sensory domain, anosognosia has been
reported also for tactile anesthesia, and even in this domain,
the unawareness of the disorder seems to be related to lesion
intrinsic to somatosensory areas. The notion that the conscious
perception of the presence or absence of a sensorimotor experi-
ence depends on the same system involved in the processing of
the primary function giving rise to that experience, can be seen
as evidence in favor of a modular structure of consciousness.
Anosognosia for More Generalized Cognitive Impairment
As mentioned earlier, unawareness of dysfunctions can be
observed for more complex neurocognitive disorders such as
those characterizing Alzheimer deficits or schizophrenic symp-
toms. In these multicomponential pathologies, is very hard to
individuate a common core explaining the different denial
behaviors. The fact that in many instances there is a severe
intellectual impairment is of course a confounding factor.
However, a shared feature in Anosognosia for Alzhiemer (AA)
and Anosognosia for Schizophrenic disorder (AS) is the dam-
age to frontal lobe circuits. In AA, a significant association with
brain areas, which are considered to be responsible for tracking
the physiological state of the body (like the right hemisphere
ventromedial prefrontal cortex) has been found. This would be
generally consistent with the role of medial and orbital frontal
structures in signaling the deviations from expected outcomes.
In the case of AA, the deviation to be detected would be the
divergence between one's own actual versus expected perfor-
mance. This interpretation of AA prompts the idea of a com-
parator system whose dysfunction generates different degrees
of unawareness for the patient's deficit. Therefore, similar to
other more specific forms of anosognosia, AA might seem to be
the result of the incapability of detecting a mismatch between
a desired state and the actual status of the system. Other inter-
pretations of AA point to the possibility that memory or execu-
tive dysfunctions, typical of this disease, may affect the
immediate ability to update and judge cognitive performance
in a domain-specific manner. Also, this interpretation seems to
explain the possible different forms of anosognosia observed
in Alzheimer disease as a result of an intramodular impairment.
As for schizophrenia, the presence of anosognosia is fre-
quently associated with bilateral volume reduction in different
areas of the frontal lobes. Awareness in these patients can be
fractionated in awareness of illness and awareness of cognitive
impairments and functioning, and some studies have shown
that awareness of schizophrenic symptoms can be relatively
poor while insight into cognitive deficits can be better or
even preserved. It is interesting to note that memory function-
ing is a strong predictor of awareness of memory deficits in
schizophrenic patients and that worse memory is associated
with overestimation of function. This seems to be again con-
sistent with the neuropsychological findings such as those
discussed earlier that cortical systems underpinning awareness
of a deficit are related and anatomically close to those under-
pinning the function whose deficit is studied. Whatever expla-
nation will turn out to be true for explaining denial of illness,
we would like to conclude this section with a quotation from
George Prigatano; "Understanding the biological and neuro-
psychological mechanisms responsible for anosognosia in
its various forms may reveal important insights into brain
Agnosia (including Prosopagnosia and Anosognosia) 67
organization and how human consciousness (subjective aware-
ness of the self and the environment) is possible. The compara-
tive study of anosognosia in patients with identifiable brain
disorders in comparison to patients with psychiatric disorders
may provide rich insight into the body-mind problem."
See also: Aphasia; Sleep, Biological Rhythms, and Performance.
Further Reading
Behrmann M and Nishimura M (201 0) Agnosias. WIREs Cognitive Science 1 : 203-21 3.
Berti A, Bottini G, Gandola M, et al. (2005) Shared cortical anatomy for motor awareness
and motor control. Science 309: 488-491.
Berti A, Spinazzola L, Pia L, and Rabuffetti M (2007) Motor awareness and motor
intention in anosognosia for hemiplegia. In: Haggard P, Rossetti Y, and Kawato M
(eds.) Sensorimotor Foundation of Higher Cognition. Attention & Performance,
vol. XXII, pp. 163-181. Oxford: Oxford University Press.
De Renzi E (1999) Agnosia. In: Denes G and Pizzamiglio L (eds.) Handbook of Clinical
and Experimental Neuropsychology, pp. 371-407. UK: Psychology Press.
Farah MJ (1990) Visual Agnosia. Disorders of Object Recognition and What They Tell
Us about Normal Vision. Boston, MA: MIT Press.
Gilleen J, Greenwood K, and David AS (2010) Domains of awareness in schizophrenia.
Schizophrenia Bulletin .
Goodale MA and Milner DA (1992) Separate visual pathways for perception and action.
Trends in Neurosciences 15(1): 20-25.
Prigatano George P (ed.) (2010) The Study of Anosognosia. Oxford: Oxford
University Press.
Ramachandran VS (1996) The evolutionary biology of self-deception, laughter,
dreaming, and depression: Some clues from anosognosia. Medical Hypotheses
47(5): 347-362.
Riddoch MJ and Humphreys GW (eds.) (1987) Visual Object Processing: A Cognitive
Neuropsychological Approach. Hove, UK: Laurence Erlbaum.
Rinaldi MC, Piras F, and Pizzamiglio L (2010) Lack of awareness for spatial and verbal
constructive apraxia. Neuropsychologia 48(6): 1574-1582.
Rosen HJ, Alcantar 0, Rothlind J, et al. (2010) Neuroanatomical correlates of
cognitive self-appraisal in neurodegenerative disease. Neuroimage
49(4): 3358-3364.
Spinazzola L, Pia L, Folegatti A, Marchetti C, and Berti A (2008) Modular structure of
awareness for sensorimotor disorders: Evidence from anosognosia for hemiplegia
and anosognosia for hemianaesthesia. Neuropsychologia 46: 915-926.
Starkestein SE and Power BD (2010) Anosognosia in Alzheimer disease: Neuroimaging
correlates. In: Prigatano GP (ed.) The Study of Anosognosia. Oxford: Oxford
University Press.
von Arx SW, Muri RM, Heinemann D, Hess CW, and Nyffeler T (2010) Anosognosia for
cerebral achromatopsia - A longitudinal case study. Neuropsychologia 48(4):
970-977.
Warrington EK (W5) Agnosia: The Impairment of Object Recognition. Amsterdam:
Elsevier Science.
Agoraphobia and Panic Disorder
G L Thorpe, S T Sigmon, and K L Yoon, University of Maine, Orono, ME, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cognitive therapy A system of psychotherapy focused upon
identifying and restructuring dysfunctional thoughts and
schemas linked to psychopathology.
Exposure in vivo The structured treatment of anxiety
disorders by systematic confrontation of feared external
situations to reduce avoidance behavior and anxiety.
Exposure to somatic cues Extends the methods of exposure
in vivo to those internal cues and bodily sensations
associated with panic attacks.
Limited symptom attack An anxiety episode with a few
subjective anxiety symptoms, insufficient in number to
qualify as a panic attack.
Panic attack A discrete period of intense fear, not explained
by a continuing organic factor, that arises rapidly with at
least 4 anxiety symptoms from a 13-item list specified in the
DSM-IV-TR.
Agoraphobia and panic disorder are anxiety disorders that
are formally classified into three overlapping syndromes in
the current psychiatric taxonomy: panic disorder without ago-
raphobia, panic disorder with agoraphobia, and agoraphobia
without history of panic disorder.
Agoraphobia is characterized by marked fear of entering
crowded, public places; of traveling away from home, especially
by public transportation; of feeling trapped or confined; and of
being separated from a place or person associated with safety.
Sudden, brief episodes of extreme anxiety - panic attacks - are
commonly associated with agoraphobia and may lead to avoid-
ance of situations in which they occur. Often there is a 'fear of
fear' pattern, in which the bodily sensations of mounting panic
are themselves a source of anxiety. Generally more debilitating
than specific or social phobias, agoraphobia causes some people
to remain entirely housebound. As a syndrome of anxiety ele-
ments in physiological, behavioral, and subjective domains,
agoraphobia represents a distinct disorder with a typical clinical
presentation and course.
Panic disorder has been formally recognized since 1980 as
an anxiety syndrome marked either by repeated, unexpected
panic attacks that are associated with distress and behavioral
impairment or by significant apprehension about having
future panic attacks.
Since the 1970s, clinical researchers have developed effective
pharmacological and psychological treatments to reduce or
eliminate agoraphobic avoidance behavior and panic attacks.
Agoraphobia and Panic: Past and Present
The term agoraphobia was introduced by the German psychia-
trist C.F.O. Westphal (1822-1890) in a classic monograph
of 1871, Die Agoraphobie. The term derives from the Greek
agora, referring to a marketplace or place of assembly. Westphal
described the abnormal fears of a series of three men who
experienced anxiety episodes when walking alone in public
places. Feared situations included city squares, conceit halls,
churches, open streets and fields, crowded rooms, and traveling
by carriage, bus, or train; typical anxiety symptoms were
trembling, heart palpitations, and 'an immediate breakout of
intense anxiety.'
The development of behavior therapy in the 1950s by
Joseph Wolpe and others was closely connected with the
study of phobias and other anxiety disorders. Interest in
agoraphobia revived with American and British research on
systematic desensitization and related methods in the 1960s,
and with the publication of Isaac Marks' Fears and Phobias
in 1969. Systematic desensitization produced disappointing
outcomes with agoraphobia, but treatment based on exposure
to relevant situations was successful in reducing avoidance
behavior and anticipatory anxiety.
Progress in psychological and pharmacological treatment
of agoraphobia in the 1970s influenced the diagnostic classifi-
cation itself in the United States, so that in 1980, agoraphobia
appeared for the first time as a distinct category. Further devel-
opments gave prominence to the panic attack as the central
feature of agoraphobia in panic disorder with agoraphobia;
panic disorder without agoraphobia was the parallel syndrome
not marked by phobic avoidance of situations.
In recent years, psychological treatment of the agoraphobia
and panic disorder syndromes has focused on therapeutic
exposure to panic sensations, on encouraging patients to make
more realistic and benign ascriptions as to the source of their
anxiety, and on helping them to question their essential prem-
ise that thoughts or concerns about emotional distress must
necessarily engage one's attention and activate affective arousal.
Exposure to somatic cues, breathing retraining, systematic con-
frontation of feared external situations, and cognitive therapy
have become the leading psychological interventions.
Descriptive Psychopathology, Diagnosis, and
Epidemiology
Description of Agoraphobia
People with agoraphobia usually fear, and often avoid, situa-
tions in which it would be difficult or embarrassing to obtain
help if overwhelmed by anxiety. Such situations include
(1) traveling away from home, especially by bus, train, or car;
68
Agoraphobia and Panic Disorder 69
(2) crowded, public places, such as government buildings,
supermarkets, concert halls, shopping malls, and places of
worship; and (3) confined places, such as elevators, the den-
tist's or beautician's chair, and - when driving - passing
through tunnels, over bridges, or along a limited-access high-
way. Agoraphobia is commonly associated with panic attacks
that often arise in the situations typically feared and avoided.
Some people with agoraphobia restrict their lives substantially,
sometimes to the point of remaining housebound, to avoid the
anxiety or panic aroused by entering public places.
Dysphoric mood, somatoform disorders, interpersonal
conflict, or substance abuse may accompany agoraphobia.
Untreated, agoraphobia tends to follow a chronic, fluctuating
course. It is common for people with agoraphobia to experi-
ence daily variations in anxiety severity. For some, there may be
weeks or months of near-normal functioning followed by a
resurgence of the original symptoms. For others, gradual
improvement leading to complete recovery may occur without
professional intervention, but this is not typical.
Description of Panic Attacks
Panic attacks are recurrent, distinct episodes of extreme anxiety
or distress, not explained by the presence of a continuing
organic factor. Panic attacks include at least 4 of a 13-item list
of typical anxiety symptoms, which are initially unexpected
and not produced in response to stimuli associated with spe-
cific or social phobias. (Panic attacks are categorized as unex-
pected or uncued, situationally bound or cued, or situationally
predisposed; only unexpected panic attacks are used in the
diagnosis of panic disorder.) Because individuals who have
unexpected panic attacks may also report situational attacks,
clinicians are careful to discern the focus of anxiety - the panic
attack itself or concerns about becoming embarrassed.
The list of typical symptoms in a panic attack includes
shortness of breath, dizziness, heart palpitations or rapid
heart rate, trembling or shaking, sweating, the sensation of
choking, depersonalization or derealization, and fear of dying,
losing control, or developing an acute mental illness. The anxi-
ety symptoms in a panic attack arise suddenly and rapidly
increase in intensity. An organic factor may have been influen-
tial in early panic attacks (e.g., the patient may have experienced
dizziness as a result of a viral infection of the vestibular system,
or depersonalization following ingestion of an illicit drug) but,
by definition, the attacks will have continued despite successful
treatment or removal of the initiating organic factor.
Diagnostic Classification: Historical Context
Before 1980, neither agoraphobia nor panic disorder was listed
as a distinct disorder in the DSM classification. Agoraphobia
could be found among lists of the Greek names for specific
phobias in textbooks on psychiatry and abnormal psychol-
ogy. It is now clear that agoraphobia is in no sense a specific
phobia - its prevalence, its resistance to treatment by system-
atic desensitization, its distressing and disabling consequences,
and the broad range and severity of its symptoms - most often
including panic attacks - all clearly set it apart from such focal
fears as phobias of heights, snakes, blood, or the number 13.
Patients with agoraphobia show various specific patterns of
avoidance, and there is no standard list of situations that must
be feared for the diagnostic criteria to be met. The current view
is that what is chiefly feared in agoraphobia is the absence of
safety signals, not the presence of disturbing objects. Most
recently, the DSM-IV-TR has cited as a central feature of panic
disorder with agoraphobia fear of situations from which it
would be difficult to escape, or in which help may be unavail-
able, in the event of a panic attack, leading to avoidance or
marked distress. Alternatively, in agoraphobia without history
of panic disorder, patients fear limited symptom attacks or
circumscribed anxiety episodes.
Anxiety restricted to social situations in which the person
fears embarrassment or humiliation under the evaluative scru-
tiny of others is better construed as social phobia.
The DSM-IV-TR Classification
Agoraphobia appears twice in the DSM-IV-TR, as panic disor-
der with agoraphobia and as agoraphobia without history
of panic disorder. Similarly, panic disorder appears twice, the
second panic syndrome being panic disorder without agora-
phobia. All three syndromes are listed as anxiety disorders. The
DSM-IV-TR lists separate criteria sets, not diagnostic categories
in themselves, for panic attack and for agoraphobia.
Panic disorder with agoraphobia
The essential elements of this diagnosis are the presence of
panic attacks and agoraphobia, as defined in the criteria sets.
Agoraphobia without history of panic disorder
A person with this disorder has never had problems that meet
criteria for panic disorder. Instead, he or she fears, and may
avoid, situations in which it would be difficult or embarrassing
to leave in the event of the sudden onset of anxiety, which may
represent a 'limited symptom attack' that would not include
the range of symptoms associated with a panic attack.
Panic disorder without agoraphobia
Patterns meeting criteria for panic disorder but not agorapho-
bia are classified as panic disorder without agoraphobia.
Epidemiology
Appropriate methodology requires assessing the prevalence
and correlates of the syndromes in the general community as
well as in clinic samples (which tend to be unrepresentative).
Because of changes in the taxonomy, allowance has to be made
for the different terms and criteria in studies conducted in
different decades. The most informative studies use accurate
community survey techniques.
In community studies, about half of the respondents
with agoraphobia are also classified as having panic disorder.
In clinical samples of agoraphobia, as many as 95% have the
combined syndrome panic disorder with agoraphobia. In the
Epidemiological Catchment Area study reported in the 1980s,
the ratio of women to men with agoraphobia was 2.7:1.
Estimates of the prevalence of agoraphobia have been con-
sistent across countries and cultures in studies using the same
instrument and careful sampling procedures, but more recent
70 Agoraphobia and Panic Disorder
studies give lower prevalence estimates. The lifetime prevalence
of agoraphobia, with or without panic, was cited as about 5%
in the late 1980s and as about 2.5% in the 1990s.
Agoraphobia is associated with more severe impairment
than other phobias and is strongly associated with dysphoric
mood. Comorbidity rates for major depressive disorder in
patients with a panic disorder syndrome range from 10%
to 65%, and in about a third of the cases, depression precedes
the development of panic attacks. In the other two-thirds of
cases, major depressive disorder occurs concurrently with or
subsequent to the development of panic disorder.
In the agoraphobia syndromes substance abuse, hypochon-
driasis, somatization disorder, and personality disorders are
often present. The usual course is chronic. The age at onset in
agoraphobia varies but is usually in the 20s or 30s with a mean
of about 28 years. There is no general agreement on an associ-
ation between agoraphobia and specific childhood experi-
ences. Maternal overprotection has been studied, but findings
are mixed.
Lifetime prevalence estimates for panic disorder increased
from 3.5% in the mid-1990s to about 4.7% in 2005. The rate
for panic disorder without agoraphobia is cited in the National
Comorbidity Survey Replication as 3.7%, and for panic dis-
order with agoraphobia as 1.1%. The lifetime prevalence rates
for isolated panic attacks without agoraphobia are much
higher at 28%.
The typical age of patients at the onset of panic disorder is
24 years. Women are 3 times more likely to be diagnosed
with panic disorder with agoraphobia than men, and twice
as likely with panic disorder without agoraphobia. Although
few theories or models address gender issues, recent research
has focused on gender-specific stressors (e.g., premenstrual
distress, sexist events) for women that may account for more
severe symptoms.
The estimated morbidity risk of anxiety disorders in the
first-degree relatives of patients with agoraphobia is 32%. Con-
cordance rates for either of the panic syndromes are significantly
higher in monozygotic than in dizygotic twins; a Norwegian
study showed 31% concordance in 32 monozygotic twins
but 0% in 53 dizygotic twins. Such results have been taken to
indicate some genetic predisposition for agoraphobia and panic
disorder.
Theories of Etiology and Maintenance
Biological Theories
The observations that anxiety syndromes run in families and
that pharmacological treatment can be helpful have led to con-
siderable interest in possible biological mechanisms - though it
has to be noted that successful pharmacotherapy would not
logically confirm a biological etiology. Attention has been paid
to the heritability of agoraphobia and to possible physical
vulnerability factors.
A predisposition toward agoraphobia and panic may be
inherited, but it is not possible to predict who will develop a
disorder even among people with a number of close relatives
with agoraphobia or panic syndromes. Those syndromes prob-
ably conform to a diathesis-stress model in which an inherited
vulnerability is necessary, but not sufficient, for the eventual
appearance of a disorder, which would require the additional
operation of certain environmental factors.
Physiological variables distinguishing agoraphobia from
normal functioning, and from less pervasive anxiety disorders
like specific phobia, include resting heart rate and forearm
blood flow (both higher in agoraphobia) and skin conduc-
tance (higher and more variable in agoraphobia). However,
such findings have not produced clear conclusions with impli-
cations for etiology or treatment.
Inherited vulnerability factors include personality traits
such as neuroticism, which is thought to result from lability of
the limbic system, of the autonomic nervous system, or of
specific neurotransmitter processes. One study showed that rats
bred for emotionality had more brain benzodiazepine receptors
than rats bred normally. Malcolm Lader noted that many of the
data on panic may be explained by an instability or hypersensi-
tivity of central noradrenergic mechanisms centering on locus
coeruleus function. Yet, few definite conclusions may be drawn
from the many physiological and endocrine studies. The best-
supported generalization is that patients with agoraphobia and
related anxiety disorders have chronically overaroused central
nervous systems and are slow to habituate to noxious stimuli.
Several physiological processes and physical disorders pro-
duce symptoms like those of panic, arousing interest in possible
mechanisms for the disorders. These include hyperventilation,
asthma, limbic seizures, abnormal thyroid function, heart dis-
ease, hypoglycemia, and mitral valve prolapse. Of interest has
been the provocation of panic by sodium lactate infusions;
people with a history of panic disorder, but not those without
prior experience of panic, tend to react to the infusion with
panic. Furthermore, pharmacological treatment by means of
imipramine can abolish the lactate provocation of panic.
Although such observations may appear to confirm a bio-
logical basis for panic disorder, the mechanism is a subtle
one that interacts with environmental and cognitive factors,
given that the lactate provocation of panic can also be blocked
by psychological treatment.
Other biological challenges that typically produce panic-
like symptoms include the inhalation of air containing greater
than normal proportions of carbon dioxide. Recent research
indicates that healthy women during the premenstrual phase
are more likely to report panic-like symptoms in reaction to a
C0 2 challenge than healthy women in the follicular phase.
No particular biological variant explains all of the features
of agoraphobia. An inherited predisposition probably interacts
with behavioral and cognitive mechanisms to produce panic
and agoraphobic syndromes. David Barlow has pointed out
that "The fact that language and meaning structures are the
most common stimuli for anxiety in humans requires a com-
plex neurobiological system."
Social, Interpersonal, and Personality Factors
From a psychodynamic viewpoint, patients' inner representa-
tions of other people are especially relevant to agoraphobia.
Disturbed 'object relations' make the person vulnerable to
insecurity and anxiety.
From an experimental perspective, Susan Mineka and others
have shown that in humans and animals, early separation
from parents can be linked to agoraphobia-like behavior.
Agoraphobia and Panic Disorder 71
An integrative theory put forward by Alan Goldstein and
Dianne Chambless in 1978 sought to explain many of
the phenomena of agoraphobia and panic, including typical
personality factors and interpersonal styles. Goldstein and
Chambless argued that the person with agoraphobia (1) fears
panic attacks rather than particular places; (2) has difficulties
with self-sufficiency, independence, and assertiveness; (3) is
unable to trace the antecedents of emotional feelings when
they arise; and (4) develops the initial symptoms of agorapho-
bia in a climate of interpersonal conflict. The interaction of
these factors produces agoraphobia. This theory is consistent
with the observations that some agoraphobia clients have
assertiveness difficulties and troubled marital relationships.
It also gave prominence to panic attacks in what was termed
the complex agoraphobia syndrome.
Similar points have been made recently by Steven Hayes
and others in the context of developing acceptance and com-
mitment therapy (ACT), which recasts older conditioning ter-
minology, such as stimulus generalization, into the language of
stimulus equivalence and the verbal classes to which anxiety-
relevant situations belong. Hayes illustrated this by describing
a client with agoraphobia who had initially felt trapped at a
checkout counter at the mall, and later started to worry about
being trapped in an unsatisfactory marriage and an unfulfilling
job. Hayes cited evidence that clients who were especially likely
to avoid rather than confront challenging situations developed
additional anxiety symptoms later. Behavioral avoidance and
attempts to suppress dysphoric private experiences led to
increased, not decreased, distress.
Behavioral and Cognitive Theories
Conditioning theories
Early behavioral theories of the etiology of agoraphobia called
attention to classical conditioning as a possible mechanism,
and treatments designed to replicate extinction procedures
have been notably successful. Exposure in vivo, in which one
learns to confront agoraphobic situations without leaving
at the onset of anxiety, helps patients overcome a pattern
of avoidance of situations and can attenuate panic attacks.
However, the success of such treatments cannot confirm that
conditioning experiences cause agoraphobia.
Panic attacks could result at least partly from conditioning
processes. A panic attack may be viewed as the result of a
vicious circle or upward spiral in which, at each point, internal
stimuli associated with anxiety, such as those produced by
increases in heart rate, elicit conditioned anxiety responses
themselves in a process known as interoceptive conditioning.
It follows from this view that it will be helpful therapeutically
for the patient to confront anxiety sensations specifically rather
than simply the external situations in which they arise. The
exposure principle predicts that systematic confrontation of
sensations such as breathlessness will ultimately diminish
their power to evoke anxiety.
If any arousal of anxiety leads inexorably to a vicious circle
that culminates in a panic attack, then people with panic
disorder would never experience limited episodes of mild
or moderate anxiety. However, it is usual for panic disorder
patients to display moderate levels of generalized anxiety
between their panic attacks. Unadorned conditioning theories
do not easily explain why mild anxiety does not always escalate
into panic attacks. The cognitive therapy approach attempts to
address this problem.
Cognitive theories
Aaron Beck's cognitive therapy rests upon theoretical assump-
tions that center upon the individual's appraisal of events.
Such appraisals range from fleeting automatic thoughts in
the form of accessible, though covert, verbalizations (e.g.,
"Oh, no. I knew I'd get anxious if I came to the mall, and
I feel slightly dizzy already!") to deeper and more enduring
cognitive schemas, not necessarily verbalized, reflecting a more
fundamental attitude (e.g., strange feelings could indicate an
impending medical crisis).
Central to the application of cognitive therapy to panic
attacks is the patient's appraisal of the bodily sensations or
somatic cues connected with mounting anxiety. David Clark
has argued that people with panic disorder have developed
cognitive schemas concerning vulnerability to medical cata-
strophes, and he and others have demonstrated that people
with panic disorder show cognitive biases in that direction.
This model complements the conditioning of somatic cues
model by indicating who is vulnerable to panic and why not
all anxiety episodes culminate in panic. Variations in cognitive
appraisals between and within individuals may account for the
unpredictability of panic attacks. In Clark's model, patients
make catastrophic misinterpretations of bodily sensations
related to anxiety, viewing them as signals of a medical disaster
such as a heart attack. The misinterpretation itself arouses
increased anxiety, and the vicious circle continues when further
alarming appraisals are made.
One of our clients associated a sense of warmth, especially
in her face, with the onset of a panic attack. On entering an
overheated consulting room she said: "Oh no! Is it hot in
here?" When the therapist remarked that it was, stiflingly so,
the client was relieved. She said: "Thank goodness! I thought
I was having a panic attack! " The therapist reflected that simply
stating that the room was hot was enough to forestall a panic
attack for this client, which led immediately into a fruitful
discussion of the rationale for cognitive therapy.
Experimentation in the tradition of cognitive psychology
has produced intriguing findings about the attentional and
interpretive biases that are associated with, and may play a
role in maintaining, anxiety syndromes. Two experimental
procedures in which potentially disturbing words, such as
'choking' or 'fainting,' are presented to research participants
on computer screens - the emotional Stroop task and the dot-
probe task have been used most frequently to assess visual
attentional biases toward threatening information. The Stroop
task calls for naming the color in which a word is printed, and
the dot-probe task calls for signaling when a symbol on the
screen replaces a significant word; delays in color naming, and
rapid responses to a probe, indicate that the word is emotion-
ally poignant for the respondent. Studies employing these two
tasks provide evidence that individuals with panic disorder
exhibit selective attentional biases for panic-relevant stimuli.
Research has demonstrated that individuals with panic disor-
der have a tendency to interpret ambiguous bodily sensations
and external (environmental) stimuli as threatening. Similarly,
they tend to overestimate aversive consequences following
72 Agoraphobia and Panic Disorder
panic-relevant stimuli; that is, they exhibit a covariation bias.
Furthermore, individuals with panic disorder tend to remem-
ber panic-related situations and panic-related words especially
well compared to healthy controls.
It is important to note, however, that most studies have
relied solely on self-report, and the findings, thus, are vulnera-
ble to experimental demand and response bias. To overcome
these limitations, researchers have recently started to employ
online, indirect indicators of cognitive biases. One study, for
example, used the contingent negative variation (CNV), a
neurophysiological measure of expectancy, to examine covari-
ation bias in panic disorder. The late CNV component was
higher for panic-relevant pictures compared to both phobic-
relevant (but panic-irrelevant) and neutral pictures in indivi-
duals with panic disorder; this, however, was not the case in
healthy controls. A recent functional magnetic resonance imag-
ing (fMRI) study replicated previous findings that individuals
with panic disorder exhibit longer latencies to name the color
of panic-related words. More importantly, this study demon-
strated that the latency to color name panic-related words
correlated with increased activation of the right amygdala,
hippocampus, and prefrontal areas. Thus, neuropsychological
evidence seems to corroborate the self-report findings that
individuals with panic disorder tend to engage in biased pro-
cessing of panic-relevant stimuli.
A comprehensive model
David Barlow's comprehensive theory suggests that panic
results from activation of an ancient alarm system, and repre-
sents the basic emotion of fear, while anxiety is a more general
cognitive-affective structure. Panic occurs in response to true
alarms (panic attacks elicited by genuine danger), false alarms
(panic attacks in the absence of objective danger that result from
a genetically determined predisposition in interaction with an
accumulation of general stress), and learned alarms (panic
attacks that are triggered by external or internal cues). False
and learned alarms are relevant to the panic and agoraphobia
syndromes. Anxious apprehension also plays a part; a cognitive
schema containing propositions concerning anxiety elicits neg-
ative affect when triggered, resulting in directing attention to
internal self-evaluations, increased arousal, narrowing of atten-
tion, and hypervigilance concerning sources of apprehension.
Barlow's model of agoraphobia is his model of panic
disorder with the addition of the development of agoraphobic
avoidance. Biological vulnerability interacts with objective
stress to produce an initial uncued panic attack, or false
alarm. The connection of the panic attack with interoceptive
cues leads to the development of cued learned alarms. As a
result, there is a psychological vulnerability characterized by
anxious apprehension about future panic attacks. Next, panic
attacks are triggered unpredictably by a combination of auto-
nomic and cognitive symptoms of anxiety with additional
somatic cues. Avoidance behavior may develop, giving rise to
the panic disorder with agoraphobia syndrome.
Clinical Assessment
In assessing clients with the agoraphobia and panic syn-
dromes, the first step involves establishing the appropriate
diagnosis. Second, identifying the specifics of a patient's levels
of distress and impairment allows development of an indivi-
dualized treatment plan. Third, evaluating concomitant pro-
blems permits employment of adjunct treatments or influences
the sequence in which treatments are provided. Fourth, moni-
toring the patient's progress throughout the course of therapy
is essential in determining response to treatment and alerting
the clinician to needed procedural changes.
Diagnosis
People with agoraphobia and panic disorder syndromes may
self-refer for treatment, having made a self-diagnosis after
reading a magazine article or viewing a television presentation.
Patients may be referred to a mental health professional by
emergency room staff after one or more visits for urgent treat-
ment during panic attacks. Because many people with agora-
phobia are either entirely housebound or have a limited range
of travel, clinicians working with this disorder become accus-
tomed to making home visits, at least in the early stages of
assessment and treatment.
Because there are several physical conditions that give rise
to symptoms like those of agoraphobia and panic, it is impor-
tant that patients receive a physical examination before mental
health interventions begin. If anxiety persists despite successful
treatment of a precipitating or complicating physical condi-
tion, then treatment of the anxiety disorder proceeds.
Assessment is needed to identify comorbid psychiatric
disorders. Also relevant for assessment are issues like marital
conflict, social skills deficits, and difficulty with personal auton-
omy that may not require a formal diagnosis but may yet be
important foci for intervention. Not all people who experience
anxiety when in public places or who have had panic attacks
show patterns that meet diagnostic criteria for agoraphobia or
panic syndromes.
Assessing the Range and Extent of Anxiety and
Avoidance Behavior
Simply applying the appropriate diagnostic label is insufficient
to guide treatment. The clinician seeks to know the patient as a
unique individual, and accordingly conducts the usual psycho-
social history and mental status examination. Beyond that, the
nature and extent of the anxiety problems will need to be
charted in sufficient detail to allow formulation of an appro-
priate individual treatment plan and continued evaluation of
progress toward treatment goals.
Developed by David Barlow and his colleagues, the Anxiety
Disorders Interview Schedule-IV (ADIS-IV) allows detailed and
accurate characterization of the person's anxiety problems
and permits authoritative diagnosis in DSM-IV-TR terms. The
instrument is primarily employed in research trials to ensure
uniformity of diagnostic practices. Although the complete pro-
tocol is too lengthy for routine clinical use, subsets of the
ADIS-IV may be used appropriately and conveniently in most
clinical settings.
Clients are asked to self-monitor general anxiety, panic
attacks, and agoraphobic avoidance daily. Individualized forms
may be used so that details of the specifics of the patient's
situation may be accommodated. For example, daily ratings
Agoraphobia and Panic Disorder 73
may be made of avoidance of, fear in, and self-efficacy
concerning each item in a customized graded hierarchy of
feared situations. Daily ratings of the frequency and intensity
of panic attacks allow the patient to record the circum-
stances surrounding each episode - situational, cognitive,
and interpersonal.
A hierarchically ordered behavioral test may be designed for
patients with agoraphobic avoidance patterns. This takes the
form of an unaccompanied journey - walking, driving, or
using public transportation - to take in as many situations
relevant to the patient's fear and avoidance as is feasible. The
clinician asks the patient to proceed as far as possible, and
takes the distance actually traveled as a helpful datum in
sampling current levels of agoraphobic avoidance.
Physiological monitoring has been a customary component
of research trials designed to provide generalizable informa-
tion on treatment effectiveness, but is far less common in
routine clinical practice. Typically, measures of anxiety in the
different domains - self-report, behavioral observation, and
psychophysiological - do not covary as might be predicted.
When all such measures are available, it is recommended that
treatment proceed until clear reductions have been seen in
each measurement modality.
Treatment
Pharmacological Treatment
Pharmacological treatment has several advantages for the
patient, and significant progress has been made in this area
since 1970, improving the general outlook for agoraphobia
and panic disorder. Medication is readily available and conve-
nient to use. However, concerns arise from medication side
effects, from high relapse rates when the medication regimen is
discontinued, and - in the case of minor tranquilizers - from
the possibility of substance dependence.
Antidepressants
Medications initially developed to treat depression can also be
helpful in the management of anxiety disorders. The term
antidepressants may be misleading in this context because
there is controversy about their role in treating agoraphobia
and panic syndromes (do they attenuate dysphoric mood,
facilitating other treatments, or do they act specifically to
block panic attacks?).
Antidepressants often used to treat panic disorder include
selective serotonin reuptake inhibitors such as fluoxetine
(Prozac), sertraline (Zoloft), escitalopram (Lexapro), and cita-
lopram (Celexa). Tricyclic antidepressants such as imipramine
(Tofranil) continue to be helpful. Early studies showed that
imipramine reduced panic attacks, but patients continued to
avoid agoraphobic situations. Later studies demonstrated imi-
pramine's superiority to placebo medication and indicated
that it brought additional benefit when added to behavioral
treatment. However, this was not attributable to the blockade
of panic. When imipramine is used in conjunction with the
antitherapeutic recommendation to avoid confronting feared
situations, improvement in mood, but not in agoraphobia,
is the result. Imipramine plus exposure therapy seems more
effective than either treatment alone. Inconsistencies in
research findings with imipramine may result from marked
differences in dosage across studies.
Monoamine oxidase inhibitors such as phenelzine (Nardil)
are also used. Whereas some studies have shown little if any
difference between phenelzine and placebo in application
to agoraphobia, another has shown that phenelzine reduces
general disability and avoidance behavior. In that study, phe-
nelzine was more effective than imipramine. For reasons that
are unclear, phenelzine appears to potentiate self-initiated
exposure.
Benzodiazepines
The benzodiazepines are minor tranquilizers that have been
extensively prescribed for various forms of anxiety and stress
reactions, clinical and subclinical, for decades. Donald Klein's
initial work on imipramine had suggested that it is specific
for blocking panic, whereas the benzodiazepines are effective
only with generalized or anticipatory anxiety. Later work sug-
gests that high doses of benzodiazepines may be effective in
treating panic attacks. The development of high-potency ben-
zodiazepines like alprazolam (Xanax) and lorazepam (Ativan)
has brought clear benefit in the treatment of agoraphobia
and panic. Alprazolam has been the subject of a multicenter
worldwide double-blind study of people with panic disorder
(with and without agoraphobia). Fifty percent of the alprazo-
lam patients and 30% of placebo patients were panic-free
3 weeks after the start of the trial.
Strong withdrawal reactions after discontinuance of alpraz-
olam pose a significant problem, as does the phenomenon of
rebound panic in which a minority of patients may experience
even worse panic attacks after withdrawal from medication
than before treatment.
Summary
Antidepressants and benzodiazepines are helpful in the treat-
ment of agoraphobia and panic. Concerns include side effects,
rebound panic after withdrawal from the drug, and chemical
dependency. Since the late 1990s, the selective serotonin reup-
take inhibitors and other new medications have been widely
prescribed for people with agoraphobia, and there is a ferment
of pharmacological research activity. The mechanisms under-
lying successful pharmacological treatment are unclear.
Psychological Treatment
Psychodynamic approaches to agoraphobia and panic have
received far less attention than biological, behavioral, and
cognitive approaches in recent decades, and there is no corpus
of empirical research on psychodynamic formulations of etiol-
ogy or on the results of psychodynamic treatment. Behavior
therapists have progressed from desensitizing anxiety to treat-
ing avoidance behavior and panic attacks.
Treatment ot agoraphobic avoidance behavior
Despite initial enthusiasm for Joseph Wolpe's technique of
systematic desensitization as a therapeutic breakthrough for
phobias, its application to agoraphobia in controlled clinical
trials in the 1960s brought disappointing results. Develop-
ments in the 1970s established flooding in fantasy and graded
practice in real life as effective treatments. Researchers in
74 Agoraphobia and Panic Disorder
Vermont led by Stewart Agras and Harold Leitenberg showed
that graded practice - with or without praise for specific accom-
plishments - could quickly reduce agoraphobic patients'
avoidance of unaccompanied journeys away from the clinic.
This work converged with that of Isaac Marks, Andrew
Mathews, and others in the United Kingdom to identify
exposure in vivo as the central ingredient of psychological
treatment for agoraphobic avoidance.
Procedural variations such as brief or prolonged exposure
duration, massing or spacing of treatment sessions, and termi-
nating exposure at the point of increasing or decreasing anxiety
were examined assiduously by clinical researchers, but the
consensus is that these technical details are less important
than the general recommendation to confront, rather than
avoid, feared situations. This exposure principle is as well
founded as any in the entire field of mental health work.
Treatment of panic
The essential technique in the psychological treatment of panic
is exposure to somatic cues, or reproduction of and confronta-
tion by the bodily symptoms that the patient associates with
panic attacks. The patient is asked to create sensations of panic
deliberately during and between treatment sessions. Running
in place, voluntary hyperventilation, and spinning around in a
swivel chair are examples of procedures for creating such sen-
sations. Clinicians match particular procedures to patients'
most troublesome symptoms; someone who is most alarmed
by dizziness will practice spinning around, while someone
disturbed by the sensations of a rapid heart rate will run up
and down the stairs.
In early trials, this approach brought the most impressive
results yet seen in the treatment of panic and agoraphobia, the
success rates approaching 100% in some studies. Advances in
methodology that have allowed the daily monitoring of panic
attacks have permitted accurate tracking of panic attack fre-
quency. 'Percentage of patients panic free' has become a stan-
dard datum to report in contemporary treatment trials.
The efficacy of exposure to somatic cues has been attributed
to various theoretical processes. These include the exposure
principle, possibly resting upon the extinction or habituation
of conditioned anxiety responses to panic sensations, or upon
the development of coping skills by the patient. The success of
the method is consistent with the specific hypothesis that
chronic hyperventilation underlies panic disorder. It is also
consistent with the cognitive therapy view that the patient
makes catastrophic misinterpretations of the bodily sensations
of panic, ascribing to them morbid significance as harbingers
of a medical emergency.
Parallel to exposure to somatic cues is cognitive therapy in
the contemporary treatment approach to panic. Consistent
with David Clark's model of an interaction of sensitivity to
somatic cues and the catastrophic misinterpretation thereof,
patients are engaged in a cognitive treatment process of collab-
orative empiricism in which implicit schemas construing panic
sensations as signals of dire illness are carefully assessed, gently
challenged, and empirically tested. Cognitive therapy involves
exploring, in a sympathetic and accepting way, the specific
idiosyncratic cognitions that are assumed to underlie emotional
distress. Real-life 'experiments' are undertaken in attempting
to challenge unrealistic assumptions. There is no standard,
structured format that must be applied systematically to all
patients; rather, the principles of cognitive therapy guide a crea-
tive treatment approach with each individual. The results of
preliminary trials of cognitive therapy have been as encouraging
as those of exposure to somatic cues, and the combination of
these treatments has brought the best outcomes.
Guidelines by the National Institute for Health and Clinical
Excellence in the United Kingdom state that cognitive-behavior
therapy is now accepted as, and should be employed as, the
treatment of choice for panic disorder. Structured treatment
protocols for the comprehensive treatment of anxiety and
panic have been developed and researched by Michelle Craske
and David Barlow. Therapist guides and client workbooks are
available.
Comprehensive treatment of agoraphobia
In addition to the central psychological treatment approaches
of exposure in vivo, exposure to somatic cues, and cognitive
therapy, relaxation training, and breathing retraining have
been found helpful in the treatment of agoraphobia and are
recommended as optional components of a treatment plan.
There is a consensus that in the typical case of panic disorder
with agoraphobia, treatment should proceed employing all of
these techniques in sequence, beginning with self-paced expo-
sure in vivo. Experts argue that, because it is not associated with
deleterious side effects or complications from withdrawal, psy-
chological treatment should be used first, and pharmacological
treatment brought in as necessary subsequently.
Alternatively, using short-acting anxiolytic medication to
help patients with severe agoraphobia embark upon an
in vivo exposure venture, and then prolonging the exposure
while the effects of the medication wane, has been validated in
controlled studies by Marks and his associates as an ideal
method for combining the advantages of both treatments
rationally in a carefully structured treatment program.
Conclusions and Prospects
Although agoraphobia is fragmented by the current nomencla-
ture into two distinct disorders, it is a coherent syndrome with
a range of symptomatology extending far beyond the limited
compass of specific phobias. Panic disorder, which overlaps
with agoraphobia in one of the syndromes, is associated with a
degree of distress similar to that seen in the mood disorders.
Agoraphobia and panic disorder have yielded in recent years to
effective pharmacological, behavioral, and cognitive treatments.
Currently in need of further attention by clinicians and
researchers are the following. Whereas pharmacological treat-
ment is readily available, it is difficult for many patients to
gain access to specialized psychological treatment, especially in
rural areas. Innovations in service delivery are needed, and
studies should address the viability of psychological treatment
of agoraphobia from remote sites by means of the latest
communications technology. Many communities are under-
served by mental health professionals, and individuals who
are housebound by agoraphobia have even greater difficulties
than most people in gaining access to needed psychological
services. Encouraging developments in this context include the
recent finding in the United Kingdom that brief training for
Agoraphobia and Panic Disorder 75
primary care therapists in cognitive therapy for panic disorder
produced significantly better outcomes than treatment as
usual. Self-directed treatment with limited therapist contact
has been shown to be effective for clients in rural areas with
sparse professional resources.
See also: Anxiety and Fear; Anxiety Disorders; Clinical Assessment;
Cognitive Behavior Therapy; Depression; Phobias.
Further Reading
Barlow DH (2002) Anxiety and Its Disorders: The Nature and Treatment of Anxiety and
Panic, 2nd edn. New York: Guilford.
Barlow DH and Craske MG (2007) Mastery of Your Anxiety and Panic: Workbook,
4th edn. New York: Oxford University Press.
Beck AT, Emery G, and Greenberg RL (1985) Anxiety Disorders and Phobias:
A Cognitive Perspective. New York: Basic Books.
Chambless DL and Goldstein AJ (eds.) (1 982) Agoraphobia: Multiple Perspectives on
Theory and Treatment. New York: Wiley.
Clark DM, Salkovskis PM, Hackmann A, Middleton H, Anastasiades P, and Gelder M
(1994) A comparison of cognitive therapy, applied relaxation and imipramine in
treatment of panic disorder. British Journal of Psychiatry 164: 759-769.
Craske MG and Barlow DH (2006) Mastery of Your Anxiety and Panic: Therapist Guide,
4th edn. New York: Oxford University Press.
Grey N, Salkovskis P, Quigley A, Clark DM, and Ehlers A (2008) Dissemination of
cognitive therapy for panic disorder in primary care. Behavioural and Cognitive
Psychotherapy 36: 509-520.
Hecker JE, Losee MC, Fritzler BK, and Fink CM (1996) Self-directed versus therapist-
directed cognitive-behavioral treatment for panic disorder. Journal of Anxiety
Disorders 10: 253-265.
Hecker JE and Thorpe GL (1 992) Agoraphobia and Panic: A Guide to Psychological
Treatment. Boston, MA: Allyn and Bacon.
Kessler RC, Chiu WT, Jin R, Ruscio A, Shear K, and Walters EE (2006) The
epidemiology of panic attacks, panic disorder, and agoraphobia in the
National Comorbidity Survey Replication. Archives of General Psychiatry
63: 415-424.
Knapp TJ and Schumacher MT (1988) Westphai's “Die Agoraphobie" . Lanham, MD:
University Press of America.
Marks IM (1 987) Fears, Phobias, and Rituals: Panic, Anxiety, and Their Disorders.
New York: Oxford University Press.
Mathews AM, Gelder MG, and Johnston DW (1981) Agoraphobia: Nature and
Treatment. New York: Guilford.
Panic disorder, http://www.nimh.nih.gov/health/topics/panic-disorder/index.shtml
(accessed August 2009).
Rapee R (1 998) Fight or Flight? Overcoming Panic and Aagoraphobia [Videotape
Presentation and Demonstration], New York: Guilford.
Schartel JG (2009) The effect of anxiety sensitivity and menstrual cycle phase on
psychological and psychophysiological reactivity to a carbon dioxide challenge
task. Unpublished doctoral dissertation, University of Maine.
Sigmon ST, Dorhofer DM, Rohan KJ, Hotovy LA, Boulard NE, and Fink CM (2000)
Psychophysiological, somatic, and affective changes across the menstrual cycle in
women with panic disorder. Journal of Consulting and Clinical Psychology
68: 425-431.
Thorpe GL and Burns LE (1983) The Agoraphobic Syndrome: Behavioural Approaches
to Evaluation and Treatment. Chichester: Wiley.
Walker JR, Norton GR, and Ross CA (eds.) (1 991 ) Panic Disorder and Agoraphobia:
A Comprehensive Guide for the Practitioner. Pacific Grove, CA: Brooks/Cole.
Wittchen H-U, Gloster AT, Beesdo-Baum K, Fava GA, and Craske MG (2010)
Agoraphobia: A review of the diagnostic classificatory position and criteria.
Depression and Anxiety 27 : 113-133.
Relevant Website
http://www.nimh.nih.gov/health/topics/panic-disorder/index.shtml - National Institute
of Mental Health in the United States.
Agraphia and Alexia
D Fiset, Universite du Quebec en Outaouais, Gatineau, QC, Canada
D Bub, University of Victoria, Victoria, BC, Canada
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cognitive architecture The component processes and their
interconnections that make up a more complex mechanism
involved in a task like reading single words aloud or writing
words to dictation.
Grapheme The smallest combination of letters associated
with an elementary sound unit. Graphemes can be as small
as a single letter. For example, the letter P corresponds to the
pronunciation 'puh' while the letter pair PH (a bigram)
corresponds to 'fuh.'
Hemifield One half of the field of view defined according to
retinal coordinates.
Hemispatial neglect A neuropsychological condition
that affects attention, exploration, and awareness
of the hemispace opposite the damaged hemisphere.
Clinical manifestations of neglect include bumping
into objects and walls, ignoring objects, persons, and
sounds coming from the affected side, forgetting to
shave (or applying make-up) to one part of the
face, etc.
Modular Refers to the idea of separable cognitive
components working together to carry out a task.
For example, the orthographic lexicon is a component
that is modularly distinct from the phonological lexicon.
Modularly distinct components can be independently
affected by neurological damage.
Orthographic lexicon The stored representation of the
spelled form of words in the reader's vocabulary. The
orthographic lexicon maintains each word as a sequence of
abstract letter identities.
Phonological lexicon The stored representation of the
pronunciation of the words in a speaker's vocabulary.
Retinocentric A spatial coordinate system centered on
the retina.
Syndrome In cognitive neuropsychology, the term refers to
a cluster of impairments on a number of different tasks, and
the co-occurrence of symptoms reflects a theoretically
important principle. The term is not applied to impairments
that co-occur simply because neurological damage has
fortuitously affected a number of unrelated processes.
The advent of the visual word as a means of communicating
ideas and preserving knowledge is of fundamental importance
to human development. Because of the crucial role played by
written language in so many domains it is no surprise that
learning how to fluently read and spell occupies much of our
initial efforts at school. As a general rule, good readers are also
good spellers, while bad readers have a harder time producing
the correct form of written words. This correlation suggests that
these two tasks share common substrates. But given the
demands specific to each task, reading and writing cannot
depend on exactly the same underlying mental processes. For
example, the rapid identification of a letter string depends on
certain mechanisms unique to vision, whereas the retrieval and
execution of letter shapes require motor processes specific to
handwriting.
The main goal of this chapter is to describe and understand
how certain neurological lesions can induce deficits in reading
and writing (i.e., alexias and agraphias). What these different
pathologies can tell us about the organization of the mental
processes responsible for reading and writing in normal indi-
viduals will also be addressed. Neurological damage can selec-
tively affect the peripheral components of word processing
(e.g., the perception or production of letters) or its central
components (e.g., the relationship between the orthographic
form of a word and its sound or meaning). Generally speaking
the more peripheral the damage, the more likely the impairment
will be specific to either reading or writing. Throughout this
article, the reading and writing system will be deconstructed
into different components, and the disorders associated with
their impairment will be described. First, the neuropsychologi-
cal disorders related to the uptake of visual information will be
presented. Next, deficits resulting from damage to more central
mechanisms will be explained. These central mechanisms relate
to orthographic, phonological and semantic knowledge, and
often affect both reading and writing simultaneously. Lastly,
the deficits that follow damage to orthographic output or,
more specifically, to letter production, will be described.
Peripheral Dyslexias
Skilled reading demands accurate and rapid visual access to the
symbols displayed on a page or computer screen. The quality
of the visual information perceived is in part determined by the
capacity to efficiently direct one's gaze and/or attention toward
the text on the page (e.g., find the first paragraph). At the same
time, in order for reading to be rapid and efficient, one must
quickly recognize the letters in each word while maintaining
their relative positions (e.g., LISTEN is not the same word as
SILENT). In brain-lesioned patients who suffer from reading
disorders as a result of damage to peripheral systems, the visual
information perceived is inadequate to support accurate or
fluent reading. Although these lesions will frequently have
major repercussions on reading abilities, they will not affect
spelling. Three types of peripheral alexias will be described in
this section: pure alexia, attentional dyslexia, and neglect dys-
lexia. Note that we will use the term 'alexia' for some varieties
of reading disorder and 'dyslexia' for others. 'Alexia' has
76
Agraphia and Alexia 77
generally been the term that denotes reading disorders acquired
as a result of neurological injury, while 'dyslexia' has more
commonly been used to refer to developmental disorders of
reading. However, the term 'dyslexia' has also been applied to
certain forms of acquired reading disorder (e.g., deep dyslexia,
attentional dyslexia, and neglect dyslexia) as has the term
'dysgraphia' to certain acquired writing/spelling disorders. We
have varied our labeling, therefore, so as to be consistent with
the published nomenclature.
Pure Alexia (Alexia Without Agraphia)
A dramatic example of a reading disorder that occurs without
corresponding difficulties in spelling or writing is pure alexia
(or alexia without agraphia). Jules Dejerine first described this
syndrome over a hundred years ago at the Biological Society in
Paris. In this seminal work, Dejerine documented the remark-
able case of 'Monsieur C' both behaviorally and anatomically.
Monsieur C, a highly educated businessman who had suffered
a left hemisphere stroke, showed preserved writing and spoken
language skills. He wrote fluently both spontaneously and to
dictation but showed severe word blindness; that is, Monsieur
C could not even read sentences he himself had written once
the memory of the text had faded.
From neuroanatomical and behavioral observations,
Dejerine concluded that pure alexia is a deficit caused by a
disconnection between vision and language-based areas of the
brain, a theory that Geschwind also resurrected in the mid-
twentieth century. The disconnection theory Geschwind pro-
posed entails that at least two lesions are necessary to isolate
words from visual input, resulting in pure alexia: one in the left
occipital lobe and the other in the splenium of the corpus
callosum. Although it is a possible explanation for some of
the cases reported over the years, current research suggests
a more parsimonious neuroanatomical explanation. Most
patients with pure alexia show a lesion of a small section of
the cortex in the left mid-fusiform gyrus. This region is the only
visible damage in at least one documented case of pure alexia.
However, the specific role of this brain region and its impor-
tance for efficient reading remains a matter of intense debate.
The majority of cases showing alexia without agraphia are
not as severely impaired as Monsieur C, who was unable to
identify even single letters, treating them as unfamiliar shapes
(e.g., he indicated that the letter A looked something like an
artist's easel and S like a snake). The majority of pure alexics are
able to identify individual letters but are forced to adopt a
laborious approach to decipher even short familiar words. As
a result, pure alexic readers (or letter-by-letter dyslexics as the
syndrome is now often called) show very slow word identifica-
tion (e.g., more than 2.5 s on average for even short, familiar
words). A defining feature of letter-by-letter reading is the so-
called word-length effect; that is, a linear relationship between
the number of constituent letters and the time taken to read a
word. The abnormally large word-length effect in this reading
disorder is in striking contrast with the reading profile of
normal readers, who show no word-length effect in naming
latencies for words of less than seven letters.
The effect of length on performance suggests that patients
with this form of reading disorder are limited to deciphering
words by analyzing each letter in sequence, so that the longer
the word, the slower the reading speed (hence the term letter-
by-letter dyslexia). Interestingly, though, this surface feature of
the disorder is misleading. Good evidence indicates that it is
generally not the number of constituent letters that impacts the
reading of many letter-by-letter dyslexics but the extent to
which the word contains letters that are perceptually confusa-
ble with other letters of the alphabet. This confusability index
tends to be greater for longer words, simply because the longer
the word, the more letters there are to be identified that are
potentially confusable with other letters. It is possible to vary
length while holding constant the summed confusability scores
of the letters making up the word. When word-confusability
scores are controlled in this way, letter-by-letter readers continue
to read slowly, but their performance is generally not slower for
longer than for shorter words. Thus, reading in this syndrome,
despite the modern label for the disorder, does not seem to be
invariably confined to a sequential analysis of letter identities.
Rather, simultaneous analysis of multiple letters generates
perceptual noise that delays recognition, and the degree of per-
ceptual noise is determined by the overall confusability of the
letters in a word.
Pure alexia offers a rich window into the visual mechanisms
dedicated to word recognition. How specific is the disorder to
reading? There are two alternative views on this question. The
first attributes the reading disorder to damage of a specialized
system for the visual analysis of alphabetic stimuli (i.e., words
and letters). This account is based on the hypothesis that visual
experience and expertise with words is sufficient to create
area(s) of visual cortex responsive uniquely to visual words.
Theories postulating that only alphabetic stimuli are impaired
in pure alexia are examples of this specific account. The second
account proposes that no brain part is uniquely devoted to
visual word recognition per se but that some neural mechan-
isms that determine efficient perception of words are less cru-
cial for other kinds of objects (e.g., faces, animals, man-made
three-dimensional objects, etc.). This last account predicts that
pure alexia is not a deficit specific to alphabetic material but
rather includes an impairment in any task where the necessary
visual resources coincide with those needed for reading. For
example, one account proposes that visual word recognition
requires simultaneous integration of multiple letters and that
this integration mechanism is not specific to reading.
All thoroughly tested patients with pure alexia show some
form of visual processing impairment not specific to alphabetic
stimuli. Hence, the hypothesis of a general visual deficit is
actually plausible. But the nature of the perceptual mechan-
isms, so important for reading yet of much less significance in
other visual tasks, remains to be determined.
Attentional Dyslexia
Attentional dyslexia is a rare form of reading disorder that
occurs after a brain lesion to the left parietal cortex. The main
behavioral symptom is a failure to correctly perceive one item
of a given category (e.g., the letter A) when this item is pre-
sented simultaneously with other items of the same category
(e.g., BDGAH). A typical patient is able to read isolated words
or letters with relative ease but his performance declines when
78 Agraphia and Alexia
he is asked to identify those same items in a sentence or a letter
string - because many items are visible at the same time. These
patients also make many errors when they are asked to identify
a single digit flanked by other digits but the error rate decreases
significantly when the target is a set of dots to be counted. This
behavioral observation implies that naming (in this instance,
of numbers) is not accountable for the deficit. The main deficit
in attentional dyslexia is a lack of control of a filtering mecha-
nism that suppresses visual processing of unattended items in
the visual field. Thus, these patients are unable to direct their
attentional spotlight to a specific region of interest. As a result,
more input than expected falls within the attentional spotlight
and the wrong features are integrated together leading to per-
ceptual errors. Within this framework, error rates are larger
when the surrounding stimuli are from the same category.
Neglect Dyslexia
Although it is frequently associated with hemispatial neglect
and right hemisphere damage, neglect dyslexia can also be
observed in isolation. In fact, at least one patient has been
known to have a right-side hemispatial neglect following a
left hemisphere lesion in conjunction with left-side neglect
dyslexia following right hemisphere damage. This exceptional
patient undoubtedly proved the existence of a double dissocia-
tion between nonverbal neglect and neglect dyslexia. In this
form of peripheral dyslexia, reading is impaired because the
patients tend to omit or misread the left side of a page or the
first letters of letter strings or words of a sentence. At the word
level, visual errors made by left neglect dyslexics can be addi-
tions of letters at the beginning of words ('love' read as 'glove'),
omissions ('cowboy' read as 'boy') or substitutions ('mother'
read as 'bother'). In right neglect dyslexia, these errors occur at
the end of the word. By itself, the existence of neglect dyslexia
suggests that spatial attention plays an important role in nor-
mal reading. However, neglect dyslexia is quite variable
between patients and different attentional mechanisms have
been proposed to explain this variability.
It has been argued that at least three spatial frames of
reference are necessary for understanding visual attention in
words. The first frame is retinocentric; spatial information is
represented by means of a coordinate system centered on the
retina. When this form of spatial attention is impaired in
neglect dyslexia, omissions and errors tend to be on the letter
strings that appear in one hemifield, generally the left in
patients with right hemisphere damage. Indeed, a word written
on the left side of a sheet will be ignored or misread while it
will be read correctly if it is presented on the right, no matter
the word's orientation (right to left, left to right, vertically, etc.).
The next frame of reference is a stimulus-based one. When
this frame is impaired for reading, words' spatial representa-
tions are altered. The left side of the word as a string of letters is
misread, regardless of whether the orientation demands a left-
to-right or right-to-left perusal. A word such as 'marble' might
be read 'cable' if presented from left to right but as 'march' if
demanding a right-to-left analysis of letters (i.e., 'elbram'). In
contrast to retinocentric impairment, damage to the second
frame of reference leads to the same error rates no matter
where the letters string is presented in the visual field.
The third and last frame of reference implicates cognitive,
higher-level representations of known words, and the proces-
sing is no longer at the visual extraction level. When this
reference frame is impaired for reading, a word such as 'bible'
could be read as 'cable' even if presented from left-to-right or
right-to-left. Interestingly, this kind of patient tends to commit
similar errors in writing or spelling, thus suggesting that at this
level, attention is applied to more central orthographic repre-
sentations, akin to the reading and spelling errors in central
alexias and agraphias.
Central Dyslexias and Dysgraphias
Central dyslexia and dysgraphia are the result of a disruption to
the linguistic mechanisms (i.e., orthographic, semantic, or
phonological) that mediate reading and spelling, respectively.
Neuropsychological studies of brain-damaged patients have
contributed substantially to the development of a model spe-
cifying the cognitive processes and their interconnections that
determine reading and spelling (see Figure 1). The structure
of the model - its cognitive architecture - is founded on the
assumption that separate routines from print to sound (or in
the case of writing, from sound to print) are recruited for
familiar and novel words. A familiar word contacts its stored
visual form and meaning, and these representations are then
available for reading aloud and writing. If a novel word is
encountered (or in a task devised for experimental purposes,
a pseudoword), no stored form or meaning is available.
Reading aloud or writing to dictation is then based on subword
graphemic units assigned to corresponding phonemic seg-
ments; for example, the written sequence FENT is pronounced
Spoken input/output
0
£
T ©
F -
I 8
-s -c --
_ o
Q_ CO
<t aj
e>
Acoustic/phonemic buffer
Phonological lexicon
Semantic system
Orthographic lexicon
Graphemic buffer
Letter-name Letter-shape
conversion conversion
Visual analysis/
letter identification
Written word
Figure 1 Modular architecture of reading and writing.
Agraphia and Alexia 79
by knowing the sound associated with the letter F, and the
typical rendering of the segment ENT. The framework offers an
account of certain aspects of normal as well as pathological
performance but is not without controversy. An alternative
connectionist architecture rejects the view of two independent
procedures for transcoding familiar and novel words. This rival
approach has three sets of simple processing units, and so is
referred to as the 'triangle model': grapheme units, representing
orthography, phoneme units representing sound, and an array
of semantic units for meaning. A spelling pattern generates
recurrent activity over the grapheme units which then propa-
gates through the network, resulting in a pronunciation for
reading aloud (writing would entail a similar propagation of
activity but from phoneme-to-grapheme units).
The triangle model, like its rival, includes two pathways
between print and sound; a direct mapping between gra-
phemes and their corresponding pronunciation, and a second
pathway from print to sound via semantic units. Neverthe-
less, there are important differences between connectionist
approaches and the more traditional dual-route model
depicted in Figure 1, the most basic of which is that the same
processing mechanism is applied to the way familiar and unfa-
miliar words are mapped between spelling and sound. For
expository purposes, we will follow the logic of the traditional
model to describe central reading and spelling disorders. We
will therefore begin our discussion on the central deficits of
reading and writing with a short portrayal of the main compo-
nents of this model.
In order to read, children must learn to assign a pronuncia-
tion to spelling units (e.g., the pronunciation of a sequence
like INE). For writing, a similar mechanism is required but
operating in reverse, to convert a pronunciation into a valid
graphemic pattern. Note that the cognitive constraints on
spelling-to-sound and sound-to-spelling conversion procedures
are rather different. For example, the letter string HAIR has only
one possible correct pronunciation but the spoken sequence/
hsar/has at least two corresponding orthographic forms (e.g.,
HARE, HAIR).
As reading and spelling develop, children learn to access or
retrieve familiar words by directly contacting their stored visual
forms. The ability to read or spell by using whole words rather
than subword units implies three processing components for a
word, one for its orthography, the second for meaning, and the
last for pronunciation. This lexical-semantic route has some
similarity with a dictionary: the orthographic lexicon contains
the word's spelling (e.g., A-P-P-L-E), the semantic system
containing its definition (e.g., the fleshy usually rounded
red, yellow, or green edible fruit) and the phonological lexi-
con containing its pronunciation (e.g., /’a-pal/). Theoretically
speaking, the orthographic lexicon contains representations of
all the words that a normal subject has learned to recognize in
a single glance or to spell with high efficiency. The semantic
system is responsible for storing the meanings of words. It is
shared by other language mechanisms such as oral or sign
language. The phonological lexicon is a system involved in
the retrieval of spoken words in speech production. It plays
an essential role in the spoken component of reading.
We will see that each of these processing routes and sub-
systems can be damaged independently and cause different
patterns of dyslexia and dysgraphia.
Phonological Dyslexia and Dysgraphia
In phonological dyslexia and dysgraphia, patients have diffi-
culty in reading and spelling unknown words and pseudo-
words. Their performance, though, is clearly better or even
in the normal range for known words, particularly for high-
frequency words. Interestingly, most errors made by these
patients are real words that are visually similar to the target
word (this kind of error is termed a lexicalization; e.g., the
nonsense word SIFE read as 'sift').
The cognitive architecture implies a division between the
mechanisms responsible for known words and those responsi-
ble for unknown words and pseudowords. The fact that a
cerebral lesion can affect reading and writing of pseudowords
without affecting known words is consistent with the claim
that distinct mechanisms are implied for processing these sti-
muli. In phonological dyslexia and dysgraphia, the lexical
route is preserved, enabling the reading and spelling of familiar
words, while the nonlexical route is impaired, affecting perfor-
mance on spelling patterns that have no representation as
real words.
Surface Dyslexia and Dysgraphia
At this point in our description of central deficits, the assump-
tion that a distinction exists between a lexical and a sublexical
pathway is not the only plausible interpretation of phonologi-
cal dyslexia and dysgraphia. It is possible that pseudowords are
simply harder to read and write than known words and that
neurological damage only exacerbates this difference. The fact
that normal subjects are slower and less accurate when reading
and writing pseudowords versus real words is consistent with
this possibility.
The finding of patients with the opposite behavioral
profile - that is a deficit for words but no deficit for pseudo-
words - would eliminate this alternative possibility, however.
Marshall and Newcomb were the first to describe such a patient
in 1973. They documented what we now call surface dyslexia,
that is, an acquired reading deficit in which the reader has
difficulty reading 'exception' words (i.e., words with an irregu-
lar spelling-to-sound mapping; e.g., colonel) but makes fewer
errors when asked to read regular words (e.g., fairy) and pseu-
dowords (e.g., peratine). The initial cases described were less
clear than expected given the assumption of a categorical dis-
tinction between whole-word and subword procedures for
translating spelling to sound. Surface dyslexics often produced
errors that indicated impairment of both procedures. This lead
to some doubt about the validity of the distinction, until a
number of unambiguous cases were documented. From the
perspective of the model, the reading impairment in surface
dyslexia results from damage to the lexical route. Since this
processing route is unavailable, surface dyslexic patients
attempt to read and spell by the sublexical (i.e., grapheme-
to-phoneme and phoneme-to-grapheme conversion) processing
systems. Put differently, surface dyslexic patients often treat
irregular words as if they were regular words or pseudowords.
For example, they will pronounce COLONEL as /'koulaneil/
instead of /'k3rnl/. Such errors are called regularizations and
constitute the diagnostic pattern of surface alexia. Importantly,
80 Agraphia and Alexia
these errors are more likely to occur for less common exception
words (e.g., MOTHER might be read and spelled correctly but
not LEOPARD).
Similar symptoms have been described for writing and
spelling. In this case, patients make a large quantity of spelling
errors for irregular words all the while correctly spelling regular
words and pseudowords. The error rate in general is actually
higher for this form of dysgraphia than for the corresponding
surface dyslexia. In writing, the number of words with unpre-
dictable phoneme-to-grapheme correspondences is quite
high (e.g., the pronunciation 'boat' can be spelled in at least
two ways, if we simply apply general principles of spelling-
to-sound translation, including BOTE). Of course, some
words, for example, DOG or CAT, have highly predictable
sound-to-spelling correspondences. These words can be cor-
rectly spelled using the nonlexical conversion module. But
many words have an unpredictable orthography, and the lexi-
cal route is needed for these to be correctly spelled.
It seems reasonable that surface dysgraphia should be
seen together with surface dyslexia; both reading and writing
errors can be attributed to a failure to contact a central repre-
sentation of the word's orthographic form, and responses are
based on more generic procedures that yield regularization
errors. In very rare cases, though, surface dysgraphia occurs
selectively (i.e., reading via word-specific procedures is demon-
strably intact) or in association with another form of central
dyslexia (e.g., phonological dyslexia). What do these unusual
dissociations imply about the orthographic representations
accessed for reading and spelling? One possibility is that differ-
ent word-specific representations are available for input
(reading) and output (spelling). The alternative is that the
representations are indeed unitary but that distinct access and
retrieval procedures exist that can be separately damaged to
yield the dissociations seen in rare cases. The fact that such
dissociations are so infrequent, though, indicates that the
neuroanatomical substrates underlying any functional dis-
tinction between word-specific representations involved in
reading and spelling must be so organized that they are mostly
affected together.
Surface dyslexic and dysgraphic errors are the result of a
failure to retrieve a word's pronunciation based on its stored
orthographic form, and an overreliance on more general cor-
respondences between spelling and sound units. We turn now
to a very different phenomenon, involving reading or spelling
impairment that ensues when the meaning of the word
remains the only point of translation between its visual form
and pronunciation, because all other routes have been
destroyed.
Deep Dyslexia and Dysgraphia
In any cognitive neuropsychological task, the occurrence of
semantic errors is probably one of the most spectacular kinds
of error. In word reading, semantic errors are the main diag-
nostic behavior of deep dyslexic patients. For example, the
typical patient could produce 'freedom' when confronted
with the word 'LIBERTY' or 'knight' in response to 'CASTLE.'
The first case of deep dyslexia was described in 1931 but the
burgeoning interest in this reading disorder began in the late
1960s with the report of Marshall and Newcombe. In writing,
the first case was described by one of us (D.B.) in the early
1980s. The limits on the performance of patients with deep
dyslexia or dysgraphia are rather similar to the constraints
suffered by patients with a phonological alexia or agraphia.
In both types of disorder, performance for nonsense or unfa-
miliar words is affected while words yield better accuracy.
However, deep dyslexic or dysgraphic patients are unable to
cope with even a short nonsense word made up of a single
syllable, thus suggesting a total destruction of the sublexical
pathway. Of course, the syndrome is also associated with
semantic errors produced by familiar words.
A comparatively large amount of research has been con-
ducted on deep dyslexia. The reading pattern is probably the
most complex and dramatic of all acquired reading impair-
ments. Although semantic errors are the most frequent and
diagnostic of errors in deep dyslexia, this kind of mistake
is only one of the many characteristics of the syndrome. The
most prominent features are (1) semantic errors, (2) visual
errors (e.g., SYMPHONY read as 'sympathy'), (3) function-
word substitutions (e.g., BUT read as 'and'), (4) derivational
errors (e.g., MARRIAGE is read 'married'), (5) a complete failure
to read pseudowords, (6) some classes of words found to be
harder than others in reading aloud (function words > (i.e.,
harder than) verbs > adjectives > nouns), (7) strong concrete-
ness (i.e., imageability) effect (TABLE easier than FATE).
To explain deep dyslexia, the dual-route model needs to
postulate a number of deficits caused by independent damage
to different functional components. Damage to the sublexical
print-to-sound translation pathway is needed to explain the
striking impairment in pseudoword reading. The presence of
semantic errors as well as the concreteness effect suggests
impairment to the semantic system itself. Destruction of the
direct pathway between the stored visual form of the word and
its pronunciation is also necessary since this procedure is
clearly not available in deep dyslexia. If it was, the semantic
errors would not occur because the pronunciation of the
word could be retrieved without requiring that its meaning
be contacted. Finally, visual errors suggest damage to the
structural analysis of words; a possible source of these errors
would be faulty processing within the orthographic lexicon.
This unlikely coincidence of deficits observed in most deep
dyslexic has led some researchers to argue that deep dyslexia
is of no relevance for comprehension of normal reading
mechanisms. Indeed, these researchers propose that deep dys-
lexics' left hemisphere lesion is so extensive that it is no longer
available for reading. In this extreme case, reading would be
mediated by a weakly literate right hemisphere. The attractive-
ness of this hypothesis comes from the reading similarities
observed between deep dyslexics and patients with an isolated
right hemisphere. However, the idea seems less plausible given
evidence from a single case demonstrating many deep dyslexic
characteristics, but whose reading was abolished after a second
left hemisphere stroke.
The question remains, then. Why does the complex of
symptoms characteristic of deep dyslexia occur, assuming that
the pattern reflects the performance of a partially damaged left
hemisphere reading system and not the normal reading cap-
abilities of a linguistically restricted right hemisphere. Some
progress on the issue has been made by researchers
Agraphia and Alexia 81
constructing and analyzing connectionist models of reading
that map the meaning of a visual word to its pronunciation.
The relevant architecture consists of a primary layer of graph-
eme units representing the visual form of words, connected via
intermediate units to 'sememe' units. The only function of
intermediate units is to learn the pattern of associations that
map specific letter patterns to a particular meaning, while
sememe units represent the meaning of a word in terms of a
limited set of semantic features like 'can be eaten,' or 'made of
wood.' Sememe units connect to phoneme units via an addi-
tional set of intermediate mapping units.
This type of architecture has trouble learning to associate
visually similar words with distinct semantic representations;
however, the inclusion of a layer of 'clean-up' units that inter-
acts with the sememe layer dramatically improves the mapping
between the orthographic form of words and its meaning, by
minimizing the confusion between visually similar words. Of
great interest is the fact that damage to a percentage of clean-up
units (by randomly altering a subset of weights) yields a num-
ber of the symptoms of deep dyslexia: visual errors occur as
well as semantic errors, because the clean-up units no longer
can differentiate between the patterns of activation generated
by visually or semantically similar words. In addition, both
deep dyslexic cases and the damaged connectionist model
produce more complex errors, involving a semantic error
driven by a visual substitution (SYMPATHY read as 'orchestra,'
via the visual confusion 'symphony'). A concrete word like
BOOK resists damage to the network more than an abstract
word like FATE, because the former has many more semantic
features than the latter. The clean-up units have greater redun-
dancy to work with and so converge on the right meaning even
after significant damage. Finally, because correct recognition of
concrete words depends heavily on clean-up units, severe
enough damage to this layer reproduces a striking reverse-
concreteness effect also observed in one rare case of deep
dyslexia; abstract words are read better than concrete words.
Despite the undeniable value of these computational accounts,
however, this approach has not yet produced a unified, coher-
ent interpretation of the full symptom complex.
Deep dysgraphia is of considerable theoretical interest
because the syndrome may sometimes occur in the absence
of deep dyslexia, though the writing disturbance is invariably
seen in association with some form of reading and language
disorder. The dissociation between deep dyslexia and dysgra-
phia is evidence that procedures affording access to meaning
from print may be partially distinct from procedures that oper-
ate in reverse to retrieve the spelled from of a word from its
meaning. In addition, cases of deep dysgraphia show errors in
their written performance indicating a failure to maintain letter
identities in an output buffer holding active the graphemic
representation of a word for sequential output. We discuss
the reason for this surprising association of seemingly dispa-
rate effects when we consider writing disorders linked to
impairment of the graphemic buffer.
Semantic Dyslexia and Dysgraphia
One possible outcome of severe damage to semantic represen-
tations is surface dyslexia, that is, a deficit in reading irregularly
spelled words. It seems reasonable to expect that words will be
treated as nonsense words given severe enough damage to
semantic representations. If the sublexical conversion pathway
remains intact, orthographically irregular words will be regu-
larized. Surprisingly, some patients - despite a severe semantic
impairment - retain an ability to read and even spell both
regular and irregular words. Semantic dyslexic cases suffer
from Alzheimer's dementia or semantic dementia. They are
able to read any kind of letter string (i.e., regular and irregular
words as well as pseudowords), but they show no understand-
ing of the words they can produce correctly. This kind of
patient demonstrates that three processing routines are avail-
able to access pronunciation from print; the spelling-to-sound
conversion procedure, the lexical-semantic pathway and a
lexical nonsemantic pathway directly mapping word-specific
orthographic forms to a pronunciation. It is this route that is
used in semantic dyslexia to read irregular words correctly.
Notably though, semantic impairment often does lead to sur-
face dyslexia rather than semantic dyslexia. Why the different
outcome in different cases? An interesting answer is that there
are substantial individual differences in the utilization of
semantic representations for reading. Some readers may need
to rely more on semantic access to read irregular words whereas
other readers are less dependent on word meaning.
Graphemic Buffer Impairment
As depicted on Figure 1, the spelled form of words is produced
by accessing a graphemic buffer, a component that receives
information from both the orthographic lexicon and the
phoneme-grapheme conversion mechanism. This component
maintains abstract letter information (i.e., case invariant infor-
mation) while peripheral conversion mechanisms operate
sequentially on each grapheme to produce the spelled form
(e.g., letter-name conversion for oral spelling, letter-shape
conversion for writing, letter-motor scheme for typing).
How should performance be affected if the graphemic
buffer were impaired, assuming that the damage is not severe
enough to preclude any form of output? The buffer holds the
final spelling of both familiar words as well as graphemic
sequences generated by spelling-to-sound conversion. Perfor-
mance should therefore be unaffected by word-specific vari-
ables like concreteness, familiarity, and grammatical class.
If the capacity of the buffer to maintain letter identities has
been reduced, longer words should yield more impairment
than shorter words. In addition, if activation in the buffer has
been compromised so as to generate interference between
letter identities, spelling errors should be more frequent in
the middle than the end of words, because terminal letters
have fewer adjacent letters to compete with their activation.
This profile has been documented in a substantial number
of dysgraphic cases. Their errors include doubling applied
to the wrong letters (e.g., STREET spelled as STRETT; consis-
tent with the idea that the doubled status of the letters is
represented separately from the letters themselves), ordering
errors (transpositions), substitutions, and omissions. A suc-
cessful computational account of this dysgraphic subtype has
been implemented (a 'competitive queuing model') based on
principles derived from a class of models that deal with the
82 Agraphia and Alexia
control of sequential output. The model includes the idea that
response elements organized for sequential output are selected
by means of an activation gradient; the more active an element
the sooner it is produced. In addition, competition occurs
between response elements. Random noise added to the
sequence generation process yields the key features of a gra-
phemic buffer dysgraphia.
A surprising variant of the syndrome shows many of the
cardinal signs indicative of a damaged buffer but in addition,
elements of deep dysgraphia, including word class effects (like
concreteness) and semantic errors. Nonsense words are spelled
very poorly. Like the more typical instances of the syndrome,
letter deletions, substitutions, and transpositions occur and
spelling is worse for longer than shorter words. However,
errors increase monotonically from the beginning to the end
of the word (whereas in the more typical case, as we have seen,
lower error rates occur at the start and end of words). This
variant can be understood if the output of the semantic path-
way yields a degraded representation of letters in the graphe-
mic buffer. Sub threshold activation of letter identities results in
numerous errors, including premature termination of the word
if the very low signal from the last few letters is taken to
indicate that the end of the word has been reached.
It was originally assumed that the graphemic buffer's
role was restricted to the writing domain. However, the buffer
may also be involved in reading, maintaining the level of
activation generated by letters in the word. Accordingly,
patients with a lesion to the graphemic buffer frequently have
difficulty reading items comprising many letters, especially
when pseudowords must be read and the orthographic lexicon
cannot contribute to the response. Normal readers supply
converging evidence for a short-term visual memory needed
to derive the pronunciation of unfamiliar words. In this frame-
work, the only strictly peripheral agraphias are those coming
from lesions to the letter-shape and letter-name conversion
mechanisms.
More Peripheral Agraphias
Impaired writing can occur even though the spelled form of the
word has been correctly retrieved. In this type of agraphia, the
word TABLE would be spelled aloud as 'tee, ay, bee, ell, eeh,'
yet errors occur when the letters are produced in written form.
The spelling of the word is clearly available, therefore, at least
in a form suitable for naming (but also for other tasks; e.g.,
constructing the spelled word using block letters), but their
shapes are not correctly translated into movements. If the
writing impairment is due to a failure in motor processes that
govern neuromuscular execution and control, written letters
are distorted often to the point of being unrecognizable, and
errors include misplaced as well as incorrectly formed strokes
of the pen.
More surprisingly, it is possible to observe instances of
peripheral agraphia where the dominant error consists of well-
formed letter shapes; TABLE would be written as F-A-P-L-E. The
question of interest is whether such errors are visually related
to the target or whether some other relationship exists, based
on the sequence and placement of the movements needed to
construct the letter. A question of additional import is whether
the action representation at this level of organization in the
writing mechanism concerns movements of the hand or
more generally, to any action that is intended to construct
letter forms. The answer to the second question is straight-
forward; the representation affected is sufficiently abstract
that impairment is seen even when letters are produced by
tracing the shape with a foot. The first question is more diffi-
cult. Letters that require similar strokes for their production
tend also to be visually similar; how then can we distinguish
visually based from action-based errors? A reasonable way
of proceeding is to invoke two empirically driven methods
to arrive at separate metrics of visual and motor similarity
between letters. For example, visual confusability can be
assessed by examining the errors of normal readers under
restricted viewing conditions; the higher the probability a letter
is confused with another, the more they are perceptually simi-
lar. Motor similarity can be operationally defined by construct-
ing a reasonable taxonomy involving, say, number of strokes,
shape (line vs. curve), direction (up, down, clockwise, coun-
terclockwise), and so on. In this way, it is possible to isolate
letter pairs that are more confusable with respect to an under-
lying metric of motor than visual similarity. An analysis of
errors using this methodological approach has disclosed that
in some cases of peripheral agraphia, errors are based on stroke
similarity between letters rather than visual similarity.
Peripheral agraphias offer the possibility of valuable
insights into the processes that govern the production of let-
ters, as the previous example illustrates. Additional evidence
from individual cases reveals a degree of categorical organiza-
tion imposed on the processes that translate letter identities
into movement. Impairment can affect the formation of letters
while numbers are unaffected; the letter O may be incorrectly
written, in other words, but not the number zero! Other forms
of peripheral dysgraphia disclose mechanisms that concern the
distinction between upper and lower case forms of the same
letter (allographs); patients may have severe difficulty in pro-
ducing uppercase printed letters for example, but not lowercase
cursive script. Finally, dysgraphia may involve a disconnection
between hemispheres, so that one hand no longer has access to
the representations that govern the production of letter shapes.
See also: Dyslexia; Motor Control; Neuroimaging of Dementia;
Psychology of Reading; Reading and Phonological Processing;
Semantic Memory; Spatial Perception.
Further Reading
Bub DN, Arguin M, and Lecours AR (1993) Jules Dejerine and his interpretation of pure
alexia. Brain and Language 45: 531-599.
Bub DN, Cancelliere A, and Kertesz A (1985) Whole-word and analytic translation of
spelling to sound in a nonsemantic reader. In: Patterson K, Marshall JC, and
Coltheart M (eds.) Surface Dyslexia, pp. 15-34. Hillsdale, NJ: Lawrence Erlbaum.
Bub D and Kertesz A (1982) Deep agraphia. Brain and Language 17: 146-165.
Bub D and Kertesz A (1982) Evidence for lexicographic processing in a patient with
preserved written over oral single word naming. Brain 105(4): 697-717.
Caramazza A and Miceli G (1990) The structure of graphemic representations.
Cognition 37: 243-297.
Cipolotti L, Bird CM, Glasspool DW, and Shallice T (2004) The impact of deep
dysgraphia on graphemic buffer disorders. Neurocase 10: 405-419.
Agraphia and Alexia 83
Coltheart M, Curtis B, Atkins P, and Haller M (1993) Models of reading loud:
Dual-route and parallel-distributed processing approaches. Psychological Review
100: 589-608.
Coltheart M, Patterson K, and Marshall JC (1998) Deep Dyslexia. London: Routledge,
Kegan and Paul.
Coltheart M, Rastle K, Perry L, Langdon R, and Ziegler J (2001) DRC: A dual route
cascaded model of visual word recognition and reading aloud. Psychological
Review 108: 204-256.
Costello AD and Warrington EK (1987) The dissociation of visuospatial neglect and
neglect dyslexia. Journal of Neurology, Neurosurgery and Psychiatry 50:
1110-1116.
Dehaene S (2009) Reading in the Brain: The Science and Evolution of a Human
Invention. New York: Penguin.
Fiset D, Arguin M, Bub DN, Humphreys GW, and Riddoch J (2005) How to make the
word length effect disappear In letter-by-letter dyslexia: Implications for an account
of the disorder. Psychological Science 16: 535-541.
Funnell E (2000) Case Studies in the Neuropsychology of Reading. Hove:
Psychology Press.
Geschwind N (1 965) Disconnexion syndromes in animals and man. II. Brain 88: 585-644.
Glasspool DW, Shallice T, and Cipolotti L (2006) Towards a unified process model
for graphemic buffer disorder and deep dysgraphia. Cognitive Neuropsychology
23: 479-512.
Hinton GE, Plaut DC, and Shallice T (1993) Simulating brain damage. Scientific
American 269: 76-82.
Hinton G and Shallice T (1991) Lesioning an attractor network: Investigations of
acquired dyslexia. Psychological Review 98: 74-95.
Jackson NE and Coltheart M (2001) Routes to Reading Success and Failure: Toward an
Integrated Cognitive Psychology of Atypical Reading. Ann Arbor: Sheridan.
Marshall JC and Newcombe F (1973) Patterns of paralexia: A psycholinguistic
approach. Journal of Psycholinguistic Research 2: 175-199.
Mayall K and Humphreys G (2002) Presentation and task effects on migration
errors in attentional dyslexia. Neuropsychologia 40: 1506-1515.
Plaut DC, McClelland JL, Seidenberg MS, and Patterson K (1996) Understanding
normal and impaired word reading: Compufational principles in quasi-regular
domains. Psychological Review 103: 56-115.
Plaut DC and Shallice T (1993) Deep dyslexia: A case study of connectionist
neuropsychology. Cognitive Neuropsychology 10: 377-500.
Rapp B and Caramazza A (1997) From graphemes to abstract letter shapes: Levels of
representation in written spelling. Journal of Experimental Psychology (Human
Perception and Performance) 23: 1130-1152.
Shallice T (1985) From Neuropsychology to Mental Structure. Cambridge:
Cambridge University Press.
Woollams AM, Lambon Ralph MA, Plaut DC, and Patterson K (2007) SD squared:
On the association between semantic dementia and surface dyslexia.
Psychological Review 114: 316-339.
Relevant Websites
http://www.facebook.com/video/video. php?v=5385357731 43 - Biomapping
the Brain.
http://wires.wiley.com/WileyCDA/WiresArticle/wisld-WCS68.html - Models of the
Reading Process.
http://www.nytimes.com/2005/04/04/books/04enge.html - Mysteries of the Brain
Vex a Novelist and His Hero.
http://www.facebook.com/video/video. php?v=53313221 6913 - Rewiring the Brain.
http://scienceblogs.com/cortex/2009/11/the_reading_brain_1.php - The Reading
Brain.
http://www.thepadc.com/sdud/index.php - What is Surface Dyslexia?
Alcohol: Psychosocial Effects
K J Sher, R Winograd, and A Verges, The University of Missouri and the Midwest Alcoholism Research Center, Columbia, MO, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Acute tolerance The phenomenon of alcohol having less
effect on the descending limb of the blood alcohol curve
(i.e., when blood alcohol levels are falling) than on the
ascending limb (i.e., when blood levels are increasing even)
at similar blood alcohol levels.
Alcohol myopia A prominent theory of acute alcohol effects
that posits that many of the behavioral and emotional
effects of alcohol can be explained as a consequence of the
narrowing of attention that alcohol produces. Such
narrowing results in the drinker attending to only the most
salient aspects of the environment which, depending upon
the nature of the environment, can result in a range of
emotional and behavioral extremes, both 'prized' and
'dangerous.'
Alcohol outcome expectancies Beliefs or associations
regarding the effects of alcohol on the drinker. Expectancies
can either be positive (e.g., 'drinking alcohol makes me
happy') or negative (e.g., 'I get sick when I drink'). Such
expectancies can be acquired directly via the personal effects
of alcohol or through vicarious learning. Expectancies have
been show to predict drinking behavior both cross-
sectionally and prospectively.
Allostasis A chronic adaptation to a substance or activity
that is reflected in homeostatic changes (e.g., changes in
hedonic set point) associated with tolerance, withdrawal,
and negative affective states. Allostatic changes induced by
chronic alcohol consumption are associated with
dysregulation of the hypothalamic-pituitary-adrenal (HPA)
axis and alcohol dependence.
Balanced-placebo design A type of experimental design
used to disentangle the pharmacological effects of alcohol
from those attributable to alcohol expectancies. The design,
in simplest form, experimentally manipulates both the
belief that one has consumed alcohol (i.e., expect alcohol vs.
expect no alcohol) and alcohol consumption (i.e., alcoholic
beverage vs. a control beverage). This design extends the
placebo-controlled design which merely holds expectancy
constant and, therefore, doesn't evaluate expectancy effects.
Chronic tolerance The phenomenon of individuals
experiencing reduced effects of alcohol with increasing
drinking experience and/or needing increasing doses of
alcohol in order to achieve a given effect of alcohol.
Pharmacodynamics The study of the physiological effects
of drugs on the body.
Pharmacokinetics The study of how the body absorbs,
distributes, metabolizes, and eliminates drugs.
Stimulation and sedation Two broad classes of effects on
subjective experience. Stimulating effects occur primarily at
low doses and early in a drinking episode; sedation occurs
primarily at high doses and when blood alcohol levels are
falling.
Introduction
From the beginning of recorded history, historians and anthro-
pologists have documented the prominent role of alcohol in
the daily life of humans across diverse societies. Thus, when we
examine alcohol use and its consequences, we should keep
in mind the fact that we are looking at a phenomenon that
has been a significant part of human experience for thousands
of years.
However, a scientific understanding of alcohol and its
effects is only a very recent achievement. Today, epidemiologi-
cal studies inform us about the widespread use of alcohol in
many societies; this makes it a public health concern in those
countries where it is used in excess or in hazardous ways. The
World Health Organization suggests that as many as 2 billion
people consume alcohol in the world, and that about
76.3 million have an alcohol use disorder (AUD); leading it
to be considered as one of the most prevalent mental disorders
in the world.
This article provides an overview of research on both the
acute and the chronic psychosocial effects of alcohol use. Such
effects are often highly influenced by moderating factors that
are specific to the occasion (i.e., the particular drinking con-
text) or the individual (i.e., some people tend to respond
differently than others); therefore, characterizing the effects of
alcohol use requires the consideration of a number of factors,
including those beyond the drug effect. Although a compre-
hensive description of alcohol effects is beyond the scope of
this article, we highlight the nature and determinants of some
of the most widely known psychosocial effects of alcohol,
including increased aggression, sexual disinhibition, injury,
stress reduction, and human capital accrual.
Acute and Chronic Effects of Alcohol
Acute Effects
When discussing the effects of alcohol we need to consider
both the acute effects associated with having recently con-
sumed alcohol and the chronic effects associated with a pattern
of alcohol use over time. Both the acute and the chronic ef-
fects of alcohol can be understood in terms of how alcohol acts
on the central nervous system (CNS) which is largely responsi-
ble for regulating cognition, affect, and behavior. The acute
effects of alcohol can be classified into three broad cate-
gories: (1) positively reinforcing effects (e.g., euphoria, arousal),
(2) negatively reinforcing effects (e.g., anxiety reduction, antide-
pressant effects), and (3) punishing effects (e.g., depression,
84
Alcohol: Psychosocial Effects 85
hangover, vomiting). These effects are likely to vary across indi-
viduals and contexts, they rely upon different underlying neuro-
biological processes, and are probably a combination of
pharmacological processes as well as individual alcohol expec-
tancies (for further discussion on this topic please see the section
'Beliefs and expectancies'). Controlled experimental studies are
better able to disentangle the effects of alcohol and these expec-
tancies compared to real-world environments as the pharmaco-
logical and expectancy effects are inextricably confounded in the
latter.
Most of the positive reinforcing effects tend to be largely
associated with an enhanced activity of the monoaminergic
system, that is, neurotransmitters such as dopamine, seroto-
nin, and norepinephrine. The arousing effects of alcohol can
be explained by the enhancement of these neurotransmitters,
and probably have a lot to do with the resulting increase in
talkativeness and excitement that is often seen when someone
is under the influence of alcohol. Negative reinforcing effects,
however, are most likely the result of increased activity in the
GABAergic system through the GABAa receptor in a way that is
not too different from that described for the benzodiazepines.
Research has suggested that the anxiolysis (decreased anxiety),
and possibly some antidepressant effects, are due to this increase
in GABAa activity, and could explain why some people use
alcohol as a means of reducing stress and tension. Lastly, the
immediate punishing effects of alcohol (like facial flushing,
nausea, and vomiting, for instance, which are mostly pharma-
cological in nature) represent a heterogeneous set of mechan-
isms that include ethanol metabolism, direct effects of alcohol
on the gastrointestinal system, brain mechanisms, and the
behavioral consequences of drunken excess (e.g., unintentional
injury).
Regarding the neurocognitive consequences of heavy drink-
ing episodes (i.e., drinking to intoxication), researchers are
still investigating for any evidence of significant cognitive
impairment. Psychopharmacological studies suggest that, in
the case of a single, standard binge episode, once the breath
alcohol levels have returned to zero, there is little or no linger-
ing impairment; although there are studies with rodents that
suggest that very high ethanol exposures can cause a more
persistent impairment. While researchers have yet to fully char-
acterize the relationship between a single drinking episode and
a lasting impairment in cognitive function, there are a number
of acute effects on cognition. For example, researchers have
studied the phenomenon of blackout, which is a period of
time (during a single, heavy drinking episode) during which
recall is impaired. This memory impairment can either be 'En
bloc' (when a large segment of the drinking episode was never
encoded and thus impossible to recall), or 'fragmented' (when
memories were encoded though not readily recalled unless the
person is presented with a cue which serves as a reminder).
These acute memory impairments are thought to be due to a
disruption in hippocampus activity which can be caused both
directly (by interfering with circuitry within the hippocampal
CA1 cells) and indirectly (such as inhibiting communication
between the hippocampus and the medial septum).
While blackouts may be one of the most dramatic and
recognizable acute impairments associated with excessive alco-
hol consumption, it can also acutely affect a number of cogni-
tive functions, especially those that involve effortful processing
of information, task switching, divided attention, rehearsal,
and self-regulation. A slower rate of information processing
(e.g., visual encoding) can lead to delayed reaction times, as
well as poorer performance on divided attention tasks (such as
driving, which requires the ability to engage in multiple tasks,
simultaneously). Alcohol consumption remains a major factor
in motor vehicle crashes, which are a major source of unin-
tended injury. Indeed, acute alcohol consumption is associated
with a range of unintentional injuries; with the risk of injury
dramatically increasing with an increase in alcohol levels (for a
further discussion on this topic please see the section 'Unin-
tentional and intentional injuries).
Chronic Effects
Those who drink heavily for long periods of time are likely to
experience a range of chronic changes in emotional function-
ing, which are perhaps a result of frequent and reoccurring
withdrawal symptoms. When present, withdrawal symptoms
are associated with affective disturbance, particularly anxiety,
which can lead to the initiation of a vicious cycle of continued
drinking motivated by this affective disturbance. More impor-
tantly, some mood and anxiety disorders have been found to
be induced by alcohol (i.e., substance-induced disorders), and
these resolve after a period of abstinence. Additionally,
substance-independent mood and anxiety disorders (i.e., they
do not resolve with abstinence) often have an onset that follows
the occurrence of an AUD at rates that are 'above chance expec-
tation,' thus suggesting prottacted affective consequences.
The broad term given to the changes associated with
chronic dmg or alcohol use is allostasis, which is the result
of repeated drug/alcohol challenges to the brain and their
resulting counterdirectional (i.e., compensatory, homeostatic)
responses that strengthen when such challenges are repeated
over time. In addition to providing a basis to the process of
chronic tolerance (i.e., the need for increasingly larger doses to
obtain a given effect) the allostatic perspective also provides an
explanation for chronic mood changes. Specifically, the allo-
static burden of adjusting to alcohol-induced homeostatic
changes results in a downward drift of the hedonic set point
(in the negative direction) via its effects on the hypothalamic-
pituitary-adrenal (HPA) axis which result in a chronic negative
affective state.
In addition to the chronic emotional effects, prolonged
and heavy consumption of alcohol can result in cognitive
impairment that ranges from mild to severe. The long-term
effects depend on the quantity of alcohol consumed, with greater
quantities being associated with a decrease in the chances of
recovering lost cognitive functions (for further discussion on
this topic please see the subsection 'Chronic effects' within the
section 'Individual differences in moderating effects').
In addition to these examples of neurological disease,
chronic alcohol use is associated with a broad range of physical
health and organ system problems. Even for diseases that are
not directly caused by alcohol use, the social and psychologi-
cal factors associated with AlIDs are inextricably linked
to one's overall health (e.g., noncompliance with medical
Ueatment). Irrespective of whether the relationship between
alcohol and poor health is direct or indirect, statistics indicate
that alcohol can be attributed to over 100 000 deaths per year
86 Alcohol: Psychosocial Effects
in the United States alone, making it the third leading cause of
death in this country (following tobacco use and poor diet/lack
of exercise). Approximately 15-20% of alcohol-related deaths
are a direct result of alcohol use (e.g., cirrhosis of the liver or
alcohol poisoning), while the remainder are due to disease
or injury in which alcohol plays a significant role. Although
the overall effect of alcohol on health is negative, moderate
alcohol consumption can have a salutary effect on some dis-
eases (e.g., coronary artery disease and ischemic stroke) except
perhaps in moderate-drinking, middle-aged men who are at
higher risk of heart disease and stroke, where these effects
would be expected to be greatest.
Moderators of Alcohol Effects
Drinking Occasion Specific
Dosing and limb effects
The effects of alcohol are highly dose dependent and when
discussing what alcohol can do, we need to be specific about
the dose under consideration. Very low doses, the kind of dose
often taken by moderate drinkers in the context of a meal, may
have minimal effects on mood and behavior while binge or
heavy drinking can lead to extremes in behavioral disinhibi-
tion and an impairment of cognitive and motor performance.
Second, there are considerable individual differences in how
quickly alcohol is metabolized, so that a specific administered
dose (even after taking into account body mass and gender) of
alcohol can result in very different blood and brain levels
of alcohol in these individuals, depending upon the variability
in pharmacokinetics. While most individuals eliminate alco-
hol from their system at a rate of about 0.015-0.020% h -1 (so
that it would take, on average, 4.5 h for someone to have his or
her blood alcohol concentration (BAC) go from 0.08% to
00%), there is as much as a threefold difference in the rate of
metabolism in the general population which results in individ-
ual differences in the time taken to clear alcohol from the
system.
Moreover, even with comparable blood and brain levels of
alcohol, there are considerable individual differences in the
magnitude of the physiological and behavioral responses or
the pharmacodynamics. Thus, when we seek to explain the
effect of alcohol, we must remember to consider the indi-
vidual differences that occur at the pharmacological and phys-
iological levels.
The effects of alcohol are highly dependent on the amount
consumed with higher doses being associated with greater
behavioral impairment, behavioral excesses, and a range of
acute adverse physical, behavioral, and social consequences
(for further discussion on this topic please see section 'Acute
and chronic effects of alcohol'). While there is some degree of
folklore surrounding the beverage-specific effects of alcoholic
drinks, research has demonstrated little evidence for the type of
beverage making a difference in the level or the type of intoxi-
cation (after controlling for differences in ethanol concentra-
tion). The alcohol content of beverages sold in retail outlets in
the United States can range from 2% ('near bear') to 95%
ethanol (Everclear), and what is considered a standard drink
varies from country to country. For example, in the United
States, a standard drink (equivalent to ~12oz. of beer,
1.25 oz of distilled spirits, or 5 oz. of wine) is usually defined
as 14 g of pure ethanol. However, there is less standardization
of ethanol concentration and drink size than these guidelines
imply. Beers tend to have about 5% ethanol and can have
concentrations that are more than double this amount.
Wine, which is typically around 12% ethanol, can have up to
16% or more. Distilled spirits are typically 40-50% ethanol
but can be much higher. This variability, coupled with varia-
bility in the size of the drinking vessel and pour size, can make
it difficult for individuals to gauge the number of standard
drinks they have consumed. Moreover, across the world, what
is officially considered a standard drink ranges from 6 g etha-
nol (Austria) to 19.75 g (Japan), a consideration to be mindful
of when comparing drinking statistics and behavior across
cultures.
Although a number of adjectives can be used to describe the
subjective effects of alcohol, stimulation, and sedation are two
'umbrella' terms for classifying the subjective responses. The
term stimulation refers to subjective states such as elation,
vigor, loquaciousness, sociability, and euphoria. Sedative refers
to subjective states such as sluggishness, difficulty in concen-
trating, and slowed thinking. Alcohol effects are frequently
described as biphasic, a term that is used to refer to the differ-
ences associated with the dose administered and, as discussed
below, with whether the blood alcohol level is rising (ascend-
ing limb; as when someone is initiating drinking) or declining
(descending limb; as when someone has recently consumed
alcohol but has now ceased and sufficient time has elapsed,
and alcohol is actively being eliminated from the system more
quickly than it is being absorbed). The limbs of the blood
alcohol curve correspond to the biphasic effects of alcohol.
On the ascending limb stimulation predominates, while on
the descending limb stimulation decreases and sedation begins
to predominate. While these are general reports, the magnitude
of these subjective effects tends to vary as a function of the
individual's typical drinking pattern, with heavier drinkers
being more likely to associate alcohol with stimulant effects
in general, and lighter drinkers reporting more sedative effects;
this partially explains why the heavier drinkers consume more,
as stimulation is usually more desirable than sluggishness.
The same BAC can produce different effects depending on
the limb of the curve. For example, individuals typically per-
ceive themselves as being more intoxicated and impaired, at
the same blood alcohol level, on the ascending limb than on
the descending limb. That is, one might feel much more intox-
icated at a BAC of 0.08% when his or her blood alcohol level is
rising than at the same blood alcohol level when it is falling.
However, this acute tolerance (less response on the descending
limb than on the ascending limb) to perceived intoxication can
diverge from other alcohol effects. Most critically, when blood
alcohol is rising, many individuals can perceive that their
cognitive or behavioral abilities are impaired (in the ways
discussed in section 'Acute and chronic effects of alcohol').
However, as blood alcohol begins to fall these same indivi-
duals may subjectively perceive themselves as being no longer
impaired. However, objective impairment can show much less
acute tolerance (i.e., individuals can still show high levels of
objective impairment on cognitive and motor tasks). This
combination of perceived low impairment in the presence of
objective high impairment is characteristic of the descending
Alcohol: Psychosocial Effects 87
limb and can lead the drinker to misjudge his ability to safely
engage in certain tasks (e.g., operating a motor vehicle) and is
probably a major factor in hazardous use.
Situational context
In many, if not most societies, alcohol is typically consumed
in a social context. Experimental research has shown that the
presence of other people strongly influences the effects of
alcohol. Research findings are consistent in showing that sub-
jects, in a social drinking condition, report more positive
subjective effects of alcohol than those who drink alone. More-
over, the presence of drinking companions can affect the
amount and the rate of alcohol being consumed. In fact,
research has shown that participants drinking in a social situa-
tion might consume almost twice as much as those drinking
alone. However, this effect seems to depend on the specific
composition of the group, since people in groups drink faster
than those in pairs; mixed-gender pairs drink faster than same-
sex pairs, and among males, those in all-male groups tend to
drink more than those in mixed-gender groups. At least two
different mechanisms are likely to be responsible for this pat-
tern of findings. One is the imitative modeling of other drin-
kers. A number of studies have shown that heavy drinking
models (usually the confederate of the experimenter) tend to
increase the amount of alcohol being consumed by a companion
(the experimental subject). Another potential factor is that alco-
hol modulates (via effects on the dopamine neurotransmitter
system) natural reinforcers, and the presence of others is natu-
rally rewarding for humans (especially in adolescence when
alcohol use is typically initiated), as is in a number of animals.
However, there are a number of contextual aspects of drink-
ing, beyond the presence of conspecifics, that can affect both
mood and behavior, and these can range from the physical
aspects of the drinking environment to specific interpersonal
events. A variety of influential cognitive theories have been pro-
posed to explain why there are such large situational determi-
nants of, ostensibly, pharmacological effects of alcohol. Steele
and Joseph have stated that alcohol intoxication reduces the
ability to attend to and process multiple internal and external
cues. Under these circumstances, only those situational cues that
are most immediate and salient are likely to be attended to. Such
alcohol myopia is predicted to lead to disinhibited behavior
(e.g., increased aggression or risky sexual behavior) in situations
in which the sober individual would normally experience an
inhibition. In other words, in circumstances in which a sober
person would perceive strong cues, either provoking or inhibit-
ing a certain behavior, the intoxicated individual will only pro-
cess the more immediate and salient cues that provoke the
behavior. Similarly, salient pleasant stimuli would be more
likely to improve mood while salient aversive stimuli would be
more likely to worsen mood; both happen to a greater extent in
intoxicated individuals compared to sober individuals. Other
cognitive theories focus on how alcohol intoxication affects
appraisal processes, such as, by diminishing the elaboration
and integration of new information that typically takes place
when one is confronted with a stressor. There is increasing
evidence that the cognitive mediation of alcohol on emotion,
through processes related to impaired cognitive processing, is
most applicable to situations where low and moderate doses of
alcohol are consumed. When intoxicating doses of alcohol are
consumed, it appears that alcohol directly affects the brain cir-
cuitry that is associated with reward and behavioral inhibition.
Individual Differences in Moderating Effects
Age
The acute and chronic effects of alcohol are known to vary and
have different implications when considered as a function of
the life course. Thus, in addition to considering variables such
as dosing, limb, and situational context, we need to consider
developmental factors that shape the immediate and lasting
effects of alcohol.
Acute effects
In most Western countries, self-administration of alcohol does
not begin until adolescence with rapidly escalating rates of
both any and heavy use occurring across the teen years. Animal
research has shown that adolescent rodents are less sensitive
than adults to some effects of alcohol such as sedation, hang-
over, or motor impairment, and exhibit better acute tolerance.
Moreover, repeated exposure to alcohol may further reduce
their sensitivity to these acute effects. Such diminished sensi-
tivity to some negative consequences of alcohol makes it pos-
sible for adolescents to drink relatively large amounts, even if
they are still only novice drinkers. However, adolescents are
more sensitive to the effects of alcohol on social facilitation
and social interaction, a critically important activity at this
stage of life. This combination, of diminished punishing alcohol
effects with increased reinforcing effects (coupled with relative
drinking inexperience and incomplete development of executive
cognitive functions), probably represents the key vulnerability
to hazardous and excessive alcohol use in adolescence.
The profile of alcohol effects changes over the life course.
Studies have also shown that older adults tend to be more
sensitive to the negative effects of alcohol than younger adults.
For example, older adults have been found to demonstrate an
increase in psychomotor impairment than younger adults. This
difference most likely explains, in part, the decrease in alcohol
use and AUDs that occurs with age. However, older adults
develop alcohol problems with lower amounts of alcohol
use, and in some situations, they even fail to report the
impairment caused by alcohol, which can, alternatively, be
observed by objective measures.
Chronic effects
The long-term effects of alcohol consumption on a developing
organism can be traced back to the first trimester of prenatal
development as prenatal exposure to alcohol can lead to a wide
range of behavioral effects that are manifested in childhood,
adolescence, and adulthood. Such fetal alcohol spectrum dis-
orders (FASD) can range from attention and impulsivity
problems, similar to those seen in attention deficit disorder,
to full-blown fetal alcohol syndrome (FAS), which is charac-
terized by distinctive facial dysmorphology (e.g., indistinct
groove between the upper lip and nose, small eyes), brain
abnormalities, a range of behavioral problems, and (usually)
mild retardation.
During adolescence, the brain appears to be especially sen-
sitive to the long-term effects of heavy drinking, presumably
88 Alcohol: Psychosocial Effects
because the brain is undergoing extensive remodeling during
this time with dramatic 'pruning' of synaptic connections
and the continued development of some brain areas (espe-
cially the frontal lobes). Structural and functional neuro-
imaging studies have shown that adolescents with patterns
of excessive consumption show evidence of neurological
impairment, and behavioral testing reveals that the impairment
is of such a magnitude that it has real-life consequences on
learning and self-regulation. At this point in time, neither the
levels of exposure that are necessary to induce prolonged
changes in brain function in adolescents nor the potential
moderating factors such as pre-drinking brain function and
other co-occurring substance use are well characterized.
Although it appears that in adults a level of social drinking
is not associated with persistent cognitive changes, it is well
documented that as drinking increases (e.g., exceeds ~21
drinks/week) various kinds of neurobehavioral deficits, in-
cluding poor visual-spatial coordination, abstracting perfor-
mance, verbal associative memory, and verbal fluency, can be
observed. Much of this loss in function (e.g., new word
learning) can return with the mere passage of time if drinking
is reduced or eliminated, but other functions require active
retraining (e.g., visuospatial learning, abstraction, problem-
solving). In the extreme, well-characterized dementias, such
as Korsakoff s psychosis, can occur with prolonged, excessive
drinking patterns in vulnerable individuals.
Beliefs and Expectancies
Alcohol expectancies are beliefs that people have about the effects
of alcohol. These usually refer to what an individual thinks will
be the effects of alcohol on him or herself as well as on others.
Expectancies begin to develop in childhood, and grade-school
children can differentiate alcoholic beverages from other types of
beverages and can express the knowledge that individual behav-
ior can change as a function of drinking. During childhood, most
expectancies surrounding drinking are negative but early in the
course of adolescence a tipping point is reached where increasing
positive expectancies begin to outweigh the negative expectan-
cies. These positive expectancies continue to strengthen during
adolescence, but weaken a little in early adulthood. This pattern
of change in positive expectancies is consistent with the typical
course of drinking in the general population of many cultures;
this also peaks during later adolescence and decreases in early
and middle adulthood. Although different types of expectancies
have been identified across different studies, expectancies about
reduction in tension, positive mood enhancement, social and/or
sexual facilitation, and enhanced cognitive or motor perfor-
mance, as well as expectancies for negative effects such as behav-
ioral impairment and adverse physical consequences, have been
recurrently found in a number of studies.
The strength with which an individual holds these positive
alcohol expectancies has been shown, both cross-sectionally
and prospectively, to be strongly correlated with alcohol use
(both initiation and level of use) and drinking problems. Such
alcohol expectancies have also been found to predict treatment
outcome among people in recovery for alcohol dependence.
While most expectancy research, to date, has focused on
explicit expectancies (i.e., what someone tells us verbally about
the anticipated outcome of consumption), in the recent years,
implicit expectancies, measured unobtrusively and indepen-
dently of typical explicit expectancies, have also been found
to be associated with alcohol use. For example, the Implicit
Association Test (IAT) measures the differential associations of
alcohol-related words and positive and negative adjectives.
Scores in the IAT have been shown to have some incremental
predictive power of alcohol use above and beyond the associa-
tion of explicit measures of expectancies, thus highlighting the
fact that these implicit cognitions tap those relevant alcohol-
relevant motivations that are not easily captured by what the
drinker can tell us explicitly.
Alcohol expectancies change over the course of the lifespan,
probably because the experiences that people have when con-
suming alcohol can modify their beliefs about the conse-
quences of drinking in the future. Some authors have also
suggested that expectancies may act as a self-fulfilling proph-
ecy. That is, the beliefs that an individual has about the effects
of alcohol can make that individual more susceptible to expe-
rience those effects.
In order to understand the nature of alcohol effects, it is
important to distinguish between the pharmacological effects
of alcohol and the effects attributable to alcohol expectancies.
In order to accomplish this, researchers have used the so-called
balanced-placebo design (BPD). The BPD is an experimental
design in which participants are assigned to four different
conditions: (1) subjects in the 'alcohol' condition are told
that they will drink alcohol and indeed they are given alcohol
to drink; (2) subjects in the 'placebo' condition are told that
they will drink alcohol but instead they are given a nonalco-
holic beverage to drink; (3) subjects in the antiplacebo condi-
tion are told that they will drink a nonalcoholic beverage but
are, instead, given alcohol to drink; and (4) subjects in the
control condition are told that they will drink a nonalcoholic
beverage and indeed that is what they are given. Although it is
hard to successfully implement the BPD at high doses of alco-
hol (i.e., the deceptions in the placebo and antiplacebo are
difficult to accomplish), the BPD has shown that, in some
cases, the effects that people experience after drinking alcohol
are not completely due to its the pharmacological effects, but
are due to the expectancies that they bring to a drinking occa-
sion. That is, sometimes, subjects in the placebo condition
exhibit effects that are at least as strong as those of the subjects
in the alcohol condition (for a further discussion on this topic,
please see sections on 'Aggression' and 'Sex' below). However,
it is also important to point out that the expected and the
pharmacological effects of alcohol can be in opposite direc-
tions. For example, in one study, men who believed that they
had consumed alcohol while watching an erotic film showed
an increase in sexual arousal (regardless of whether they actu-
ally received alcohol) while those who actually consumed
alcohol showed a decrease in sexual arousal, regardless of
whether they believed they had received alcohol or not. These
experimental findings lend credence to Shakespeare's observa-
tion that alcohol "provokes the desire but it takes away the
performance. Therefore much drink may be said to be an
equivocator with lechery: it makes him and it mars him; it
sets him on and it takes him off." Whether or not the placebo
effects are iso- or counterdirectional to an alcohol effect
appears to be dependent upon a number of factors. When
Alcohol: Psychosocial Effects 89
counterdirectional, it is often viewed as unmasking a prepara-
tory or compensatory adjustment to maintain homeostasis.
Personality Differences
More than 80 years ago, William McDougall conjectured that
an extraverted personality was particularly susceptible to the
effects of alcohol. Although this is consistent with clinical and
anecdotal evidence, research on personality-based differences
in alcohol sensitivity has been sparse. However, recent interest
in the neuropharmacological basis of personality has provided
a further rationale for these ideas by providing a neurobiologi-
cal foundation to the association between personality type and
alcohol sensitivity. That is, if alcohol affects the major neuro-
transmitter and hormonal systems that are thought to underlie
variations in personality, then these individual differences in
personality could reflect variation in the baseline functioning
of these systems. This baseline functioning could determine, in
part, the nature and extent of the effects of alcohol on these
systems.
This line of research has shown that some specific
personality traits are associated with a higher sensitivity to
alcohol. For example, individuals who are high on the trait
of impulsivity/disinhibition are more sensitive to the stress-
reducing properties of alcohol. With respect to overt behavior,
individuals who are high on trait hostility or low on empathy
are more likely to exhibit aggression when intoxicated. This
means that individual differences in alcohol effects are asso-
ciated with individual personality differences.
Family History of Alcoholism
Research in the last decades has confirmed the long-standing
clinical observation that AUDs tend to run in families, with
individuals who report alcoholism in a first-degree relative
being at 86% greater risk for alcohol dependence than those
without a family history of alcoholism. This phenomenon has
led to an increasing interest in the family history of alcoholism
as a predictor of alcohol consumption and AUDs, as well as in
the mechanisms through which this variable has an impact on
alcohol use.
One potential explanatory mechanism for the effect of fam-
ily history of alcoholism is its impact on the subjective
response to alcohol. Specifically, researchers have considered
the possibility that the acute effects of alcohol are different
among children of alcoholics compared to children of nonal-
coholics. The accumulated evidence shows that, on one hand,
children of alcoholics are less sensitive to the subjective intox-
icating effects of alcohol which may lead them to drink greater
amounts of alcohol, while on the other hand, they are more
sensitive to some other effects of alcohol, such as stress
response dampening and hyperreactivity due to the stimulat-
ing properties of alcohol. Newlin and Thomson have hypothe-
sized that differences in alcohol effects between the children of
alcoholics and nonalcoholics can be resolved by considering
the limb of the blood alcohol curve. Specifically, those with a
family history of alcoholism may be more likely to experience
an enhanced reinforcement from alcohol early in the course of
intoxication. Later in the drinking episode, when blood alco-
hol levels are decreasing, children of alcoholics seem to
become less sensitive to the effects of alcohol, which in this
phase of the blood alcohol curve, tend to be more punishing.
This greater sensitivity to the reinforcing effects of alcohol, and
decreased sensitivity to the negative effects of alcohol, could
lead to greater amounts of consumption with relatively fewer
aversive side effects.
Selected Important Acute and Chronic
Effects of Alcohol
Aggression
Alcohol is known to reduce inhibitions and alter perceptions
of risk. Epidemiological studies have found that alcohol use is
associated with episodes of aggression and violence. For exam-
ple, estimates on the percentage of homicides that have been
preceded by alcohol use range from 7 to 85%. Although this
range suggests a low reliability of the findings, most studies
have found that 60% or more of homicides have been com-
mitted by individuals who had consumed alcohol prior to the
offense. Also, 40-50% of violent crimes, and 30% of cases of
intimate partner violence occur under the influence of alcohol,
with greater alcohol use being associated with more severe
violence. Moreover, some prospective studies have shown
that alcohol problems, or alcohol-related disorders, predict
subsequent aggression. However, increased aggression has
also been found to predict alcohol problems in the future.
With regard to the mechanisms that explain this strong associ-
ation, studies using the BPD have, generally, found that the
effects of alcohol on aggression are primarily due to the phar-
macological effects of alcohol.
Although the facilitative effect of alcohol for the expression
of aggression is strong, there are some factors that moderate
this association. For example, although alcohol increases
aggression in both men and women, the effect is much stron-
ger in men and, often, can only be demonstrated if there is
direct provocation and sufficient dosage. Other variables, such
as personality traits, related to aggressivity and neurocognitive
executive functioning appear to moderate the effect. Thus,
although there is a clear strong association between alcohol
and aggression, some of it is due to aggressive people tending
to drink more. When there is a direct alcohol effect, it appears
to be limited to certain individuals, drinking heavily, and
under certain contexts (e.g., provocation).
Sexual Behavior
Alcohol use has been found to alter sexual behavior in both men
and women. Moreover, the consumption of alcohol is asso-
ciated with impairments in judgment and decision-making.
Findings from studies using the BPD indicate that, at least at
lower doses, the expectancies regarding the effects of alcohol on
sexuality are more powerful than the pharmacological effects of
alcohol. However, this effect has been consistently found only
in men but not in women. Although these studies provide
evidence for expectancy-based theories, there is also research
that provides evidence for the alcohol myopia theory in the
context of alcohol-sexual behavior association. Research has
90 Alcohol: Psychosocial Effects
shown that impelling (e.g., sexual arousal) or inhibiting (e.g.,
fear of contracting AIDS) cues can affect the likelihood of risky
sexual behavior. That is, whereas sober individuals can weigh
both the impelling and the inhibiting cues that are related to
risky sexual behavior, the behavior of intoxicated participants
reflects the more salient cues in their environment.
Across adolescent and adult samples, both homosexual and
heterosexual, cross-sectional data show that those who drink
more heavily are more likely to engage in risky sexual practices.
Moreover, greater levels of alcohol consumption relate to an
increase in the likelihood of contracting a sexually transmitted
infection. However, methodological limitations, and the
inherent difficulty of conducting research in this area, make
causal inferences, regarding the role of alcohol in risky sex,
extremely difficult. Furthermore, some event-based studies
(i.e., studies focusing on one particular occasion of sex, such
as the first or the last intercourse) have not been able to
replicate this association and there is good reason to suspect
that third variable explanations (e.g., sensation seeking or
impulsive personality traits and situational factors) mediate
much of this relation between alcohol consumption and
high-risk sexual behaviors; although more recent studies have
found consistency in the findings from between subjects and
the event-based studies.
Recent studies suggest that certain individual and situa-
tional characteristics may increase the likelihood of engaging
in risky sexual acts following drinking. Variables such as age
(the association between alcohol and risky sexual behavior
increases with age among men who have sex with men) and
gender (rates of unprotected sex increase with number of
drinks in men but not in women) or partner's level of alcohol
consumption (when partner has also been drinking, the asso-
ciation between alcohol use and unprotected sex becomes
stronger among women), and partner type (i.e., casual vs.
steady; such that the use of a condom is less likely in women
having sex with casual partners after drinking) may provide
important information with regard to predicting, under what
conditions and for whom there is, an association between risky
sexual decisions and the consumption of alcohol on a given
occasion.
Unintentional and Intentional Injuries
Alcohol consumption contributes to both intentional and
unintentional alcohol-related injuries. Motor vehicle crashes
are one of the most common forms of unintentional injury due
to alcohol and have resulted in 11 773 deaths in 2008. Blood
samples are positive for alcohol in approximately half of the
traffic fatalities. Injury deaths account for 45% of all alcohol-
related deaths each year, and, in addition to vehicular acci-
dents, can occur as a result of falling, fires, sports activities,
being cut, or overdosing. Findings from hospital studies sug-
gest that each episode of alcohol consumption significantly
increases one's short-term risk for injury, with a greater increase
in risk for violence-related injury. Interestingly, people who
typically drink the least have the greatest risk for injury after
alcohol consumption.
Intentional injuries from alcohol consumption can be self-
inflicted injuries, assaults, or homicides, with self-inflicted
injuries being the most common. Alcohol-related homicides
are most likely to occur with the presence of firearms (stabbing
and strangling may be just as common but are less likely to
result in death). Motivations behind these intentional injuries
can be understood through the effect of alcohol on aggression,
lowered inhibitions, or low mood. As noted repeatedly, indi-
vidual differences such as drinking context, peers, drinking
pattern, and baseline mood have a large impact on the
subsequent effects of alcohol, making the alcohol-injury asso-
ciation highly conditional upon multiple cofactors.
Coping/Stress Reduction
The concept of people drinking to cope with negative emotions
is clearly connected to the concept of self-medication. Exten-
sive research on drinking motivations has demonstrated that in
addition to enhancement (i.e., drinking to have fun), another
common reason to consume alcohol is to reduce or manage
dysphoria (i.e., drinking to cope with negative emotions).
Many people consume alcohol when they are anxious or overly
aroused, with the belief or expectation that the alcohol will
reduce their stress levels. Though alcohol is not uniformly
stress-reducing across people and situations, people who are
strongly motivated to drink, on a regular basis, for coping, have
been found to drink larger quantities of alcohol, to drink more
frequently, and experience more negative consequences as a
result of drinking than those who do not use alcohol to cope;
perhaps because they are more likely to drink at inappropri-
ate times or when negative consequences are more likely to
result - like before work or during an emotionally vulnerable
period. One explanation for people who frequently use alcohol
as a means of regulating negative emotions is that they lack
other ways of effectively coping with stress.
Human Capital Accrual
Because alcohol can cause role and cognitive impairment and
because excessive and problematic alcohol use are most preva-
lent during adolescence and young adulthood, when schooling
is a primary role, the effect of alcohol on human capital accrual
is an important issue. That is, to the extent that alcohol inter-
feres with learning and schooling, the effects of alcohol are
likely to be long lasting since the human capital that accrues
during this period of life plays a large role in determining the
economic well being of the individual over his/her life span.
When studying the effect of alcohol misuse on educational and
occupational attainment, it is important for researchers to look
at other potential explanatory variables like general deviance
and a family history of alcoholism, as these could clearly be
highly related to both the above outcomes (alcohol abuse and
poor academic or job performance). Studies that have accounted
for these factors have found that, while alcohol abuse in late
adolescence does have a modest association with academic and
vocational achievement, this association is partly attributable to
background variables such as family history, gender, parental
education, and previous (early) academic achievement. Also,
individuals who have a history of good academic performance
(in high school, for example) are most likely to experience
negative academic consequences as a result of excessive alcohol
abuse (in college). Regarding occupational attainment, research
suggests that only those with the most extreme alcohol problems
Alcohol: Psychosocial Effects 91
(early in their career) experience negative impacts in this area.
Thus, alcohol misuse can have a negative effect on educational
and occupational attainment but the relationship is not particu-
larly strong when other characteristics of the drinker are taken
into account. Indeed, there is some evidence that alcohol use, in
general, is associated with good educational attainment, proba-
bly because it facilitates engagement during college among those
who attend, which result in higher rates of degree attainment.
Alcohol use appears to be most problematic with respect to
educational and vocational achievement when use is early, exces-
sive, and persistent.
Summary
Throughout much of the world, alcohol is a frequently used
drug that has been integrated into many aspects of social life
but also has a range of effects on health. As a drug, its effects are
highly variable, both acutely and chronically, and this varia-
bility is due to many factors that relate to the way in which the
drug is used, such as individual differences among drinkers,
and the context in which it is used.
See also: Adolescence; Addictions and Adolescence; Aggression;
Caffeine; Cognition and Personality; Drugs, the Brain, and Behavior;
Risk-Taking Behavior (Young Male Syndrome); Sexual Behavior.
Further Reading
Brown SA, McGue M, Maggs J, et al. (2008) A developmental perspective on alcohol
and youths 16 to 20 years of age. Pediatrics 121: S290-S310.
Chermack ST and Giancola PR (1997) The relation between alcohol and aggression:
An integrated biopsychosocial conceptualization. Clinical Psychology Review
17: 621-649.
Cooper ML (2002) Alcohol use and risky sexual behavior among college
students and youth: Evaluating the evidence. Journal of Studies on Alcohol
14(Suppl.): 101-117.
Davis kC, George WH, Norris J, et al. (2009) Effects of alcohol and blood alcohol
concentration limb on sexual risk-taking intentions. Journal of Studies on Alcohol
and Drugs 70: 499-507.
Giancola PR (2000) Executive functioning: A conceptual framework for alcohol-related
aggression. Experimental and Clinical Psychopharmacology 8: 576-597.
Gilbertson R, Ceballos NA, Prather R, and Nixon SJ (2009) Effects of acute
alcohol consumption in older and younger adults: Perceived impairment
versus psychomotor performance. Journal of Studies on Alcohol and Drugs
70: 242-252.
Graham K, Leonard KE, Room R, et al. (1998) Current directions in research on
understanding and preventing intoxicated aggression. Addiction 93: 659-676.
Hull JG and Bond CF (1986) Social and behavioral consequences of alcohol
consumption and expectancy: A meta-analysis. Psychological Bulletin
99: 347-360.
Jones BT, Corbin W, and Fromme K (2001) A review of expectancy theory and alcohol
consumption. Addiction 96: 57-72.
Leonard KE and Blane HT (eds.) (1999) Psychological Theories of Drinking and
Alcoholism, 2nd edn. New York: Guilford Press.
Martin CS, Earleywine M, Musty RE, Perrine MW, and Swift RM (1993a) Development
and validation of the Biphasic Alcohol Effects Scale. Alcoholism: Clinical and
Experimental Research 17: 140-146.
Martin CS and Sayette MA (1993b) Experimental design in alcohol administration
research: Limitations and alternatives in the manipulation of dosage-set. Journal of
Studies on Alcohol 54: 750-761 .
Newlin DB and Thomson J (1990) Alcohol challenge studies in sons of alcoholics:
A critical review and analysis. Psychological Bulletin 108: 383-402.
Parsons 0A and Nixon SJ (1998) Cognitive functioning in sober social
drinkers: A review of the research since 1986. Journal of Studies on Alcohol
59: 180-190.
Sayette MA, Reichle ED, and Schooler JW (2009) Lost in the sauce: The effects of
alcohol on mind wandering. Psychological Science 20: 747-752.
Sher KJ and Grekin ER (2007) Alcohol and affect regulation. In: Gross J (ed.)
Handbook of Emotion Regulation, pp. 560-580. New York: Guilford.
Sher KJ and Wood MD (2005a) Subjective effects of alcohol II: Individual differences.
In: Earleywine M (ed.) Mind Altering Drugs: Scientific Evidence for Subjective
Experience, pp. 135-153. New York: Oxford.
Sher KJ, Wood MD, Richardson AE, and Jackson KM (2005b) Subjective effects of
alcohol I: Effects of the drink and drinking context. In: Earleywine M (ed.) Mind
Altering Drugs: Scientific Evidence for Subjective Experience, pp. 86-1 34.
New York: Oxford.
Spear L (2000) Modeling adolescent development and alcohol use in animals.
Alcohol Research and Health 24: 1 1 5-1 23.
Steele C and Josephs R (1990) Alcohol myopia: Its prized and dangerous effects.
American Psychologist 45: 921-933.
Sullivan EV, Harris RA, and Pfefferbaum A (2010) Alcohol’s effects on brain and
behavior . Alcohol Research and Health 33: 127-143.
Wiers RW and Stacy AW (eds.) (2006) Handbook of Implicit Cognition in Addiction.
Newbury Park, CA: SAGE.
World Health Organization (2004) Global Status Report on Alcohol 2004. Geneva:
Author.
Relevant Websites
http://www2.scc.rutgers.edu/alcohol_studies/alcohol/index.htm - Alcohol Studies
Database.
http://www.apa.org/divisions/div50/-American Psychological Association Division on
Addictions.
http://alcoholstudies.rutgers.edu/ - Center of Alcohol Studies.
http://www.alcoholdependence.org/ - Midwest Alcoholism Research Center.
http://monitoringthefuture.org/ - Monitoring the Future.
http://www.niaaa.nih.gov - National Institute on Alcohol Abuse and Alcoholism.
http://www.rsoa.org/ - Research Society on Alcoholism.
http://www.casacolumbia.org/ - The National Center on Addiction and Substance
Abuse at Columbia University.
Alexithymia
V Panaite and L M Bylsma, University of South Florida, Tampa, FL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Alexithymia Inability to describe emotions in a verbal
manner.
Construct An abstract collection of ideas that serves as an
explanatory concept.
Dysphoria A negative mood state characterized by
prolonged bouts of sadness.
Emotion regulation The process through which emotional
arousal is redirected, controlled, or modified to facilitate
functioning.
Empirical Controlled observation and measurement.
Validity Refers to the extent to which what was observed is
the same as what was claimed to be observed.
History
The alexithymia constmct has a history that goes back about
40 years. The term alexithymia came from the Greek phrase
a lexis thymos, which literally means having no words for emo-
tions; the term, however, was initially used in its current form
by Sifneos, a psychoanalyst who coined it in 1973. Some of the
early insights we have about alexithymia came from observa-
tions made by clinicians in the late 1940s, such as Ruesch and
MacLean, who described patients struggling to express emo-
tions; these patients seemed to channel their feelings inward
rather than outward through expressive behavior.
Following these initial descriptions, additional reports of
patients with alexithymia, who were unresponsive to psycho-
analysis, became known. Psychoanalysts Homey and Kelman
described their patients as having a limited internal emotional
activity, including a decreased awareness of feelings, lack of
emotional experience, and what seemed like a lack of creativity
and openness. La pensee operatoire (or operational thinking)
was the next step in conceptualizing alexithymia. Marty,
deM'Uzan, and David described patients with psychosomatic
illnesses who, while being treated with psychoanalysis, were
less aware of their inner worlds and more preoccupied with
their physical experiences.
The information that surfaced over time from these observa-
tions became significant once Sifneos and Nemiah's systematic
investigations and descriptions of patients with psychosomatic
illnesses culminated in alexithymia becoming a formalized
construct. Specifically, they described patients with alexithymia
as having a 'marked difficulty describing subjective feelings and
a communicative style characterized by a preoccupation with
minute details of external events and a paucity or absence of
drive related fantasies' (p. 28).
Much of the initial evidence for the constmct of alexithymia
came from observations of patients with psychosomatic ill-
nesses. Continuing this trend, Krystal and Raskin and Brusch
reported characteristics similar to those described earlier in
patients with other psychiatric illnesses, such as posttraumatic
disorders and anorexia nervosa. Ever since, considerable
research has been conducted, bringing support for this con-
struct. The operational definition of the alexithymia constmct
has changed little over the years (e.g., La pensee operatoire is now
maintained under the facet of externally oriented thinking),
although the perspective from which we describe this constmct
has moved from one of a psychodynamic nature to a view
focused on interconnections between alexithymia and both
physical and mental health. The development of scales and
of studies looking at the relationship between alexithymia
and other constructs has afforded a more in-depth understand-
ing of this concept. This has brought about the development of
theoretical subtypes of alexithymia, as well as a clearer perspec-
tive on its relationship with both physical and psychological
health.
Measurement of Alexithymia
The field of alexithymia measurement had been initially
concerned with validation of the construct alexithymia itself.
In this process, several measures were developed and investi-
gated. Available measures for alexithymia vary widely in their
presentation and focus. While historically this area focused on
projective techniques, such as TAT, Rorschach, or the Objec-
tively Scored Archetypal Test, such instmments have become
obsolete with the development of other scales for measuring
alexithymia.
More recently, a new self-report scale, the Toronto Alexithy-
mia Scale (TAS), and its revised version, the 20-item Toronto
Alexithymia Scale (TAS-20), have brought support for the alex-
ithymia construct with a more robust psychometric profile.
TAS-20 is the latest version of the scale, and it has been
shown to have good internal consistency and test-retest reli-
ability. Factor analytic investigations brought up the three
factors supporting the alexithymia constmct: (1) difficulty in
identifying feelings, (2) difficulty in verbalizing emotions, and
(3) externally oriented cognitions.
Haviland and colleagues developed the Observer Alexithy-
mia Scale (OAS). The 33-item scale was designed for observers
(e.g., family, friends) to report on the alexithymia features of a
patient. The items can be divided into five subscales: distant,
uninsightful, somatizing, humorless, and rigid. The scale and
its subscales have been psychometrically investigated and good
reliability and validity were found for clinical and nonclinical
populations. The scale was also found to be a useful tool in
differentiating between people with clinical and nonclinical
levels of alexithymia.
92
Alexithymia 93
One measure, Levels of Emotional Awareness Scale (LEAS),
was proposed as a valid measure of level of emotional aware-
ness. Although not specifically created to measure alexithymia,
it was proposed by Lane and colleagues as an alternative
conceptualization of alexithymia - a topic developed in more
detail below. Unlike the previously presented measures, LEAS
uses descriptions of emotion-evoking situations to which
subjects are asked to respond by describing emotional reac-
tions from both personal and other perspectives. LEAS has
empirical support for good internal consistency and inter-
rater reliability.
One of the most recent self-report measures was devel-
oped by Vorst and Bermond: the 40-item Bermond-Vorst
Alexithymia Questionnaire (BVAQ). In the scale development,
alexithymia was conceptualized as having five facets. Using
the original TAS scale as a model, the five factors of the
BVAQ are emotionalizing, or the incapacity to experience emo-
tions; fantasizing, or having a poor fantasy world; identifying/
differentiating emotions; analyzing or showing low levels
of emotional analysis; and verbalizing emotions. Validity and
reliability were investigated initially through two studies using
samples from three different countries. Convergent validity
was established by showing correlations between BVAQ and
TAS-20.
Related Constructs
Although alexithymia has been conceptualized as a distinct
construct, its independence has been questioned. Several
other constructs have been proposed as potentially overlap-
ping with alexithymia. This has raised the question of whether
alexithymia is a unique construct or whether it can be fully
explained by combinations of other existing constructs.
Given alexithymia's conceptualization as a personality trait,
its relationship with other personality traits, particularly neu-
roticism, extraversion, and openness, has been investigated.
While positive correlations between alexithymia and neuroti-
cism and a negative correlation with extraversion and openness
were indicative of the relationship between all these traits,
important differences also surfaced. For example, the ease of
emotional expressivity of introverts given their introspective
nature is not common to people high in alexithymia. This
feature sets introverts and alexithymics apart given the diffi-
culty with which people with alexithymia express emotion.
In a similar vein, neuroticism is connected to strong emo-
tionality, which is one of the core deficits in alexithymia; low
scores on openness related to externally oriented thinking and
low levels of openness to feelings in alexithymia.
Other constructs that have been proposed are repressive
coping style, inhibition, and big five personality dimensions.
Bonnano and Singer originally proposed that alexithymia may
be part of or connected to the repressive coping style identified
by defensiveness. Despite defensiveness being in line with the
emotional dysregulation found in people with alexithymia,
their relationship was not detected by subsequent research.
King found that inhibition and alexithymia are positively cor-
related, although while the former entails control, the latter is
primarily conceptualized as a deficiency. Luminet and collea-
gues also supported the proposition that alexithymia may be
related to other personality traits. However, while alexithymia
was related to a certain degree to traits such as proneness to
depression and negatively related to other traits such as prone-
ness to positive emotions or openness to feelings, it was not
entirely represented by one trait and so, finally, investigators
concluded that the construct of alexithymia is distinguishable
from these other traits.
Constructs that have raised further questions about alex-
ithymia's independence are somatization, dysphoria, and
depression. Somatization and, more specifically, somatosen-
sory amplification - the tendency to focus on bodily responses
to emotional stimuli - have been associated with alexithymic
individuals. However, somatization does not account for all
the clinical facets that create alexithymia. Both somatization
and dysphoria have been conceptualized as potential out-
comes of emotional dysregulation. High correlations between
depression and alexithymia have raised questions about the
latter being a part of depression. Although, as noted earlier,
while depression may exacerbate alexithymia traits, alexithy-
mia tends to remain stable even once depression subsides. In
addition, Parker and colleagues found that by factor analyzing
the Beck Depression Inventory (BDI) and TAS together, they
loaded on separate orthogonal factors, further supporting the
independence of the constructs of depression and alexithymia.
Asperger's and Autism spectrum disorders more recently
have gained general interest and have come to alexithymia
researchers' attention, given the similarities in the lack of emo-
tional expressivity between these constructs. However, empiri-
cal work directly examining the relationship between the two is
still sparse. A letter to the editor makes a compelling argument
for the importance of investigating these constructs in connec-
tion with alexithymia given extensive similarities. The authors
talk about similarities in the lack of social interest exhibited by
people with either of these disorders. People with alexithymia
and Asperger's show a lack of interest and pleasure from inter-
acting with others, and interactions are generally detached.
Another central similarity noted by the authors is the poor
emotional understanding and functioning, among others.
However, research investigating similarities between these con-
structs has yet to fully develop, although strides are being made
in this direction.
Looking at the relationship between alexithymia and other
constructs has been an important part in the process of validat-
ing alexithymia as a distinct construct. While clinical and
empirical evidence points to the similarity and some overlap
between alexithymia and other personality traits (i.e., neuroti-
cism) and mental disorders (i.e., depression), the body of work
that exists to date has provided the basis to speak of alexithy-
mia as a distinct constmct.
Cognitive Aspects
One core characteristic of alexithymia is the difficulty in
identifying and describing one's feelings. Lane and Schwartz
created a cognitive development model for emotion on the
basis of Piaget's theories of cognitive development. They pro-
posed that this ability to identify and describe feelings develops
over time just as cognition does. The cognitive development
model as applied to emotion has five stages of emotional
94 Alexithymia
experience, starting with physical sensations, action tendencies,
single emotions, blends of emotion, and blends of emotional
experience. Additionally, the level of emotional development
also depends on what and how much people know about
emotions.
Research shows that people who score high in alexithymia
not only have difficulty in verbal and nonverbal emotion
recognition but also show impaired recognition of facial
expressions. Cognitive deficits in alexithymia have also been
proposed to potentially be adaptive, to the extent that people
capable of a complex cognitive understanding of emotions
also being more likely to potentially be more impacted by
extreme emotional experiences (i.e., traumatic experiences).
Given that alexithymia has often been observed in people
with mental or physical illnesses that are likely to be associated
with increased emotional distress, such a trait can be protec-
tive in the short term, although marked deficits in emotional
processing most often have been shown to lead to protracted
anhedonia and apathy.
Physiological Aspects
As alluded to earlier, the relationship between alexithymia
and psychosomatic disorders has a long history with a con-
siderable amount of literature supporting it. An important
subsection of this research has focused on aspects of physiol-
ogy in people who have high alexithymia scores. It has been
proposed that alexithymia is related to a higher baseline
arousal. This was viewed as a dysregulation of the autonomic
nervous system and system linked to pervasive dysphoria in
people high in alexithymia and various other mental and
physical disorders. Anxiety disorders, such as panic disorder
and posttraumatic stress disorder (PTSD), depression and
personality disorders, eating and substance abuse disorders,
coronary heart disease, diabetes mellitus, and hypertension,
all have been seen as related to the deleterious effect of a
prolonged state of arousal.
Initially, people high in alexithymia were also thought to
experience higher reactivity to emotional stimuli compared
to healthy people. Three different theories have been proposed
to explain the increased physiological arousal in response to
emotional stimuli: Papciak and colleagues' decoupling hy-
pothesis, Martin and Pihl's stress hypothesis, and Cacioppo
and colleagues' discharge theory. The decoupling hypothesis
proposes that in people high in alexithymia there is a discon-
nect between emotional experience and physiological auto-
matic response when exposed to stress that results in high
somatic response but lack of emotional expression. The stress
hypothesis proposes that through the inability of alexithymic
people to identify emotion and emotionally charged stimuli,
they are also unable to identify stressful situations and, there-
fore, are not signaled to engage adaptive emotional regulatory
tactics, leaving them vulnerable to experience more stressors,
or more prolonged stress. Discharge theory views emotional
expression as the outward channeling of energy. Alexithymia
is characterized by an inward emotional expression through
increased activation of the nervous system, given the deficit of
individuals high in alexithymia to direct expression of emotion
outward through speech or emotional behavior.
More recently, in contrast to previous theories, Linden and
colleagues proposed the hypoarousal theory, which suggests
that people high in the alexithymia trait will react less to
emotional stimuli than other individuals. Evidence for the
physiological aspects of alexithymia is still equivocal, although
more recent studies have found more considerable support for
the theory that people with alexithymia experience low physi-
ological arousal to emotional stimuli. This is perhaps a direct
result of decreased awareness of emotions in people high in
alexithymia.
Neurological Aspects
Alexithymia is primarily conceptualized as a personality trait,
and personality theories emphasize genetic attributes as impor-
tant contributors to personality development. Research and
theory have proposed neural mechanisms as propellers of
what constitutes personality. In explaining alexithymia, two
main views have crystallized to date, although much is left
unknown about this area. One view proposes that alexithymia
is a result of poor communication between the left and right
hemispheres of the brain. This view is supported by a body
of research in split-brain patients. What investigators have
found is that such patients exhibit traits similar to those
that describe people high in alexithymia: difficulty in recog-
nizing and describing emotions. This view has also been
investigated among patients whose brains were intact, by
looking at the efficiency or deficit in interhemispheric com-
munication. It was observed that while the transfer of linguis-
tic information was similar among alexithymic and normal
individuals, the two differed in the processing of nonlin-
guistic information, those high in alexithymia showing a
deficit in interhemispheric communication. This is meaning-
ful in interpreting the lack of recognition of emotional faces
in people high in alexithymia.
Another view proposes that the difficulty in emotional
processing experienced by people with alexithymia is due pri-
marily to a dysfunction in the right hemisphere of the brain.
This view has been supported by observations of patients with
damage to this part of the brain. The right hemisphere has been
found to have a role in information processing about emotion
of facial expression. This observation has often been documen-
ted in people with depression, for example, who have similar
difficulties with emotional recognition, which potentially
points at one explanation for the relationship between depres-
sion and alexithymia. But the role of emotional processing by
the right hemisphere goes beyond that of recognition of facial
expression, to facilitating information processing of other
stimuli with an emotional characteristic. Without a capacity
to process different emotional stimuli, a poverty in experience
with emotion often occurs.
Two specific areas of the brain that have been theoretically
and empirically connected to emotional processing are
the amygdala and the hippocampus, both of which have
been investigated using diverse methods (i.e., fMRI, PET,
EEG) . The amygdala has been primarily connected to giving
emotional meaning to different sensory information, thus
helping individuals integrate experience and memories with
an emotional value. The amygdala has also been connected
Alexithymia 95
to emotional behaviors, such as facial expressions. All of
these pieces seem to be interconnected in that individuals'
experience of an event is determined by the emotional valence
attributed to the event, which may dependent on the emo-
tional behavioral experience. The hippocampus, on the other
hand, has an important role not only in emotional proces-
sing, but also in memory, which may be important in creating
the emotional schemata that individuals use in expediting
information processing. It becomes apparent that a dysfunc-
tion in either of these areas can impact the emotional devel-
opment outlined in Lane and Schwartz's model.
The findings of these different studies attest to the complex-
ity and breadth of alexithymia. The studies looking at the
potentially poor interhemispheric communication bring sup-
port for the lack of verbal recognition and expression of emo-
tion. Poor emotional experience of people high in alexithymia
is mirrored by dysfunction in the right hemisphere and the
amygdala corresponds to a low behavioral emotional expres-
siveness. The hippocampus was connected to emotional mem-
ories that may connect further to a poor fantasy world in
people with alexithymia.
Alexithymia and Emotional Expression/Regulation
Given that alexithymia has been conceptualized as involv-
ing deficits in emotion recognition and expression, one
line of research that has looked at emotional expression in
alexithymia has been research on crying behavior. Crying,
like other emotional behaviors, is sparse among people
with alexithymia, possibly due to an unpleasant arousal
during crying. Rottenberg and colleagues also found a con-
nection between alexithymia and increased negative affec-
tivity after a crying episode, potentially further supporting
the lack of emotional recognition as well as identification
of potential roots to the negative affect among people high
in alexithymia.
As a consequence of the decreased ability to express and
recognize one's emotions, emotion regulation is a major prob-
lem in people high in alexithymia. Unable to recognize their
emotions, they often mislabel physical symptoms as illness
rather than emotional responses to stimuli. Individuals with
alexithymia are rarely able to intervene and regulate these
emotions. Additionally, people high in alexithymia traits are
poor evaluators of others' emotions and so, rarely offer
empathic support that would in turn create the social support
often employed in adaptive emotion regulation. Alexithymia
has also been connected to compulsive behaviors, such as
binge eating and drinking, or starvation in anorexia nervosa,
all potential indicators of dysfunctional regulatory attempts.
An inability to recognize emotions among people high
in alexithymia can potentially interfere with the identification
of the source of emotions for adequate problem solving and
downregulation of negative emotions. Consequently, people
with alexithymia may be prone to experiencing more negative
emotion and protracted dysphoria may interfere with the
experiencing of positive emotions. Poor emotional processing
interferes, not only with personal experience, but also with
interpersonal affective exchanges that may lead to poor social
experiences.
Alexithymia and Physical Health
A significant body of literature on alexithymia proposes
and supports its connections to physical health. Lumley and
colleagues wrote a critical review of the literature where they
established a model of the connections between physical
health and alexithymia. One of the proposed links is that
alexithymia is an antecedent to somatic illnesses. Lumley pro-
poses two potential mechanisms by which alexithymia may
cause physical illnesses: through physiological changes and
unhealthy behaviors. Empirical studies have found thus far
that alexithymia is related to a low immune function and,
consequently, connected to diseases with an immune compo-
nent, such as precancerous deficits. There also seems to be a
connection between cardiac activity and alexithymia. People
with alexithymia have been found to have an increased
resting physiological arousal, although high reactivity to
stressors has yet to be found. On the contrary, attenuated
reactivity as shown by a low RSA fluctuation has been seen in
people with alexithymia. A low immune function, high resting
arousal, and low reactivity can impact physical health by
creating vulnerability for somatic disease. People high in alex-
ithymia have also been found to engage in a variety of
unhealthy behaviors, such as unhealthy eating, smoking,
drinking, drug abuse, and gambling as well as unhealthy nutri-
tion and lifestyle in general.
According to Lumley and colleagues, 'illness behavior,'
which was conceptualized as somatization and health care
seeking, is another pathway between alexithymia and physical
illness. Research studies have found a relationship between
alexithymia and symptom endorsement, without any objective
evidence for illness that would warrant the experience of those
symptoms. So, in short, alexithymia seems to be connected
to the subjective experience of illness-like symptoms, but not
to the objective illness. It seems likely that this subjective
experience of different symptomatology prompts people with
alexithymia to become more engaged in seeking health care
services, which is what some empirical studies have shown.
Interestingly, however, this finding seemed to be highly
dependent on the specific facet of alexithymia for which an
individual is elevated. So, Lumley concludes that, at least given
what we know now, people with alexithymia who have diffi-
culty in identifying and describing emotions, rather than
those highly externally oriented, may have a higher awareness
of their bodies and be more likely to identify and seek help on
the basis of subjective experiences of illness.
Another proposed pathway between alexithymia and phys-
ical illness that Lumley and colleagues identify in their model
is physical illness as an antecedent of alexithymia or, as they
call it, 'secondary alexithymia.' In support of this possibility,
the authors present empirical evidence from literature primar-
ily on stress and trauma that have been shown to result in
outcomes similar to qualities of alexithymia. However, the
authors recognize that given the strong evidence for alexithy-
mia being a stable trait, it is more plausible that it is an
antecedent. Although, given the scarcity of this literature, it is
plausible that 'secondary alexithymia' is in the realm of possi-
bilities and this area would benefit from longitudinal studies
following healthy people at risk for future trauma or stress
(e.g., soldiers).
96 Alexithymia
One last arm proposed by Lumley and colleagues in
the model connecting alexithymia and physical illness intro-
duces a third factor as the connecting antecedent of the
two. While there is some literature, primarily on depression
and other negative affect constructs, the overall status of the
evidence currently is that the 'third variable' pathway is
not supported. It is possible that with further studies examin-
ing genetic and temperamental third variables, they may sur-
face as potential connectors between alexithymia and physical
illness.
In conclusion, while there is literature supporting the dif-
ferent proposed pathways, the bottom line is that much of the
literature on alexithymia and physical health seems to be
mostly based on cross-sectional and correlational studies.
While this evidence is an adequate start up point, further
investigation using more complex models would be warranted
to further the state of this area.
Alexithymia and Mental Health
It is often the case that extreme or chronic stress and trauma
result in mental and emotional disorders that are characterized
by deficits in affect regulation, and hence have been identified
with alexithymia. Two areas relevant to alexithymia that have
received extensive attention and empirical support are anxiety
and mood disorders. The symptoms associated with these dis-
orders have been conceptualized as potentially adaptive in
normative samples in the form of signals of a negative event
that is either about to happen (i.e., anxiety-like symptoms) or
that has happened (i.e., sadness due to loss). At clinical levels,
these symptoms often indicate a dysregulation of affect due to
an overwhelming overpowering by emotions resulting from
extreme or traumatic events.
Anxiety disorders that have been particularly viewed as
portraying a profound deficiency in emotion regulation are
panic disorder and PTSD. In people with panic disorder, the
focus on bodily sensations and symptoms becomes excessive
as they become tuned in with these feelings in an attempt to
recognize when the next attack will happen. Because of this
high focus on physical arousal and low capability to cogni-
tively regulate these feelings, alexithymia is highly prevalent
among people with panic disorder. Similarly, there is con-
siderable overlap between the presentation of PTSD and alex-
ithymia, including rich evidence for a relationship between
alexithymia and childhood abuse or adult trauma. Moorman
and colleagues developed the Reality Escape Model to explain,
among others, the role of alexithymia in victims of childhood
sexual abuse: the manifestation of a 'nonfeeling state that is
so characteristic of traumatized individuals' (p. 82). In this
model, it is thought that alexithymia victims of sexual abuse
are able to switch off negative feelings and emotions that are
too unbearable to deal with in order to cope with the traumatic
events that often are a repeat offense. Unfortunately, as the
authors note, the impact of such a manner to 'regulate' emo-
tions is farther reaching, potentially affecting multiple areas
and resulting in a generalized state of anhedonia that has
been connected back to alexithymia. Abuse in childhood,
according to these authors, is particularly deleterious and likely
to be connected to alexithymia due to its timing in a period
during development when personality formation may be most
vulnerable. Interestingly enough, alexithymia in adults who
have experienced childhood sexual abuse seems to be of a
certain kind: experientially, they seem to be unharmed, but
cognitively they are incapable of recognizing and accepting
emotions, making these adults good candidates and poten-
tially rather responsive to cognitive interventions.
The connection between depression and alexithymia is
less obvious, given that depression is often conceptualized as
a disorder of mood, characterized by intense feelings of
sadness, guilt, and excessive ruminative negative cognitions.
However, these symptoms are not necessary for a diagnosis of
depression. In western cultures, for example, a depression that
has marked anhedonia as the primary symptom is clinically
recognized. In non-Western cultures, a focus on negative affec-
tivity is not the norm. Somatic symptoms rather form the
construct of depression in these cultures and are often misla-
beled and missed altogether, especially when somatization is
coupled with a culturally encouraged lack of emotional expres-
sion. Both anhedonia and somatization have been connected
to alexithymia and draw the link between alexithymia and
depression.
A substantial body of research has also been developed on
the relationship between alexithymia and substance abuse,
also characterized by emotional dysregulation. Empirical stud-
ies have shown that about 50% of people with clinical levels of
substance abuse (primarily alcoholism) score in the clinical
range of alexithymia scale. It has been proposed that the pri-
mary link between alexithymia and substance abuse disorders
is emotional dysregulation. Studies have shown alexithymia to
be a stable trait in alcoholics; however, longitudinal studies
with younger populations have been proposed to further sup-
port this contention. Given that many alcoholics have battled
their condition for long periods of time, this may have enabled
changes in personality and physiology that are similar to those
observed in people with alexithymia.
Another area of psychopathology that has been related to
alexithymia is eating disorders, also through a deficit in emo-
tion regulation. An additional aspect that links alexithymia to
eating disorders is a difficulty in people with various eating
disorders to identify or recognize bodily signals (e.g., those
that would signal hunger) and the difficulty to recognize
emotions in people with alexithymia. Empirical studies have
shown that these two deficiencies correlate closely, and that
about 50-60% of people identified as having an eating disor-
der will also meet the criteria for alexithymia. Similar to other
psychopathologies, alexithymia was found to be a stable trait
in people with eating disorders, although alternative explana-
tions have been proposed for this finding, such as comorbid
psychopathology and continuation of subclinical levels of the
eating disorders.
Treatment
Historically, given the initial popularity of psychodynamic
therapy, it was the treatment of choice when the first patients
with alexithymia were observed and described by clinicians
such as Sifneos and Nemiah. However, it was observed that
such therapies were not adequate given the reliance on free
Alexithymia 97
verbalization of emotions and internal processes, which go
counter to what has been observed in people with alexithymia.
Alternative treatment recommendations, in the past, have
included therapies focused on support. Other therapies that
were considered to be more efficacious were therapies that
would focus on teaching how to communicate emotions rather
than what should be communicated.
Many therapies for people with alexithymia are mainly
educational in nature, teaching these patients skills such as
relaxation techniques to help regulate emotions that are often
mislabeled as physical symptoms. Although most therapies for
alexithymia target bodily sensations and emotion identifica-
tion, they also incorporate cognitive behavioral techniques
meant to help attend to physiological symptoms and under-
stand them from an emotional and cognitive perspective.
Additionally, teaching patients that physical symptoms are
time limited has also been successful in helping people with
alexithymia. Another psychoeducational technique was devel-
oped by Ronald Levant for the treatment of what he called
'normative male alexithymia' to help men express their emo-
tions verbally.
Despite showing a deficit in cognitive and affective under-
standing of emotions, alexithymia was found to have positive
effects on treatment outcomes in cognitive behavioral thera-
pies. The reasoning offered was that people high in alexithymia
are more inclined to use specific cognitive styles to manipulate
information. Consequently, therapies that target these skills
and put less emphasis on affect recognition and expression
would better fit these individuals.
Group therapy has been proposed as an option for treating
patients with alexithymia, as it affords interaction with other
people and direct observation of emotional expression. It is
important to note that people with alexithymia are less emo-
tionally responsive and so may provide little emotional
support to those in a group therapy. Empirically, group ther-
apy has been evaluated in people suffering from a physical
illness (e.g., heart disease) who were also high in alexithymia.
Group therapy, in this context, was found to be beneficial in
lowering both TAS scores as well as experience of a variety of
cardiac events. Given the connection between alexithymia and
physical health, it is difficult to disentangle the impact one
outcome can have on the other; however, it is hoped that
group therapy will be a viable option for people with alex-
ithymia. The group setting affords observation of emotional
expression as well as room to practice emotional expression
with other peers. It is also important to note that improve-
ments in physical health in conjunction with decreases in TAS
scores further reinforce the connection between alexithymia
and physical illness.
Despite the use of pharmacotherapy for the treatment
of emotion regulation dysfunction in many disorders, psycho-
tropic medications have not been evaluated in patients with
alexithymia. However, some studies have looked at the impact
of alexithymia on antidepressant treatment of depressed
patients showing that alexithymia was related to negative out-
comes of treatment. It is, however, possible that people with
alexithymia did not experience an increase in adaptive emo-
tional regulatory behaviors with the decrease in low mood,
as people without alexithymia may experience with treatment
for depression.
Although treatment has not been a focus of research in
alexithymia, what has been done so far showed overall encour-
aging outcomes: CBT and group therapy, in particular, seem to
have shown substantial impact in patients with alexithymia,
especially in areas that can potentially be life threatening, such
as cardiovascular disease. Although the importance of research
on treatment for people with alexithymia has been acknowl-
edged and commenced, a long road is still ahead of us in this
domain and further attention needs to be directed toward
finding empirically supported interventions.
Controversial Issues
Alexithymia Types
There is a limited body of research that has proposed and sup-
ported the suggestion that there are two distinct types of alex-
ithymia. Type 1 is considered alexithymia characterized by poor
affective and cognitive emotional experience and it presents with
distinctive behavioral patterns that set it apart from healthy
typologies of behavior. One core aspect of type 1 alexithymia
relates to social interaction. People with this type of alexithymia
are aversive to what they may perceive as forced social interac-
tions and intrusive social behaviors of others. Some have
described this pattern of behaviors as a fear of enmeshment,
when one's identity becomes intertwined with that of another,
making it indistinct on its own. This becomes particularly mean-
ingful for both treatment and understanding of further ramifica-
tions of this trait, as people with this type of alexithymia will not
rely on social support as a coping mechanism. Type 1 describes
the prototypical person with alexithymia.
Type 2 alexithymia is considered a lower level alexithymia:
while it is characterized by a similar deficiency in cognitive
emotional awareness as type 1, the emotional experience is
present. People with this trait have a low sense of self and,
while also retracted from social engagement, it is primarily
because of a sense of social inadequacy and insecurity, where
feelings of failure are predominant. Although through a differ-
ent underlying mechanism, just like those with type 1 alexithy-
mia, people with the type 2 trait exhibit similarly poor social
coping skills that revolve around abandonment fears and
attempts at avoiding it. Type 2 alexithymia has been proposed
to primarily identify victims of sexual abuse.
The description of these types of alexithymia affords a
look at the full range of characteristics of people high in alex-
ithymia. It becomes clear that social engagement is possibly
central in learning emotional coping and emotional under-
standing, both of which are core aspects that are missing
in alexithymia. However, this area is still controversial given
that research on the different types has been limited to certain
populations. Further research may be needed for the generali-
zation of these traits.
Trait Versus State Alexithymia
Alexithymia has been proposed as a personality dimension,
and validation of the construct comes through the work of
Taylor and colleagues on the TAS and TAS-20. Both of these
scales have been validated by examining their relationships
with scales that would tap into deficits observed among people
98 Alexithymia
with alexithymia (e.g., access to one's feelings). Concomi-
tantly, there are studies that support the idea that alexithymia
could also be a state reaction to different mental and physical
conditions. Longitudinal studies have looked at the stability of
alexithymia in the context of different levels of stress. Overall,
these studies supported the trait view by demonstrating the
long-term stability of alexithymia, thus supporting Taylor's
conceptualization of alexithymia as a stable personality trait.
In contrast, studies on clinical populations have suggested that
when connected to psychopathology, alexithymia is viewed as
state dependent, or as an outcome of the experienced vulnera-
bility during stressful situations that subsides with the allevia-
tion of symptoms.
To reconcile these two views, it was proposed that when
speaking of personality traits a distinction between relative and
absolute stability is warranted. Absolute stability is reached if
people with alexithymia score the same on an alexithymia scale
over time. Relative stability refers to temporal stability or
people's relative position on levels of alexithymia, even though
their absolute score may change. Similar trends are seen in
other personality dimensions such as neuroticism and extra-
version. Saarijarvia and colleagues offer empirical support for
the idea that the two views don't have to be mutually exclusive,
but that they can complement each other for a more complex
understanding of alexithymia. In their study, while alexithymia
scores decreased as depression decreased, overall, subjects
remained high in this trait even after reduction of depression.
In conclusion, a temporary increase in negative affect may
exacerbate alexithymia traits, but this change was not found
to explain the entire trait in any given person; hence, those
scoring high on alexithymia most likely possess this trait.
Future Directions
Despite the great advancements that have been made in the
past three decades in particular, a great deal is still to be
investigated and understood about alexithymia. One particular
area of research that has not reached a consensus despite great
advancements in measurement of alexithymia and research
methods in general is emotional reactivity and expressivity in
individuals with alexithymia. In spite of the most recent
research bringing further support for the hypoarousal theory
of alexithymia, the results are still tentative and further research
is needed. Emotional expression research has to continue to
reach a better understanding of reasons for people high in
alexithymia not to experience the benefit of expressing emo-
tional behaviors, like crying. Additional inquiry is needed in
the area of treatment for people with alexithymia, as the lim-
ited research that has been done indicates that certain types of
interventions can be beneficial. Furthermore, while investiga-
tors have started posing questions about the overlap or con-
nection between alexithymia and constructs such as Asperger's
and Autism spectrum disorders, this area is still in its infancy
and considerable work is warranted. Finally, additional empir-
ical inquiry is needed to bring further support to some of the
propositions that are still controversial in alexithymia research.
For example, longitudinal studies on samples of youth at risk
to develop alexithymia could potentially bring further clarifi-
cation to the state-trait debate.
Summary
Alexithymia as a personality trait has developed rapidly and
has attracted interest in areas as diverse as personality and
neurobiology, health psychology, and clinical psychology,
which have brought support for this construct with diverse,
yet specific research agendas. Since the first reports of clinical
observations, empirical evidence has added support to this
construct. Notable advancements have been made in mea-
surement, with the goal of having a valid and reliable mea-
sure of alexithymia having been reached with scales such as
the TAS-20 or the BVAQ, among others. Neurobiology has
found specific brain structures that are related to alexithymia
and that changes in these structures could interfere with this
personality trait. Great advancements have been made in
understanding connections between alexithymia and both
physical and mental health, and consequently how it may
impact treatment strides. In spite of all the knowledge that
has accumulated about alexithymia over the past 4 decades,
areas such as emotional regulation and expression, or types of
alexithymia, need further empirical inquiry.
See also: Depression; Dissociative Disorders; Empathy; Facial
Expression of Emotion; Moral Emotions; Posttraumatic Stress Disorder.
Further Reading
Bermond B, Bierman DJ, Cladder MA, Moorman PP, and Vorst HCM (2010)
The cognitive and affective alexithymia dimensions in the regulation
of sympathetic responses. International Journal of Psychophysiology
75: 227-233.
Fitzgerald M and Bellgrove MA (2006) The overlap between alexithymia
and Asperger’s syndrome. Journal of Autism and Development Disorders
36: 573-576.
Lane RD, Ahern GL, Schwartz GE, and Kaszniak AW (1997) Is alexithymia
the emotional equivalent of blindsight? Biological Psychiatry
42: 834-844.
Lane RD and Schwartz GE (1987) Levels of emotional awareness: A
cognitive-developmental theory and its application to psychopathology.
American Journal of Psychiatry 1 44: 1 33-1 43.
Lumley M (2000) Alexithymia and negative emotional conditions. Journal of
Psychosomatic Research 49: 51-54.
Lumley MA, Beyer J, and Radcliffe A (2008) Alexithymia and physical health
problems: A critique of potential pathways and a research agenda. In:
Vingerhoets AJJM, Nyklicek I, and Denollet J (eds.) Emotion Regulation:
Conceptual and Clinical Issues, 1st edn., pp. 43-68. New York: Springer
Science+Business Media, LLC.
Lumley MA, Gustavson BJ, Partridge T, and Labouvie-Vief G (2005) Assessing
alexithymia and related emotional ability constructs using multiple methods:
Interrelationships among measures. Emotion 5: 329-342.
Martinez-Sanchez F, Ato-Garcia M, and Ortiz-Sortia B (2003) Alexithymia - State or
trait? The Spanish Journal of Psychology 6: 51-59.
Mikolajczak M and Luminet 0 (2006) Is alexithymia affected by situational stress or
is it a stable trait related to emotion regulation? Personality and Individual
Differences 40: 1399-1408.
Moormann PP, Bermond B, and Albach F (2005) The reality escape model:
The intricate relationship between alexithymia, dissociation, and anesthesia in
victims of child sexual abuse. In: Nyklicek I, Temoshok L, and
Vingerhoets A (eds.) Emotional Expression and Health: Advances in Theory,
Assessment, and Clinical Applications, 1st edn., pp. 81-98. London: Taylor &
Francis e-Library.
Moormann PP, Bermond B, Vorst HCM, Bloemendaal AFT, Teijn SM, and
Rood L (2008) New avenues in alexithymia research: The creation of
Alexithymia 99
alexithymia types. In: Vingerhoets AJJM, Nyklicek I, and Denollet J (eds.)
Emotion Regulation: Conceptual and Clinical Issues, 1st edn., pp. 27-42.
New York: Springer Science+Business Media, LLC.
Rottenberg J, Bylsma LM, and Vingerhoets AJJM (2008) Is crying beneficial?
Current Directions in Psychological Science 17: 400-404.
Taylor GJ (2005) Alexithymia: 25 years of theory and research. In: Nyklicek I,
Temoshok L, and Vingerhoets A (eds.) Emotional Expression and Health:
Advances in Theory, Assessment, and Clinical Applications, 1st edn., pp. 138-154.
London: Taylor & Francis e-Library.
Taylor GJ, Bagby RM, and Parker JDA (1997) Disorders of Affect Regulation:
Alexithymia in Medical and Psychiatric Illness, 1 st edn. Cambridge:
University Press.
Taylor GJ and Taylor HL (1997) Alexithymia. In: McCallum M and Piper WE (eds.)
Psychologycal Mindedness: A Contemporary Understanding, 1st edn., pp. 77-104.
Mahwah, NJ: Laurence Erlbaum Associates.
Vorst HCM and Bermond B (2001) Validity and reliability of the of the Bermond-Vorst
alexithymia guestionnaire. Personality and Individual Differences 30: 413-434.
Relevant Website
http://www.emotionalprocessing.org.uk/emotion%20concepts/Alexithymia.htm -
Emotional Processing.
Altruism and Helping Behavior
A A Stukas, La Trobe University, Bundoora, VIC, Australia
E G Clary, Kutztown University, Kutztown, PA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Altruistic helping Helping behavior motivated by concern
for the person in distress; selfless helping.
Egoistic helping Helping behavior motivated by concern
for the benefits and costs to the helper; selfish helping.
Helping behavior Actions intended to assist another person
with a problem or to relieve distress.
Prosocial behavior Positive forms of behavior, including
helpfulness; the concept stands in contrast to antisocial
behavior.
HELPING BEHAVIORS are activities where people provide
assistance in solving the problems of other people. In this
article, we review the characteristic forms of helping behaviors,
a continuum that begins from spontaneous help which is often
given in emergencies, to episodic help such as blood donation,
and finally to long-term help which is typically provided by
volunteers working for nonprofit organizations. Along the
way, we examine the research that has focused on the thoughts,
feelings, and behaviors of those involved in helping. We then
describe a process model that focuses broadly on the antece-
dents, experiences, and consequences of helping. Among the
many personal and situational influences on helping, we dis-
cuss its motivational underpinnings. For example, altruism
designates a case where the assistance being provided is moti-
vated by concern for the other person, whereas assistance
motivated by concern for self is regarded as egoistic. We main-
tain a consistent emphasis on the situational and interpersonal
contexts that surround any act of helping and contribute to its
form and efficacy. Lastly, we examine the consequences of
helping with respect to the recipients and the helpers, as well
as for the larger community and society.
Forms of Helping
Whereas human beings are clearly capable of extreme cruelty
and violence toward others, they are also capable of displaying
extraordinary acts of kindness, generosity, and sacrifice on
behalf of their fellow human beings. Amidst the horrors of
Nazi-dominated Europe were cases of the Gentiles rescuing
and hiding the Jewish victims of the Final Solution. In the
southern United States, during the 1950s and 1960s, white
participants in the civil rights movement were working actively
with black participants in political organizing and voter regis-
tration. These dramatic acts of helpfulness may well be infre-
quent occurrences, yet many smaller and more ordinary acts
occur with great regularity. During the course of our everyday
lives, we can assist others with disabled vehicles or dropped
articles, give directions or loose change, to name but a few
opportunities. In addition to these brief help-giving encoun-
ters, people help on a sustained basis, oftentimes in the form
of volunteer work. We find, for example, people providing
companionship to young people or the elderly, counseling
those who are psychologically distressed, and tutoring illiterate
adults or disadvantaged children. Moreover, national surveys
have found that approximately half of all American adults
reported having performed some kind of volunteer activity in
the previous 12 months.
Clearly, prosocial activities are widespread and include a
broad range of actions. One attempt to uncover the differences
among this wide assortment of helpful acts applied statistical
procedures to helping behaviors, utilized as dependent mea-
sures in research, and found helping situations to vary along
three dimensions. Table 1 presents helpful actions that are
representative of the combination of the three dimensions.
The first dimension looks at whether the helper directly
provides help or indirectly helps the person in need by pro-
viding a resource that can be used to alleviate the need. The
second dimension is whether the helping situation arises
unexpectedly, with the helper being surprised by the need for
help (e.g., emergencies), or is planned, with the helper deliber-
ating before acting and, perhaps, even seeking out the oppor-
tunity. Finally, the situation in which the need for help arises
may be serious or not serious.
Researchers have often limited their studies of helping
behavior to particular forms of helping. For example, they
have studied spontaneous or unplanned helping, episodic
helping, and long-term helping; in each case, the situation
may have been serious or not serious, and the action direct or
indirect. Many studies take just a single type of helping (e.g.,
blood donation) as their central focus. We now turn to each of
these forms of helping and the research that supports them.
Spontaneous Help
For behavioral scientists in the 1970s and 1980s, the central
question of interest was whether or not a person would help
when he/she was placed in a situation in which a stranger
needed help. A major impetus for behavioral scientists' atten-
tion to helping behaviors was a dramatic episode of a failure
to help. In 1964, Kitty Genovese was murdered by a knife-
wielding attacker as she returned to her New York City apart-
ment building in the early hours of the morning. The episode
consisted of two separate attacks over an approximately 30 min
time period and was witnessed by at least 38 people who were
watching from their apartments. None of the observers as-
sisted her. Although recent archival work suggests that some
aspects of this 'parable' may be apocryphal, the subsequent
100
Altruism and Helping Behavior 101
Table 1 Dimensions of helpful actions and representative acts
Planned help
Unplanned help
Serious situation
Not serious situation
Serious situation
Not serious situation
Indirect help (giving)
Direct help (doing)
Donate blood
Crisis counseling
Save cartons for art students
Volunteer for experiment
Telephone for a medical emergency
Perform CPR for a heart attack victim
Give money for bus fare
Look for lost contact lens
experimental research did offer important scientific insights
into the situational factors that influence whether or not
bystanders will intervene in an emergency, using the Genovese
story as a model for their studies. College students participat-
ing in an experiment, for example, have been exposed to a
medical emergency or a theft, and from these studies,
one consistent finding has emerged: the presence of other
people, and particularly strangers, inhibits intervention during
emergencies.
From the attempts which have been made to explain the
inhibitory effect of the presence of others has come a frame-
work for the intervention process in emergencies (see Figure 1).
The first condition that must be met before a potential helper
can intervene is attention to the event and, interestingly, sev-
eral factors can actually lead to a failure in noticing even a
dramatic emergency. These include being overwhelmed by
stimuli (as might be the case when one lives in a large urban
area), distraction from external events (because one is focusing
attention on one's own problems), or time pressure. Upon
noticing, however, the potential helper must then interpret
the episode as a situation that requires help. An interpretation
that the episode is an emergency is hindered by any set of
conditions that make it ambiguous - for example, is an alter-
cation between a man and woman an attack or a lover's
quarrel?
The conditions mentioned up to this point could occur
regardless of whether one is facing a possible emergency
alone or with others. More directly related to the inhibitory
effect of other people is the social process of evaluation appre-
hension. This involves people's desire to present themselves as
'calm, cool, and collected,' and the fear of any embarrassment
being caused by acting as if the situation is serious when, in
fact, it is not. The unfortunate upshot of this process is that
when a group is faced with a potential emergency, individuals
present a calm mask while looking around at others' faces to
determine if the situation is serious - and if others are acting
similarly. The end result is often an interpretation that the
event is not an emergency and so no one intervenes; this state
of affairs, in which everyone believes that others have con-
firmed that there is nothing to worry about, is known as
pluralistic ignorance. When alone, the absence of others removes
the audience and gives the individual more latitude in making
the interpretation that an emergency has indeed occurred.
Having made an interpretation that the event is an emer-
gency, a person must still decide whether or not he or she will
intervene. Again, several factors affect this final step, including
an estimation of whether one possesses the skills necessary to
assist the person in need. In terms of the inhibiting presence of
others, an important aspect of this final decision is the diffusion
of responsibility. When alone, the responsibility for helping a
victim rests solely with the individual. When others are present,
Did the potential helper notice the person in need?
No Yes
(No help) -»■
Did the potential helper interpret the situation as an emergency?
No Yes
(No help) >
Did the potential helper decide to personally intervene?
No Yes
(No help) > Help
Figure 1 The bystander intervention process.
however, several individuals psychologically distribute respon-
sibility for the action and each individual's own sense of
responsibility is lessened. Thus, while the processes of evalua-
tion apprehension and diffusion of responsibility reduce the
probability of helping amongst a collection of individuals,
recent research does offer a potential caveat to this dire claim,
as findings suggest that when other bystanders are friends or
members of the same social group they do not inhibit helping
(and may even encourage it), particularly, if the victim is a
member of the same social group.
Other situational factors may also influence spontaneous
helping. Simple acts of spontaneous helping, such as picking
up a dropped pen for someone or making change when asked,
may be affected by the immediate environment. In particular,
unplanned and nonserious helping may be inhibited in
American cities with greater population size or density (perhaps
via distraction and sensory overload or diffusion of responsibil-
ity). Another inhibition may be the perceived costs that the
helper foresees in terms of time, energy, or money. Moreover,
some situations which call for serious help are clearly dangerous
for the intervener; rescuers of Jewish victims of the Holocaust,
for example, faced possible execution if caught. Even in the
absence of physical harm, some situations involve victims who
are severely injured and suffering, and people can find exposure
to these situations psychologically distressing. Even though
helping can certainly be costly, there are also costs associated
with failing to help (e.g., the disapproval of others).
Yet, spontaneous help is not simply the result of perceiving
a lower number of potential losses. People can benefit through
helping and the availability of potential rewards increases the
likelihood of help. The benefits available to helpers often seem
to center on approval in terms of the approval of others for
helping, the gratitude of the recipient, or self-approval. With
respect to self-approval, helpers can reinforce themselves as
having acted in accordance with their values and/or for having
'made a difference.' Moreover, through helping, a helper can
experience an increase in self-confidence and, more generally,
self-esteem.
1 02 Altruism and Helping Behavior
A number of other studies have examined why situational
features may make spontaneous help more or less likely.
Manipulations of a positive mood (e.g., finding a coin in a
phone booth or having an experience of success) increase the
likelihood of help. Several psychological mechanisms appear
to contribute to this relationship, including the tendency for
people in a good mood to view things in a more positive light,
be reminded of their own good fortune, and the desire to
maintain the positive mood state. The relationship of negative
moods and helping is more complex, and varies with the
specific negative mood. Guilt is generally associated with
increased helping, and is especially so when one feels respon-
sible for the potential recipient's state of need, as helping
provides a mechanism for reducing that guilt. The relationship
of sadness and helping, however, is more variable. Helping
becomes more probable if the sadness is directed at the distress
of another person, but is less likely if the sadness is directed at
oneself. Thus, a potential helper's internal mood state, positive
or negative, which is generated by the current situational fea-
tures or carried into the situation with the helper (as a result of
prior events), can clearly influence helpfulness.
Other studies have looked not just at the ways that current
affect influences spontaneous helping, but also at how cur-
rently accessible cognitions may do so. For example, partici-
pants who are running late for another engagement have been
found to be less likely to help; thus, competing goals may
block spontaneous helping. Researchers have also primed
helpfulness in participants; for example, after describing the
characteristics of superheroes (a very helpful group), partici-
pants were more willing to sign up for volunteer work (and
actually follow through with their commitment) compared to
a control group. These primes, which often operate at an
implicit or unconscious level, may make certain goals that are
focused on helping more accessible in thought, and when
opportunities arise to act on these goals, primed individuals
are more likely to do so.
Besides goals, other cognitions may become active in some
situations and can influence spontaneous helping behavior.
Here, a great deal of research has examined the role of salient
norms in ensuring helpful action. For example, when pre-
sented with unexpected opportunities to help, witnessing
another person helping increases the incidence of behaving
helpfully; a motorist with a flat tire received more offers of
assistance when travelers had earlier observed another motorist
with a flat tire receiving assistance. In part, witnessing the
helpfulness of another person increases the observer's help-
fulness by teaching new behaviors and by demonstrating that
certain behaviors are appropriate in a situation. Modeling of
the helping behavior by others has been shown to have both
immediate effects (by making norms salient) and long-term
effects (by socializing children to learn social norms or even
personal goals). In particular, models may remind those who
witness their actions of the norm of social responsibility. Accord-
ing to this norm, we should help those who are dependent on
us, such as young children, the sick, and people in need.
Whereas research is consistent on the prediction that help-
ing increases as dependency increases, explanations which
focus on norms like social responsibility are often criticized
on the grounds that they are vague, general, and appear to be
present in situations where there is a failure to help (e.g., when
bystanders are present in emergencies). Furthermore, there are
social norms that discourage helping, as exemplified by the
admonishment 'mind your own business.' Nonetheless, many
societies have norms, such as social responsibility, which serve
as a backdrop to events in which others are in need and can
exert an influence when combined with other factors.
A clearer case of a help-related norm influencing helping
behavior concerns the norm of reciprocity - we should help
those who have helped us. This guideline is not just limited
to the realm of helping, but applies more broadly to our
relations with others. Reciprocity also appears to represent a
universal norm and some societies have very clear-cut guide-
lines about how and when one returns a gift (e.g., the Kula ring
among the Massim peoples found in several small islands near
New Guinea). Yet, receiving help does not guarantee that the
recipient will later return the help. Recipients of help seem to
feel less obligated to reciprocate when the gift represents a
small portion of the helper's overall resources or if the original
help was given accidentally rather than deliberately. Reciprocal
help also seems to be less likely if the helper is judged to have
had ulterior motives for acting helpfully. In this case, recipients
may experience 'psychological reactance' - a negative motiva-
tional state where people wish to reassert their freedom to act,
which may result in a refusal to reciprocate.
The conditions under which prior help is more likely to be
returned raise a larger question about the potential recipient of
help as a situational determinant of helping - when are people
in need more likely to receive help? Research suggests there is
an increased tendency to help when the need of the victim
increases (e.g., in clearly defined emergencies). This tendency,
however, is moderated by the potential helpers' judgment
about whether or not the victim deserves help. People tend to
help more when the victim is viewed as needing help because
of circumstances beyond his or her control, compared to when
he or she is known to have created his or her own predicament.
Subway riders, for example, were more likely to help a fallen
rider when his fall appeared to be due to sickness rather than
drunkenness. In a related vein, more help is forthcoming when
the need seems legitimate and necessary, as in requesting
money for purchasing milk, rather than a more unnecessary
and trivial need (e.g., requesting money for beer). Finally, there
is a tendency to help others more when the person in need
appears to be similar to us, as signaled by the person's salient
social group memberships.
Thus, a considerable amount of research has been devoted
to the social processes that inhibit or enhance spontaneous
helping, particularly in emergency settings, and also with less
serious behaviors. However, until relatively recently, research-
ers in behavioral sciences appeared less interested in investigat-
ing helping behaviors that occur more frequently or even for
long periods of time. We turn next to these, more sustained,
forms of helping.
Episodic Helping
When it comes to planned helping behaviors, those who wish
to assist others may choose to do so in a limited fashion by
engaging in relatively short-term helping behaviors. The most
obvious prototype of episodic helping is blood donation, but
participation in walkathons or paintathons also falls into this
Altruism and Helping Behavior 1 03
category. Indeed, a lot of literature has developed which exam-
ines the factors that influence charitable giving, and this may
sometimes be the helpful choice of those who find themselves
time-poor (conversely, those who are resource-poor may
choose volunteering time over money).
One approach to examining the giving of time, blood, and
money, has focused on the roles that the participants adopt
when they choose to engage in these forms of helping and how
these roles can become a central part of a person's identity or
the sense of self. Research on episodic helping often targets the
predictors of this ongoing commitment, rather than only that
which led the givers to engage in helping in the first place
(which we cover in a later section). Regular ongoing blood
donation has been shown to increase as a function of the
positive expectations of other people in the donor's social
network and as a function of the donor's donation history.
Indeed, at a certain point in their donation career, individuals
may experience a 'role-person merger' whereby the role of
blood donors (or volunteers) becomes a central and important
part of who they are, and this is made stronger by the fact that
other people in their social network are also likely to acknowl-
edge the importance of this role for these persons; this can help
to maintain the role-consistent behavior.
Aside from 'opportunity costs' (e.g., those costs that one
must bear if one is to use one's time to engage in helpful
behaviors rather than to pursue other goals or leisure activ-
ities), other factors may also influence whether or not someone
gets involved, episodically. For example, recalling that a parent
or another significant role model engaged in the same activity
is associated with greater commitment to episodic helping
(blood, money, and time); as is the perception that people
have a moral obligation to engage in these behaviors. Social
marketing research also suggests that episodic helping may also
be a function of catching people at the right time and place, or
of finding ways to make helping behaviors more convenient
and/or a regular part of a person's daily schedule (e.g., blood-
mobiles and workplace-sponsored helping). Finally, reducing
the costs associated with helping opportunities (such as those
involved with driving to the location or giving up time on the
weekend) can increase the likelihood of episodic helping.
Long-Term Helping
From the late 1970s onward, behavioral scientists began to
turn their attention beyond short-term or emergency helping
and the factors that influence it, to the sort of planned and
sustained helpful behaviors that are epitomized by volunteer-
ism. In many nations, volunteers provide an enormous resource
to communities, free of charge, which can be the equivalent of
many billions of dollars per year in labor costs. They get
involved in organizations that focus on helping other people
( public-serving organizations, such as the Red Cross or Habitat
for Humanity) or those that focus on bringing like-minded
people together or promoting a particular activity [member-
serving organizations, such as local arts organizations or interest
groups). These organizations provide many helping opportu-
nities that range from those focused on improving the health
and well-being of adults, youth, or children, to those that focus
on specific issues, such as education, the environment, or
human rights (including many political organizations), and to
those that help to organize arts, sports, and cultural activities.
Within these organizations, volunteers take on many roles, with
some requiring few skills and little training (picking up garbage
on beaches, delivering meals to housebound elders), and others
requiring increasing amounts of training, orientation and exist-
ing skills (serving as a mentor to troubled youth, fundraising,
coaching a sports team).
Sociologists have examined the ways in which background
characteristics may be associated with a greater likelihood of
volunteering. First, people with greater stores of 'human capital'
resources, such as higher education or higher status jobs, com-
pared to those who have fewer such resources, tend to volunteer
at higher rates, perhaps because they are more aware of the
social issues that require volunteers' services or because they
are more frequently asked to volunteer. Second, volunteering
also occurs at higher rates for those who possess greater social
resources, or social capital. Having a large social network and
belonging to more organizations can create more opportunities
for volunteering, as in the type of volunteering that involves
people helping the schools or churches that their friends and
family attend. That is, through family, friends, and other people
they know, people leam about problems and the opportunities
to solve them. They get asked to volunteer, but they also develop
shared interests and concerns, and become enmeshed in rela-
tionships that involve trusting and being trusted, potentially, as
part of a community that shares beliefs and norms about the
importance of contributing to the public good.
A Process Model of Helping
With volunteerism as a case in point, we now review the ante-
cedents, processes, and consequences of helping behavior and
the different theories and findings that have explored this
prosocial behavior.
Antecedents
The most distal cause of helping behavior considers the possi-
bility that humans may have evolved to engage in helping
behavior, with the discussion centering on the adaptive advan-
tages of helping in the survival of the individual and the
species. Evolutionary psychologists have demonstrated that
people are more likely to help those with whom they have
close biological ties than those with whom they share no
genetic similarities (known as kinship selection). Such helping
clearly promotes the transmission of one's genes to the next
generation if those carrying similar genes are assisted when in
need. But humans may also help those who are not blood
relatives, to the extent that these strangers are later, more likely,
to return the favor. Such reciprocal altruism (as it is known)
offers evolutionary advantages, at the group level (by invoking
a norm of reciprocity), that may have enabled some early
humans (our ancestors) to outlast others who did not develop
such a norm of reciprocal altruism with others. Other modern
revisions of Darwin's Evolutionary Theory point to other
advantages in forming communities, trusting others, and help-
fulness; these stand in contrast to earlier suggestions that
human selfishness is evolutionarily programmed in ways that
discourage the helping of strangers.
1 04 Altruism and Helping Behavior
A more proximal influence on helpfulness comes from
socialization experiences. Parents explicitly teach their children
about moral and social rules and principles, and about norms
for behavior, using praise and other rewards as well as punish-
ments, to guide and shape children's behavior. For example,
people who endorse the values of benevolence (seeing the
importance of helpful action toward specific known others)
or universalism (seeing the importance of promoting the wel-
fare of all) are more likely to help than those who endorse
more hedonistic values. However, parents also provide a more
implicit form of learning for children by the example of their
own behavior. Social learning theory has demonstrated that
children often engage in behaviors, either prosocial or aggres-
sive, that they have seen modeled by adults, particularly when
those adults received rewards (and not punishments) for
engaging in such behavior. And, indeed, children whose par-
ents have engaged in volunteer work are more likely to volun-
teer later in life. For example, a group of committed civil rights
activists reported that their parents had been activists of an
earlier era (i.e., modeled helpfulness), as well as having estab-
lished warm, positive relationships with their children. This
combination appears to be especially important for internali-
zation of such help-related norms and goals : children are more
likely to adopt, as their own, those standards that are based on
a nurturant model. Research has also shown that preaching
helpfulness in the absence of action is ineffective, with chil-
dren, not surprisingly, imitating this model's (nonhelpful)
behavior.
The question has also been asked whether specific person-
ality traits are associated with helping behavior. Relative
to nonvolunteers, volunteers have been observed to possess
greater degrees of self-efficacy, emotional stability, empathy, an
internal locus of control, and intrinsic religiosity. Under con-
ditions of unplanned or unexpected opportunities to help,
traits, observed to be related to helpfulness, have included
self-esteem, agreeableness, ascription of responsibility, dispo-
sitional empathic concern, and a high need for nurturance
combined with a low need for succorance. Two important
qualifications have been noted regarding the relationship of
personality traits and helping behavior, especially in unex-
pected/unplanned situations. First, a stronger relationship
between traits and helping is obtained when one combines
several traits into a composite measure, as this composite
better captures the more general predisposition to help others
and is thus seen as indicating a prosocial (or altruistic) person-
ality. Second, it appears that whether or not a particular trait
correlates with helping depends on the fit between the trait
and the specific situation. Nurturance, for example, was found
to be correlated with willingness to counsel high school stu-
dents but unrelated to willingness to collate and assemble
class materials; this latter activity, however, was related to
autonomy.
More broadly, research has shown that personality and
other personal factors (such as one's values or beliefs) are
more likely to affect helping behavior in certain kinds of situa-
tions, rather than in all. Specifically, such situational factors
seem to operate as more powerful influences in those situa-
tions where the opportunity to help arises unexpectedly and
the resulting help is unplanned (e.g., emergencies). Con-
versely, in situations where the help is planned and may even
involve the helper seeking out the opportunity to help (e.g.,
volunteer activities), there is a tendency for important personal
factors to emerge as predictors of behavior. This variation in
the strength of situational and personal factors, as a function of
the type of situation, seems to be a general relationship: situa-
tions possessing highly salient external cues for behavior reveal
situational determinants, whereas situations possessing less
salient external cues reveal personal dispositions and desires
that guide behavior.
One area of study where both personal and situational
influences may be involved is the study of prosocial motiva-
tion. Research suggests that there may be individual differences
in the ability to see the perspective of another person and,
when the other is in need, to feel empathic concern for them.
Empathic concern is an emotional state marked by feeling 'sym-
pathetic,' 'compassionate,' and 'tender' in response to another's
need and represents an orientation toward the other person's
distress. Individuals with a disposition to empathize may
be more likely to feel prosocial motivation and are therefore,
more likely to help; their distinctiveness from other people
may be most obvious in those situations when helping may
be easily avoided (e.g., when those who can't empathize avoid
helping). However, other research has shown that empathic
concern may be the result of specific situational factors,
particularly the features of the person who needs the help.
Perspective-taking and empathic concern are easily achieved
by potential helpers who are similar, in some way, to the
person needing their help. Moreover, if a potential helper
feels dissimilar from the person needing help, they may only
feel distress at seeing them in need. Personal distress, character-
ized by such feelings as being 'alarmed,' and 'upset' when
in a situation requiring help, may lead a person to try and
leave the situation, if it is possible, so as to escape their own
negative feelings rather than help. If escape is not possible,
they may help but then that help is impelled by an egoistic
desire to reduce their own distress rather than to reduce the
other person's distress.
Studies in this area have contributed to a major debate in
behavioral science which questions if helping is ever truly
altruistic and motivated solely by concern for another person's
welfare without reference to any personal benefits or costs for
the helper, or, whether it is basically egoistic and offering
benefits (or avoiding downsides) for the helper. If help to
another person is offered, even at a cost to the helper and
with no expected benefits, it may be determined to be purely
altruistic, especially if the help is given when it would be easy
to avoid doing so. On the other side of the debate, researchers
continue to point out benefits that helpers can achieve through
helping, which include a positive feeling about having done
the right thing. Another egoistic possibility available in even
the most selfless-appearing help (aside from anticipated social
approval and gratitude) is the potential for a feeling of 'one-
ness' or a merging of self and other, which creates a feeling of
shared positive outcome. This is more likely if the person
needing help has a close relationship with the helper; this
also means that in the most communal relationships, helping
the other is essentially self-help.
Research on planned helping has usually eschewed the
debate over whether helping is altruistic or egoistic and
instead, has focused on the fact that helping behavior may be
Altruism and Helping Behavior 1 05
motivated by both other-focused and self-focused goals at
the same time, and indeed, by many different goals. This
principle, and the idea that different people may engage in
the same helping action for different reasons, are hallmarks
of the functional approach to volunteerism (which investigates
the functions that volunteering serves in the lives of those who
volunteer). Researchers using this approach have identified a
number of reasons for which people get involved in volunteer
activities (see Table 2).
Processes
Indeed, researchers have long recognized that merely under-
standing the antecedents of helping behavior may not be suffi-
cient to fully explain the experience of helping and its
consequences. Actual helping is often theorized to be the result
of an interplay between background characteristics (e.g., dis-
positions, motivations) and situational features (e.g., charac-
teristics of the recipient of help, specific affect, and cognition
that are aroused by the experience). For example, the func-
tional approach to volunteerism suggests that motivations to
volunteer have an impact on positive outcomes such as satis-
faction and future intentions to volunteer only to the extent to
which the volunteers' actual experiences allow them to achieve
these instigating goals.
One important manifestation of the interplay of personal
factors and situational features concerns the actual interactions
of the helpers and the recipients, especially when helping
activities involve people from different social backgrounds or
with different group memberships. When a helper and a recip-
ient of the help belong to different social groups, a number of
issues may arise which influence how people feel about the
experience. Researchers have long suggested that receiving help
may be threatening to one's self-esteem, suggesting a weakness
or an inability to take care of oneself. That is, helping often
involves a relationship that is unequal in power, and may carry
a suggestion that the recipient is incompetent, with the result
that needing help can be embarrassing and humiliating. Also,
helpers vary in the ways in which they give assistance; some are
caring and supportive, whereas others may assist in more
negative and threatening ways. Certainly, help that is given in
a more negative and threatening manner is experienced as
being more unpleasant for the recipients.
When help comes from another person who belongs to a
higher status social group, the threat involved with receiving
Table 2 Why people volunteer
Motivation
Description
Values function
To act on and to express their prosocial values
Understanding
To gain a sense of understanding of the world, an
function
issue, themselves, or other people
Enhancement
To feel more positively about themselves
function
Protective
To protect themselves from feeling bad
function
Social function
To fulfill their social obligations and to fit in with the
expectations of others
Career function
To obtain career experience and advantages
help may take on a new meaning. Findings suggest that low-
status participants may be less willing to seek help from high
status participants (e.g., studies that have involved Israeli Jews
and Israeli Arabs), unless it is clear that the help is designed to
enable them to leam the skills that will allow them to help
themselves in the future. This type of help, epitomized by the
phrase, "Give a man a fish and he will eat tonight, teach
a man to fish and he will eat from now on," is called auton-
omy-supportive help. In contrast, dependency-oriented help does
nothing to help the low-status members of society to change
their situation and is more likely to be rejected when such group
status differences are seen as illegitimate. Following theoretical
traditions that suggest that there are personal benefits to
showing favoritism to one's own ingroup (and occasionally
derogating outgroups), research also shows that high-status
group members are less willing to offer autonomy-supportive
help to low-status group members and instead prefer to offer
dependency-oriented help. This pattern of behavior can serve
to reinforce the existing hierarchical differences between the
involved groups. As such, helping experiences may become
that much more fraught when they involve bringing together
people from unequal backgrounds; however, as we shall
see below, the opportunities for greater social benefits may
also be higher in such intergroup interactions.
Consequences
The final area to be examined concerns the impact of the help
given, and particularly, if the help given was effective. Helpful
actions do not necessarily carry a guarantee of success, as
attested to by the proverb that 'the road to hell is paved with
good intentions.' Moreover, it seems unlikely that there is
some basic skill that reliably leads to effective helping. Rather,
successful help is most probably related to specific skills being
applied to the specific demands of a situation. It is important
to point out that research has shown that possessing the rele-
vant skills also affects the decision to help in the first place. In a
staged emergency that involved a victim of electrical shock,
subjects having experience with electrical equipment were
more likely to intervene than those lacking this experience.
At the same time, a relevant skill is not just another variable
that influences the decision to help. This was demonstrated in
an experiment where subjects, either before or after receiving
Red Cross training, were exposed to a mock arterial bleeding
emergency. Whether subjects intervened or not was influenced
by some of the usual bystander intervention variables (e.g.,
the presence or absence of other people), but how these people
helped was affected by training - those with training were
more likely to provide direct help (the application of direct
pressure to the wound) whereas those without training were
more likely to engage in indirect help (finding other assistance).
Moreover, in the context of the medical emergency used here,
failure to apply direct pressure within 4-10 min usually results
in death, so indirect help may be of no help. Thus, whether a
person will help, or not, is not the same as how one will help.
Research has also examined the attributions of responsibil-
ity that helpers and the recipients of their help make about
both the causes of the current problem and its solution.
In most studies, psychological well-being was positively asso-
ciated with taking responsibility for solutions to problems.
1 06 Altruism and Helping Behavior
Investigations from several areas are generally consistent on
this point, with an improvement being more likely when the
recipients of help become active participants in solving their
own problems. Research on psychotherapy provides one such
example in which successful outcomes are associated with the
clients' attributing the change to their own efforts.
So, the consequences of helping for recipients of help may
vary depending upon the skills and abilities of helpers, the
amount of responsibility taken on by recipients (if appropri-
ate), and a number of other factors, including the magnitude of
the recipient's need, the difficulty of implementing a solution
(e.g., some problems require environmental, political, or social
changes for a solution) and the resources that are available to
assist in the act of helping. These factors may also influence
the consequences of helping for helpers, the communities to
which these individuals belong, and to the larger society.
Most research that examines whether helpers also benefit
from offering help has examined the goals that helpers may
seek to meet in their helping activities. As discussed earlier,
numerous egoistic benefits can be achieved from helping and
volunteers cite career benefits, social benefits, esteem benefits,
and gains in understanding, as desirable personal consequences
of their good work. Other research suggests that volunteerism is
associated with greater well-being, better mental and physical
health, and even longer lives for the elderly (although this latter
effect may dissipate if too many hours are committed). However,
one recent study suggests that greater well-being is only asso-
ciated with motivation to help others because of the pleasure
that can be derived, and helping motivated by external pressure
is less related to well-being. Similarly, those who perceive a
requirement to volunteer (often imposed by academic institu-
tions as a condition of graduation) to be too controlling of their
actions indicate that they will be less likely to volunteer in the
future (as compared to those who are unaffected by such require-
ments), even if they had already demonstrated a commitment to
volunteering freely in the past. These findings are not inconsis-
tent with the notion that there are egoistic benefits to be achieved
from helping, but clearly, helpers must be attuned to and value
the available benefits. When volunteers find themselves in this
situation, research suggests that self-focused motivation maybe a
greater predictor of continued volunteering than other-focused
motivation. That is, those who indicate that they are purely
altruistic and selfless may burn out faster than volunteers who
possess a mix of self- and other-oriented motives.
Finally, helping behaviors can also have consequences for
the organization that is sponsoring the helping activities (if
any), the relationship between the helper and the recipient, the
community at large, and the greater society. For example,
researchers have examined the extent to which volunteers per-
ceive a congruence between their own values and the mission
of any organization to which they contribute their time and
energy, as a predictor of continued helping (at that organiza-
tion). Indeed, the respect that an organization shows to its
volunteers and the pride that volunteers feel in being part of
the organization can contribute to the continued commitment
by volunteers, and concomitantly, to the health of the organi-
zation and its ability to achieve its mission.
Perhaps even more important for consequences at the
macro level is the way in which helping behavior can create
positive bonds of trust and reciprocity between helpers and
recipients. Theories focused on the generation of social capital
have suggested that such bonds can be created through social
participation, either as part of a member-serving organization
or a public-serving organization. Although bonds between
people from similar backgrounds, dubbed bonding social cap-
ital, may be easier to create, bonds between people from differ-
ent backgrounds, dubbed bridging social capital, may offer
more benefits to society and help override social ills that may
come about when there is hostility or prejudice between
groups in society. However, it is not yet clear if such helping
interactions can indeed create bonds of trust across previously
established group lines. Most studies of intergroup contact
suggest that interactions have more positive outcomes when
people meet in circumstances that offer equal status to all
participants; however, such is not always the case in helping
interactions where helpers and recipients may differ in power.
Nevertheless, research does suggest that trust and a sense of
community can be realistic outcomes of volunteerism and
social participation. As such, societies may need to further
promote helping behavior as a way of ensuring that these
qualities are in good supply in order to assist with the many
problems facing the world.
Conclusions
As with human behavior in general, prosocial action is a com-
plex phenomenon that is affected by a variety of factors and
considerations. Whether or not one helps in a situation where
another person is in need depends on social forces operating in
the situation, the characteristics of the particular needs of the
situation, and the motivations and abilities of the potential
helper. Having decided to help, the helper must decide how,
and in what way, to help, with the specific type of help and
approach to helping being influenced by some of the same
factors, as well as by others (e.g., the helper's fundamental
assumptions about the helping process). An act of helping,
then, can be broadly constructed as an interactive function of
the characteristics of the helper and characteristics of the help-
ing situation.
In several respects, helping behavior is not as simple an
issue as it may appear at first glance. As we have seen, there are
different influences on short-term and long-term helping and
both distal and proximal influences may guide the course of a
helping interaction and its consequences for the recipient, the
helper, and the context around them. An array of different
theoretical and empirical approaches has been developed to
examine helping, both in the abstract and in very specific
concrete situations (as with blood donation). However,
within this complexity, a generally positive view of human
behavior has emerged, with evidence which shows that peo-
ple are willing to help each other under many different con-
ditions and in many different ways. The scientific endeavor
continues to offer data to support new ways of understanding
helping behavior and speaks of ways to ensure that it leads to
benefits for all.
See also: Crowd Psychology; Empathy; Group Dynamics; Social
Values (Influence on Behavior).
Altruism and Helping Behavior 1 07
Further Reading
Batson CD (1998) Altruism and prosocial behavior. In: Gilbert D, Fiske S, and Lindzey G
(eds.) The Handbook of Social Psychology, 4th edn., vol. 2, pp. 282-316.
New York: McGraw-Hill.
Carlson M, Charlin V, and Miller N (1988) Positive mood and helping behavior: A test
of six hypotheses. Journal of Personality and Social Psychology 55: 211-229.
Dovidio J, Piliavin J, Schroeder D, and Penner L (2006) The Social Psychology of
Prosocial Behavior. Hillsdale, NJ: Erlbaum.
Eisenberg N and Mussen P (1 989) The Boots of Prosocial Behavior in Children.
Cambridge: Cambridge University Press.
Krebs D and Miller DT (1985) Altruism and aggression. In: Lindzey G and Aronson E
(eds.) The Handbook of Social Psychology, 3rd edn., vol. 2, pp. 1-71. New York:
Random House.
Mikulincer M and Shaver PR (eds.) (2009) Prosocial Motives, Emotions, and Behavior:
The Belter Angels of Our Nature. Washington, DC: American Psychological
Association.
Oliner SP and Oliner PM (1988) The Altruistic Personality: Rescuers of Jews in Nazi
Europe. New York: Free Press.
Smithson M, Amato P, and Pearce P (1983) Dimensions of Helping Behavior. Oxford:
Pergamon Press.
Snyder M, Clary EG, and Stukas A (2000) The functional approach to volunteerism.
In: Maio G and Olson J (eds.) Why We Evaluate: Functions of Attitudes,
pp. 365-393. Hillsdale, NJ: Erlbaum.
Alzheimer’s Disease
Y Gazes, Columbia University Medical Center, New York, NY, USA
A Soldan. University of Massachusetts - Dartmouth, North Dartmouth, MA, USA
Y Stem, Columbia University Medical Center, New York, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Acetylcholine A common neurotransmitter.
Amyloid angiopathy The formation of protein
fragments called amyloid inside the walls of cerebral
blood vessels.
Amyloid beta The main constituent of plaques in the brains
of AD patients.
Amyloid plaques Hard, insoluble plaques made of
protein fragments called amyloid accumulated
between neurons.
APOE (apolipoprotein) gene A gene that is involved in
regulating cholesterol levels. The E4 type of this gene is the
major genetic risk factor for late-onset AD.
APP (amyloid precursor protein) A protein from which
neurons make amyloid beta.
Autoantibody An antibody that attacks an individual's own
proteins.
Blood-brain barrier A separation between the blood and
the cerebrospinal fluid restricting molecules from entering
the central nervous system.
Cerebrovascular Of or related to blood vessels that supply
the brain.
Cholinergic Referring to cells or structures that use the
neurotransmitter acetylcholine.
Dementia Loss of cognitive abilities severe enough to
interfere with daily functioning.
Diabetes A condition in which normal insulin function is
dismpted, resulting in an overaccumulation of glucose in the
blood.
Dyslipidemia An abnormality in the lipid content in the
blood.
Enzyme Biological catalysts that speed up chemical
reactions in the body.
Hippocampus Brain structure important for memory.
Hypertension High blood pressure.
Inflammation The response of the immune system to
injury, irritants, or infection.
Inflammatory Of or referring to inflammation.
Neurodegeneration Progressive damage to and loss of
neurons.
Neurofibrillary tangles Aggregates of the protein tau found
inside neurons of AD patients.
Neuropsychiatric Of or referring to psychiatric disorders
attributable to nervous system disease, such as
hallucinations, delusions, aggression, and depression.
Neurotransmitter Chemicals used by neurons to
communicate with other cells.
Oxidative stress Increased release of unstable oxygen causes
damage and possibly death of neurons.
Plaques Abnormal deposits of amyloid cell debris in the
brains of AD patients.
Tau The main constituent of neurofibrillary tangles.
Alzheimer's disease (AD) is the most common form of demen-
tia, constituting about 60-80% of all dementia cases. Dementia
is defined as loss of cognitive abilities severe enough to interfere
with daily functioning. AD is a progressive, degenerative, and
terminal disease that currently afflicts more than 25 million
people worldwide. The disease is characterized by deterioration
of cognition and memory and a gradual loss of the ability to
perform activities of daily living. Because the occurrence of
AD is strongly associated with increasing age and because life
expectancies have been rising across the world, the prevalence
(occurrence) of the disease could quadruple by 2050, reaching
107 million people. Lack of effective treatment, high healthcare
costs, and increasing numbers of patients make AD one of the
most challenging medical conditions of our times.
History
AD was first described in 1906 by Dr. Alois Alzheimer, a
German psychiatrist. The report was based on the case of
a middle-aged woman with impaired memory, disorientation,
a gradual loss of higher mental functions, and brain
abnormalities (plaques and tangles). Dr. Emil Kraepelin, who
was Dr. Alzheimer's boss, named the disease after him in a
textbook. For many years after Dr. Alzheimer's initial descrip-
tion, AD was not considered a major disease. Cognitive decline
and dementia were considered part of normal aging and usually
referred to as senility. Not until the 1970s did AD become
recognized as a specific disease with well-defined abnormalities
of the brain, leading to increased public awareness and research
funding. It is now recognized that subtle cognitive changes do
occur with aging, but very marked changes are indicative of an
underlying disease process.
Symptoms
AD is characterized by a subtle onset and gradual decline.
Although each person with AD is different, most individuals
advance through a series of stages with progressively more seri-
ous symptoms. Because of its subtle onset, AD can progress for
many years before being diagnosed. The most prominent early
symptom is memory impairment, particularly the ability to
remember recent events or names of familiar people or objects.
108
Alzheimer’s Disease 1 09
Other types of memory are initially relatively unimpaired, such
as old memories from the person's past, memory for skills, and
the ability to benefit from prior experiences in the performance
of some task without conscious awareness of these experiences
(implicit memory). Early AD is accompanied by other, initially
subtle, cognitive deficits, such as impaired verbal, spatial, or
problem-solving skills, as well as impaired planning, cognitive
flexibility, and abstract thinking. For example, some people may
have difficulty working with numbers, following a recipe,
driving to a familiar location, following a conversation, or
keeping track of bills. Language problems are mainly character-
ized by a shrinking vocabulary and word-finding difficulties
(reduced verbal fluency). Other symptoms may include disor-
ientation, increased irritability, suspicion, mood swings, depres-
sion, anxiety, withdrawal from work and social activities, apathy
(or loss of motivation), and sleep disturbance.
As the disease progresses, delusions are common, as are
behavior problems (e.g., aggression, wandering, disregard for
normative social conduct, etc.). Reading and writing skills are
progressively lost. Complex motor skills become less coordi-
nated and there is an increased risk of falling. Patients may
have trouble recognizing close relatives. In the later stages of
the disease, basic activities of daily living, such as eating and
dressing, become impaired. Late-stage AD is marked by the loss
of recognizable speech and the inability to control bodily
functions, leaving patients completely dependent on care-
givers. Muscle mass and mobility may deteriorate to the point
where patients are completely bed-bound, can no longer feed
themselves, and may have difficulty swallowing. AD ultimately
leads to death, although the cause of death may not be the
disease itself but some external factor, such as infections.
The first symptoms of AD usually appear late in life,
although they can occur earlier. Late-onset AD occurs after
age 65 and accounts for 90% of all AD cases. Early-onset AD
begins around age 30-60 years. There do not appear to be any
differences in the symptoms of early- and late-onset AD.
Prevalence and Mortality
Today, the global prevalence of AD in individuals aged 60 and
older is estimated to be about 5-7%. This rate doubles approx-
imately every 5 years after age 65 so that up to 40% of indivi-
duals over age 85 are afflicted by AD. Overall, the prevalence
rate of AD in the United States and Europe is higher than in
developing countries. However, as life expectancy is increasing
in developing countries, the prevalence of AD is projected to
rise in those regions. Cross-cultural studies indicate that life-
style factors associated with living in developed nations also
increase the rates of AD in developed compared to developing
nations. For example, Indians from Northern India and
Yoruba from Nigeria have a significantly lower rate of occur-
rence of AD than Indians living in Pennsylvania and African-
Americans living in Indiana, respectively. Although women
are more likely to have AD than men, this is largely explained
by the fact that women live longer than men. AD significantly
shortens life expectancy, but survival among those with the
disease varies widely. Once diagnosed with the disease, the me-
dian survival time ranges from 3 to 12 years. AD is one of the
leading causes of death in developed nations.
Both the prevalence of AD and the survival time following
diagnosis of AD have been reported to vary across ethnic
and racial groups, but it is currently unclear whether those
differences are due to genetic heritage, health, or social or
cultural differences between groups. For example, compared
to Americans of European descent, African and Hispanic
Americans are at a greater risk for AD, whereas Asian and
Native Americans are at lower risk. At the same time, the life
span following a diagnosis of AD has been shown to be short-
est among non-Hispanic whites than among Hispanics and
African-Americans. When variables such as years of education,
literacy, health status, and socioeconomic status are taken into
account, these group differences are greatly reduced or disap-
pear, suggesting that they are due to factors other than race.
Causes
The causes of AD have not been well understood, and several
competing hypotheses have been proposed to explain the
causes of AD.
Cholinergic Hypothesis
According to the cholinergic hypothesis, the degeneration of
cholinergic neurons (neurons that use the neurotransmitter
acetylcholine) in the basal forebrain of patients with AD
leads to a deficiency in the neurotransmitter acetylcholine in
the cerebral cortex and hippocampus of these patients. The
cerebral cortex and hippocampus are brain structures that are
vital for cognition, behavior, and memory. The reduction in
acetylcholine transmission is thought to contribute signifi-
cantly to the progressive decline in cognitive function seen in
AD. Consistent with this hypothesis, acetylcholine has been
linked to cognition, memory, and attention in both humans
and animals. More recently, it has been suggested that non-
cognitive symptoms of AD, such as depression, aggressive
behavior, psychosis, and overactivity, may also be partially
attributable to cholinergic dysfunction.
The cholinergic hypothesis has served as the basis for the
first drug treatment approaches for AD. In support of the
cholinergic hypothesis, drugs that increase acetylcholine levels
are associated with modest improvements in cognition, func-
tion, and behavior of AD patients in clinical trials. However,
these modest benefits are not sustained in the long term and
the disease continues to progress while patients are receiving
treatment. This and other findings have led some researchers to
conclude that acetylcholine deficits may not be directly causal
to AD, but the result of widespread brain damage and loss of
neurons.
Amyloid Hypothesis
The most prominent hypothesis about the cause of AD is the
amyloid hypothesis, first put forward in 1991. This hypothesis
proposes that amyloid beta protein (also written A(3 or A-beta)
deposits are the causative factor of the disease. Amyloid beta
deposits are the main constituents of plaques (also called
neuritic plaques, senile plaques, or amyloid plaques) that are
characteristically found in postmortem brains of AD patients.
1 1 0 Alzheimer’s Disease
According to the amyloid hypothesis, soluble forms of amyloid
beta protein may be toxic to the brain, causing programmed
cell death, other brain abnormalities, such as tau tangle forma-
tion, and the clinical features of AD. Amyloid beta may also
disrupt the metabolism of neurons by inhibiting certain
enzyme functions and changing how neurons utilize glucose,
their main source of energy.
Amyloid beta protein is a fragment of a larger protein called
amyloid precursor protein (APP) that is located in the mem-
brane of neurons. The primary function of APP is not known,
but it may be important for neuron growth and survival. One
mechanism by which amyloid beta accumulates in the brains
of AD patients is an increase in the rate at which enzymes
cleave APP into smaller fragments, thereby generating in-
creased levels of amyloid beta. Other mechanisms for amyloid
beta accumulation are faulty clearance of amyloid beta from
the brain and overproduction of amyloid beta caused by
genetic mutations that have been linked to early- and late-
onset AD. Thus, individuals with early-onset AD typically
have genetic mutations in APP or the presenilin proteins, the
latter of which form subcomponents of an enzyme that cuts
APP into smaller fragments to produce amyloid beta. Both the
APP and the presenilin mutations act by increasing the overall
production of amyloid beta or by increasing the relative or
absolute level of the toxic form of amyloid beta. In addition,
elevated amyloid beta deposition has been linked to inheri-
tance of the E4 type of apolipoprotein E (APOE) gene, which is
the major genetic risk factor for late-onset AD. Additional
support for the amyloid hypothesis comes from the finding
that the gene for APP is located on chromosome 2 1 and that
patients with an extra copy of this chromosome, known as
trisomy 21 or Down's syndrome, almost invariably develop
AD by 40 years of age.
Despite all the evidence consistent with the amyloid hy-
pothesis, a number of findings have raised questions about it.
Few studies have been able to demonstrate that amyloid beta
is substantially toxic to neurons and induces cell death. Rather,
it appears that the toxicity is dependent on the presence of the
protein tau, the main constituent of the neurofibrillary tangles.
More importantly, drugs that reduce amyloid beta or clear
amyloid plaques in the brain seem to have little effect on the
symptoms of AD or the progression of the disease. Although
there are likely to be a number of reasons for the failure of
these drug studies, they strongly suggest that factors other than
amyloid beta are important for the development of AD.
A possible explanation for why amyloid-reducing drugs may
be ineffective in ameliorating the symptoms of AD is that
amyloid beta triggers tau tangle formation and other disease-
related processes that, once initiated, cause the progression of
AD despite reduced levels of amyloid beta.
Also problematic for the amyloid hypothesis are findings
from recently developed amyloid imaging studies that are able
to measure the amount of amyloid beta deposits in living
human subjects. These studies have shown that up to one-
third of cognitively normal older individuals show substantial
amyloid beta deposits at a level that is similar to that seen in
AD patients. This indicates that the presence of amyloid beta
alone is not sufficient for cognitive impairment and AD. More-
over, amyloid appears to increase linearly throughout the latter
years of life and the rate at which amyloid accumulates in the
brain does not appear to differ between cognitively normal
individuals and AD patients. It is now believed that significant
amyloid beta deposition occurs prior to the clinical symptoms
of AD, which instead are linked to the degree of neurodegen-
eration (progressive damage to and death of neurons). Recent
studies indicate that the cognitive and functional decline seen
in AD patients clearly correlates with brain atrophy (shrinkage)
that is not linked to the accumulation of amyloid beta in the
brain. An updated version of the amyloid hypothesis suggests
that a close relative of amyloid beta protein may be the culprit
in the disease, whereas amyloid beta plays a complementary
role. In this view, an amyloid-related mechanism that is
involved in pruning neural connections early in life to elimi-
nate nonproductive neural circuits may be triggered later in life
to cause the death of neurons seen in AD.
Tau Hypothesis
The third major hypothesis about the causes of AD is the tau
hypothesis. According to this view, abnormalities in the pro-
tein tau, the main component of the neurofibrillary tangles,
cause the progressive neurodegeneration and clinical symp-
toms of AD. In this model, tau proteins pair with other threads
of tau to form tangles inside neurons. These cause the disintegra-
tion of the neurons' internal support structure and transport
system (the microtubules), disrupting communication between
neurons and later leading neurons to collapse and die. Consis-
tent with the tau hypothesis, the degree of tau tangle aggregation
correlates with neuronal loss and appears to agree with clinical
indicators of disease severity. In line with the clinical progression
of AD, tau tangles first form in brain regions that are important
for memory (hippocampus and entorhinal cortex), then spread
to other cortical regions involved in cognition and behavior, and
only later affect sensory and motor cortex, which control sensa-
tion and movement. Also consistent with the tau hypothesis, the
APOE4 gene, the main genetic risk factor for late-onset AD, has
been associated with increased tangle load through a different
mechanism than its effect on amyloid beta. Based on these
findings, potential tau-based drug treatments are currently
being investigated.
Other Potential Causes
In addition to these three main hypotheses regarding the
causes of AD, a number of other mechanisms have been pro-
posed that may cause or contribute to the cause of AD. The
mitochondrial cascade hypothesis holds that late-onset AD is
caused by alterations in the function of the cell's mitochondria,
which are responsible for energy production in cells. Increasing
amounts of evidence also indicate that inflammatory processes
are involved in the neurotoxicity of AD. Inflammation is the
response of the immune system to injury, irritants, or infec-
tion. ffowever, it is not clear whether inflammation is a causa-
tive agent or a downstream reaction to other events, such as
amyloid and tau pathology. AD brains also show evidence of
injury mediated by oxidative stress, in which increased release
of unstable oxygen, such as free radicals, causes damage and
possibly death of neurons. Furthermore, insulin resistance,
which is seen in type 2 diabetes, has been linked to AD,
although the causal mechanism is unclear. Because most
Alzheimer’s Disease 1 1 1
patients with AD also suffer from small-vessel cerebrovascular
disease (narrowing of the small blood vessels in the brain),
it has been suggested that there may be a common cause for
both AD and small-vessel cerebrovascular diseases or that the
two diseases are mechanistically linked. Taken together, cur-
rent research suggests that multiple factors contribute to the
causes of AD and that the balance among these various causes
may vary from patient to patient.
Diagnosis
The most definitive diagnostic method is through postmortem
examination. The suspected AD brain is tested for amyloid beta
plaques, neurofibrillary tangles, and amyloid angiopathy (the
formation of amyloid inside the walls of cerebral blood ves-
sels) using antibody stains. According to postmortem diagno-
sis of AD, 80-90% of the antemortem clinical diagnoses are
correct.
Diagnosis of AD in a person with memory complaint is
based on criteria specified in the Diagnostic and Statistical
Manual IV - Text Revision and from the National Institute
of Neurological and Communicative Disorders and Stroke
and Alzheimer's Disease and Related Disorders Association
(NINCDS-ADRDA). The set of criteria states that memory
loss should be established and disrupted function occurs in
at least one other cognitive domain, such as language or per-
ception, or that executive function is impaired. Some addi-
tional common symptoms include confusion, hallucinations,
withdrawal, and agitation. To distinguish AD from other cog-
nitive debilitating diseases, the onset of the symptoms must be
subtle and the condition deteriorates gradually. Mild cognitive
impairment (MCI) describes the category of patients who do
not fit into the AD diagnosis but who do display noticeable
cognitive deficits such as memory complaints. The symptoms
of MCI patients do not interfere with daily life.
Many screening tests have been developed for AD diagno-
sis. The most commonly used is the mini-mental state exam
(MMSE), which assesses competence in language, memory,
basic motor skills, and orientation. Despite MMSE's wide-
spread usage, its weaknesses include lengthiness, insensitivity
to the extremes of the disease spectrum, and vulnerability to
differences in education level, gender, age, and cultural back-
ground. The modified version of MMSE includes additional
tests for executive function and visuospatial ability to improve
the tests' reliability and validity, but the modifications have
lengthened the test even further. Some medical practitioners
recommend using other more accurate cognitive tests in place
of MMSE, such as the Montreal Cognitive Assessment (MOCA),
the DemTect, the 7-minute Screen, the General Practitioner
Assessment of Cognition, and the Behavioural Neurology
Assessment Short Form. The MOCA and the DemTect, in partic-
ular, are more sensitive to MCI diagnosis. The MOCA covers a
wider array of cognitive abilities than the MMSE, such as clock
drawing, executive functions, and abstraction. The DemTect
includes word recall, semantic word fluency, backward digit
span, and number transcoding.
Screening tests enable quick detection of potential AD
occurrence, but a full neuropsychological assessment is neces-
sary to accurately diagnose the disease and to distinguish it
from other dementing diseases. A typical testing battery
assesses both verbal and nonverbal memory, language, atten-
tion, praxis, problem solving, and activities of daily living.
Examples of tests administered to assess these abilities include
the Selective Reminding Test (SRT) for verbal memory, the
Benton Visual Retention Test (BVRT) for nonverbal memory,
the Boston Naming Test (BNT) for language ability, the Digit
Span subtest of the Wechsler Adult Intelligence Scale-Revised
(WAIS-R) for attention, the Block Design subtest of the WAIS-R
for praxis, the Wisconsin Card Sorting Test for problem solv-
ing/abstract reasoning, and the Philadelphia Geriatrics center
Forms for activities of daily living.
A full neuropsychological battery is crucial to AD diagnosis.
Loewenstein et al. (2001) reported that, rather than a single
factor, multiple distinct cognitive domains are impaired in AD.
To confirm a dementia diagnosis, two or more cognitive func-
tions have to be abnormal as determined by neuropsychological
tests. Studies have further shown that different stages of AD have
distinct neuropsychological profiles and may help to identify
milder forms of AD. For example, Vliet et al. (2003) found that
memory and abstract reasoning deficits best differentiated a
group of mild AD patients from healthy elders, while scores
on memory tests separated MCI patients from healthy elders.
An important diagnostic step is to administer laboratory
tests to exclude the possibility of other diseases, such as thyroid
diseases and anemia, which may explain the observed symp-
toms and thus can potentially reverse the memory deficits.
Vitamin B12 deficiency contributes to cognitive deficits and
has to be ruled out as a possible cause of symptoms as well.
Biomarkers
Another diagnostic step is to test for specific biomarkers found
in AD. The biomarkers with the highest specificity and sensi-
tivity are decreased levels of amyloid beta, increased total tau
levels, and increased phosphorylated tau in cerebrospinal
fluid. However, the necessity of a lumbar puncture, an invasive
and uncomfortable procedure, to obtain these biomarkers
contributes to these tests being underutilized. A number of
AD biomarkers can be obtained with the less invasive method
of blood draw, including amyloid beta autoantibodies, APOE,
and isoprostanes, but these produce less reliable results.
Brain imaging has traditionally been used as a diagnostic
tool for excluding lesions as a cause of symptoms. With recent
development in imaging, however, it is becoming a valuable
tool for confirming AD diagnosis. Using high-resolution struc-
tural magnetic resonance imaging (MRI), the amount of atro-
phy of the hippocampus, a structure that is vital for forming
new long-term memories, provides high sensitivity and speci-
ficity for distinguishing AD, MCI, and healthy controls. Calcu-
lation of atrophy level is currently most accurately performed
manually and takes up to 2 h. Even though programs are being
developed to automate the process, which reduces the calcula-
tion time to 30 min, there is a lack of a well-validated auto-
mated procedure. Other structural imaging modalities being
evaluated as a potential AD diagnostic tool include voxel-
based volumetry (measures reduction in cortical gray matter),
deformation-based morphometry (quantifies the amount of
structural deformations in the brain from a standard template),
and cortical thickness measures. These methods combined
1 1 2 Alzheimer’s Disease
with multivariate statistics have great potential for clinical
diagnosis of AD and MCI.
Research efforts are also underway toward the imaging of
brain metabolism for AD diagnosis. Fluorodeoxyglucose posi-
tron emission tomography (FDG-PET) measures glucose
metabolism in the brain, which shows different patterns of
activation in the parietotemporal areas for AD and normal
controls with a sensitivity of 90% and a specificity of 70%.
Rather than relying on glucose metabolism, a more direct tech-
nique uses Pittsburgh Compound B (PiB)-PET, which binds to
amyloid plaques and fibrils to image brain regions with amy-
loid deposition. While studies have found greater PiB binding
in AD and to a lesser degree in MCI patients, conflicting results
have been reported regarding the relationship between PiB
binding and memory abilities. Nevertheless, PiB-PET is under
extensive research as a potential diagnostic tool for AD and
MCI. Functional magnetic resonance imaging also has poten-
tial utility in AD diagnosis. For example, synaptic loss can be
located with cerebral blood flow and cerebral blood volume,
allowing disease progression to be tracked over time.
A comprehensive AD diagnosis procedure should include a
selective combination of the instruments introduced here. For
example, the Preclinical AD Scale (PAS), developed to distin-
guish MCI patients who will convert to AD within 5 years of
MCI diagnosis from those who do not, includes six instru-
ments: age, MMSE score, functional impairment, cognitive
test performance, hippocampal atrophy, and APOE genotyp-
ing. As each of the instruments are developed and validated by
further research, a more comprehensive battery of instruments
will provide better accuracy in AD diagnosis.
The Brief Smell Identification Test
An interesting diagnostic test that takes advantage of olfactory
deficits occurring early in AD and progressing with disease
severity is under investigation at Columbia University. Ten
items from the lengthier University of Pennsylvania Smell
Identification Test (UPSIT) were selected to create the Brief
Smell Identification Test (B-SIT) to accommodate the need
for short tests in a clinical setting. Study results showed
promising diagnostic capability in the B-SIT to distinguish
AD, MCI, and normal controls. Furthermore, the test also
identified patients who would convert from MCI to AD in
follow-up visits. Replication is needed, however, to validate
the test by other research groups.
One criticism of using UPSIT items for testing people with
cognitive decline is its reliance on memory of odors and verbal
ability. As an alternative, the Sniff Magnitude Test (SMT) uses
foul smells to test for olfactory deficit and only asks subjects to
report the presence or absence of smell in sniff canisters, and
thus may be a better diagnostic tool.
Genetics
Genetic contributions to the development of AD differ by
disease onset. Early-onset AD, characterized by disease onset
before 65 years old, is attributed to mutation in genes that
encode APP and the presenilins (PSEN1 and PSEN2), while
the risk of having late-onset AD is associated with a gene that
encodes APOE. Hundreds of other genes have been reported to
mediate the risk for late-onset AD, but none has been consis-
tently associated with the disease. Familial AD, characterized
by two or more family members diagnosed with AD, occurs in
25% of AD cases. Familial AD can be early as well as late onset,
with mutations in the APP and presenilin genes accounting for
71% of early-onset familial AD. Mutation in the APP gene is
classified as ADI and occupies 10-15% of early-onset familial
AD. Mutations in the PSEN1 and PSEN2 genes are classified as
AD3 and AD4, respectively. Whereas AD3 occupies 20-70% of
early-onset familial AD, AD4 is rarely observed.
APOE contributes to an increased risk of having late-onset
AD via influencing amyloid beta metabolism. APOE has three
isoforms (alleles e2, e3, and e4), of which e4 is associated with
the highest risk for AD such that relative to someone without
an s4 allele, having one £4 allele increases AD risk by 2-3 times
and having two e4 alleles increases the risk by 12 times. The e4
allele may exert its effect via the onset, location, and amount of
amyloid beta accumulation in the brain. Studies have found
that in cognitively normal subjects, the amount of amyloid
fibers that make up the plaques in AD increases with the
occurrence of APOE e4 allele. Having the s4 allele does not
confirm an AD diagnosis; 42% of AD patients do not have the
APOE e4 allele. The e2 allele, however, decreases the chance of
developing AD, but the incident of individuals with this allele
is very rare, making the investigation of allele's protective
mechanism difficult to perform.
Individuals with Down syndrome have an extra chromo-
some 21, which contains the gene for APP, resulting in the
overexpression of APP and thus an accumulation of amyloid
beta in the brain. AD symptoms are observed by the age of 40
in almost all Down syndrome individuals.
Risk Factors and Prevention
Vascular Diseases and Diabetes
A number of factors have been identified to increase the risk of
developing AD, and understanding these factors can help pre-
vent the disease. Among the major risk factors are cardiovascu-
lar diseases such as hypertension (high blood pressure),
dyslipidemia (an elevation of lipids in the blood), and diabe-
tes. Hypertension in midlife has been shown to increase the
risk of AD, but the result is less consistent for hypertension
developed in late-life due to the decrease in blood pressure
prior to and after dementia onset. Antihypertensive drug trials
have shown conflicting results in the effectiveness of reducing
hypertension on lowering the risk of AD. Hypertension may
increase the likelihood of developing AD by increasing the
permeability of the blood-brain barrier to proteins, which
results in amyloid beta accumulation in the brain. There
is evidence that high cholesterol, similar to hypertension, is
most predictive of AD when it is developed in midlife. How-
ever, study results have been inconsistent, and further longitu-
dinal studies are needed to clarify its relationship to AD.
While a number of studies have shown that diabetes may
increase the risk of developing AD, a few other studies did not
find a significant correlation between diabetes and AD. The
inconsistency may be due to confounding factors such as
APOE genotype. When these effects were adjusted for, diabetes
Alzheimer’s Disease 1 1 3
was found to be a significant risk factor for AD. It can be
difficult to isolate the effects of diabetes on AD because of the
tendency for diabetes to develop in conjunction with cerebro-
vascular disease, which is a major risk factor for AD. Hypergly-
cemia (high blood sugar), which is the defining characteristic
of untreated diabetes, causes oxidative stress and damage to
brain tissue, which are associated with increased risk of AD.
Insulin and insulin-degrading enzyme have also been found
to worsen amyloid beta metabolism, further increasing the
risk of AD.
Cognitive Reserve
The idea of cognitive reserve originated from the observation
that given two different people with similar brain injury, one of
them may exhibit obvious cognitive dysfunction, while the
other may have no behavioral symptoms. This is best demon-
strated by cases in which women were diagnosed with AD at
autopsy but had no cognitive abnormality before death. Cog-
nitive reserve enables the brain to maintain cognitive abilities
despite brain damage, and thus, having a high level of cogni-
tive reserve can help reduce the risk of AD symptoms.
A number of factors are positively associated with cognitive
reserve: education, occupation, physical activities, and social
and leisure activities. The majority of studies relating the level
of education to the incidence of dementia found that educa-
tion has a protective role in developing dementia, but a few
studies have also reported the effect of lack of education with
dementia. Furthermore, even in healthy elders, a higher level
of education is associated with slower cognitive decline.
Similarly, the degree of cognitive complexity in different occu-
pations and the amount of personnel management have been
shown to be associated with a lower risk of dementia. Overall,
the risk of dementia is lowered with increased education, more
demanding occupation, greater amounts of physical exercise,
and more engagement in social and leisure activities.
A possible mechanism through which cognitive reserve
decreases the risk of AD is the availability of more neuronal
connections and more efficient processing, such that damage
in one pathway may be compensated by using another path-
way. This way, normal function can be maintained despite
damage. Life experiences may increase cognitive reserve by
inducing changes in neuronal processes such as greater cerebral
activation and increased glucose and oxygen metabolism, and
possibly reduced neurodegeneration. Higher education level
and career complexity may also be associated with a healthier
lifestyle and thus lower exposure to neurotoxins such as heavy
drinking. The most likely possibility, however, is a combina-
tion of these different causes.
Diet
Studies on the relationship between individual nutritional
components, such as antioxidants, vitamins, folic acid, and
dietary fats, and risk of AD have not produced consistent
results. An exception is moderate alcohol intake in which
moderate amounts of alcohol (less than three servings per
day) has been shown to lower the risk of AD. There is accumu-
lating evidence, however, suggesting that certain dietary pat-
terns are effective in lowering the risk of AD. One of the most
promising dietary patterns is based on the Mediterranean cui-
sine (the Mediterranean diet), consisting of vegetables,
legumes, fruits, cereals, olive oil, fish, and a moderate amount
of alcohol. The Mediterranean diet is associated with a lower
risk of AD, and oxidation, inflammation, vascular, and meta-
bolic disease. Furthermore, a reduction of 61-67% in AD risk
was reported when the Mediterranean diet was followed along
with physical exercise in elderly individuals.
Other Risk Factors
Experimental evidence suggests that traumatic brain injury
may contribute to AD development. Studies have found that
APP accumulates in injured axons (projections of neurons that
transmit information away from the neuron's cell body),
which could result in amyloid beta production. In a pig
model, amyloid beta and APP were found in neurons after
injury to axons. Similarly, amyloid beta accumulation and
amyloid plaques were also reported in patients with axonal
injury. Nevertheless, consistent evidence is lacking in epidemi-
ological studies relating head trauma to AD.
Cigarette smoking and obesity have also been shown to
increase the risk of AD, most likely because of their effect on
cardiovascular health. Exposure to metals and solvents has
been suggested to increase the risk of developing AD, but
these findings could not be confirmed by other studies. The
evidence for a link between AD and exposure to electromag-
netic fields, such as power lines, electrical wiring, radio waves,
X-rays, and microwaves, is also inconclusive. Occupational
exposure to heavy metals, such as aluminum or mercury, has
also been linked to increased risk of AD, but the evidence
is mixed.
Treatment
Currently, there is no cure for AD, or a treatment that slows the
neurodegeneration seen in AD. However, some drug treat-
ments are available that can ameliorate or slow down the
worsening of the symptoms of AD for a limited time in some
individuals. Non-drug-based psychosocial treatments are fre-
quently used in clinical practice, but few studies have examined
their effectiveness in controlled double-blind trials.
Drug Therapies
Most currently available drugs to treat AD are based on the
cholinergic hypothesis (see section 'Causes'). These drugs, also
called cholinesterase inhibitors, act by increasing levels of the
neurotransmitter acetylcholine in the brain to counteract the
loss of cholinergic neurons. Clinical trials have shown modest
improvements in cognition, function, and behavior of AD
patients treated with cholinesterase inhibitors, with the great-
est benefits being apparent in early stages of the disease. There
seems to be little difference in the efficacy of various cholines-
terase inhibitors, although some of them tend to be better
tolerated by patients and have fewer side effects. Prescribing
cholinesterase inhibitors to patients with MCI does not seem to
delay the onset of AD. Another type of drug, memantine,
reduces the effect of the neurotransmitter glutamate in the
1 1 4 Alzheimer’s Disease
brain. It was developed based on the finding that AD is asso-
ciated with excessive release of glutamate, which can be toxic to
neurons. In the United States, memantine is approved for the
treatment of moderate-to-severe AD, while in Europe it is used
to treat a variety of neurological disorders. Memantine has
shown consistent but small effects in improving cognition
and global assessment. It is associated with a few mild adverse
side effects.
In general, treatment effect sizes for both cholinesterase
inhibitors and memantine tend to be small, though statistically
reliable, for measures of cognition and global assessment.
Effects on behavior and quality of life seem to be less consis-
tent. Both types of drugs have shown some improvement of
neuropsychiatric symptoms associated with AD, such as delu-
sions, hallucinations, apathy, and aggression. The benefits of
cholinesterase inhibitors and memantine tend to be lost when
the drug is discontinued. The usage of standard antipsychotics,
antidepressants, anticonvulsants, and antianxiety drugs for the
treatment of AD-related neuropsychiatric and behavioral
symptoms is problematic, because their modest effectiveness
is offset by the risk of serious adverse side effects and increased
risk of mortality.
Several other drug treatments have been proposed, but they
are not recommended for routine use at this time because
of insufficient efficacy and/or safety in placebo-controlled trials.
These include vitamins B, C, and E, folic acid with or without
vitamin B12, the herbal extract gingko biloba, nonsteroidal anti-
inflammatory drugs (such as aspirin or ibuprofen), hormone
replacement therapy, the hormone melatonin, cholesterol-
reducing statins, and the iron-reducing agent desferrioxamine.
Currently, many other products are being tested, such as tau
and tangle-reducing drugs and stem cell therapies to replace
injured neurons.
Psychosocial Therapies
Much of the published evidence regarding psychosocial treat-
ments is characterized by several limitations, including
inadequate sample sizes, short study duration, use of non-
standardized evaluation methods, lack of proper controls,
and inadequate information on the persistence of treatment
effects. In addition, studies are often not specific to AD but
focus on dementia in general. Overall, the beneficial effects
of psychosocial treatments are small and of a short duration.
The effectiveness of particular treatments also seems to vary
greatly from patient to patient. Therefore, psychosocial treat-
ments may work best in specific, time-limited situations, tai-
lored to the needs and preferences of an individual. Some
interventions may provide pleasure for the patient and/or
caregiver and for this reason be valuable despite not being
effective in the treatment of the disease. Psychosocial thera-
pies for AD can be classified into four broad groups: behav-
ioral, cognitive, emotion-based, and stimulation-oriented
approaches.
The goal of behavioral interventions is to modify specific,
maladaptive, or self-destructing behaviors by learning more
appropriate behaviors. Overall, behavioral interventions are
not effective in treating the overall functioning of patients,
but some evidence supports their use in lessening or eliminat-
ing specific problem behaviors, such as urinary incontinence.
The benefits of behavioral treatments seem to disappear when
treatment is discontinued. The goal of cognitive therapies is to
reduce or slow the cognitive decline seen in AD. They include
cognitive orientation, cognitive retraining, and skills training.
Although modest, short-lived gains in cognition, behavior,
and mood have been demonstrated with cognitive therapies,
some patients may experience frustration. Stimulation-oriented
therapies aim to provide sensory and cognitive stimulation and
seem to be modestly effective in improving behavior and
mood, and enhancing socialization and the quality of life.
They include recreational activities (crafts, games, pets, etc.),
music and art therapies, multisensory stimulation, exercise,
and aromatherapy. Emotion-oriented treatments focus on
improving emotional and social functioning of AD patients
and ultimately, the quality of life. They include reminiscence
therapy, validation therapy, supportive psychotherapy, sensory
integration, and simulated-presence therapy. Currently, there
is limited evidence supporting the effectiveness of emotion-
oriented therapies.
Societal Impact
Caregiver
Across the world, families are fundamental for the care of
patients with dementia. This is particularly true for developing
countries, where residential care units are generally unavail-
able. Individuals with AD need assistance with various aspects
of everyday tasks, ranging from legal and financial manage-
ment to feeding and bathing. AD patients are more likely to
need help with everyday tasks than patients with other dis-
eases: 32% of AD patients need help dealing with bladder and
bowel incontinence versus only 13% of patients with other
diseases. Care is provided by both paid and unpaid indivi-
duals, in which unpaid caregivers consist of family members,
friends, or neighbors, with ages 50-64 occupying the greatest
percentage. Most family caregivers are women, but the number
of male caregivers is rising. Due to the slow progression of the
disease, providing care for AD and other dementias also lasts
for a longer duration than for other types of diseases. The need
for care increases as the disease progresses and can be up to
10 h day -1 for patients with severe AD.
The greater demand and longer duration of care leaves the
caregiver of an AD patient more susceptible to depression and
other illnesses, and can place more strain on the caregiver's
finance and career. Studies have found that unpaid caregivers
experience greater level of stress and have lower immune func-
tion. Some caregivers have to quit their jobs in order to take
care of the AD patient, losing financial support and benefits.
Caregiver interventions have been effective at increasing care-
giver knowledge about AD, improving mood of caregivers,
decreasing caregiver stress and depression, decreasing neurop-
sychiatric symptoms of patients, and delaying nursing home
placement of the patient.
Economic Impact
Due to the longer term and extensive care AD patients require,
healthcare costs for AD patients can be three times that for
individuals of the same age group and will continue to rise as
Alzheimer’s Disease 1 1 5
the number of AD cases increases with the aging population.
According to an estimate by the Karolinska Institute, the world-
wide cost of dementia amounts to US$315 billion per year,
of which 72% is contributed by high-income countries.
Healthcare costs for individuals with AD are not only paid by
patients, their families, and government agencies; private busi-
nesses also pay indirectly through work missed by employees
caring for an AD patient. In the developing world, caregiving is
associated with substantial economic disadvantage due to cuts
in paid work. A reason for the high healthcare cost for AD
patients can be attributed to the tripling of hospital stays
relative to the elderly without AD. Besides hospital stays,
costs of healthcare for AD patients include greater likelihood
of needing to stay in nursing homes and requiring home
healthcare.
See also: Aging and Cognition; Aging and the Brain; Brain and Behavior
Relationships; The Brain; Hippocampal Formation; Memory, Neural
Substrates; Memory; Neuroimaging of Dementia; Neurotechnologies.
Further Reading
Alzheimer's Association (2009) Alzheimer’s disease facts and figures. Aizheimefs &
Dementia 5: 234-270.
American Psychiatric Association (APA) (2007) Practice Guideline for the Treatment of
Patients with Alzheimer's Disease and Other Dementias, 2nd edn. Arlington, VA:
American Psychiatric Association (APA). DOI: 10.1176/appi.
books.9780890423967.1 521 39.
Bird TD (2008) Genetic aspects of Alzheimer disease. Genetics in Medicine 10(4):
231-239.
Duron E and Hanon 0 (2008) Vascular risk factors, cognitive decline, and dementia.
Vascular Health and Risk Management 4(2): 363-381 .
Feldman HH, Jacova C, Robillard A, et al. (2008) Diagnosis and treatment of dementia:
2. Diagnosis. Canadian Medical Association Journal 178(7): 825-836.
Hampel H, Burger K, Teipel SJ, Bokde AL, Zetterberg H, and Blennow K (2008) Core
candidate neurochemical and imaging biomarkers of Alzheimer’s disease.
Aizheimefs & Dementia 4(1): 38-48.
Jack CR Jr., Lowe VJ, Weigand SD, et al. (2009) Serial PIB and MRI in normal, mild
cognitive impairment and Alzheimer's disease: Implications for sequence of
pathological events in Alzheimer’s disease. Brain 132(Pt 5): 1355-1365.
Jalbert JJ, Daiello LA, and Lapane KL (2008) Dementia of the Alzheimer type.
Epidemiologic Reviews 30: 15-34.
Loewenstein DA, Ownby R, Schram L, Acevedo A, Rubert M, and Arguelles T (2001) An
evaluation of the NINCDS-ADRDA neuropsychological criteria for the assessment of
Alzheimer's disease: A confirmatory factor analysis of single versus multi-factor
models. Journal of Clinical and Experimental Neuropsychology 23: 274-284.
Macher J-P (ed.) (2009) Alzheimer’s disease and mild cognitive impairment. Dialogues
in Clinical Neuroscience 11(2) Retrieved from http://www.dialogues-cns.org/
b roch u res/41 /htm 1/41 _3 .as p
Pimplikar SW (2009) Reassessing the amyloid cascade hypothesis of Alzheimer's
disease. International Journal of Biochemistry and Cell Biology 41 (6): 1 261 -1 268.
Prince M and Jackson J (eds.) (2009) World Alzheimer Report. London: Alzheimer's
Disease International.
Ryan NS and Fox NC (2009) Imaging biomarkers in Alzheimer’s disease. Annals of the
New York Academy of Sciences 1180: 20-27.
Scarmeas N and Stern Y (2004) Cognitive reserve: Implications for diagnosis and
prevention of Alzheimer’s disease. Current Neurology and Neuroscience Reports
4(5): 374-380.
Small SA and Duff K (2008) Linking Abeta and tau in late-onset Alzheimer's disease:
A dual pathway hypothesis. Neuron 60(4): 534-542.
Terry AV Jr. and Buccafusco JJ (2003) The cholinergic hypothesis of age and
Alzheimer's disease-related cognitive deficits: Recent challenges and their
implications for novel drug development. Journal of Pharmacology and
Experimental Therapeutics 306(3): 821-827.
Vliet EC, Manly J, Tang MX, Marder K, Bell K, and Stern Y (2003) The
neuropsychological profiles of mild Alzheimer’s disease and questionable dementia
as compared to age-related cognitive decline. Journal of the International
Neuropsychological Society 9: 720-732.
Relevant Websites
http://www.alz.org - Alzheimer's Association.
http://www.alz.co.uk/ - Alzheimer's Disease International.
http://hod.kcms.msu.edu/index.php - Kalamazoo Center for Medical Studies - History
of Alzheimer's disease.
http://www.mayoclinic.com/health/alzheimers-disease/DS00161 - Mayoclinic.com -
Alzheimer's disease.
http://www.nia.nih.gov/Alzheimers/Publications/adfact.htm - US National Institute on
Aging: Alzheimer’s disease fact sheet.
http://www.ninds.nih.gov/disorders/alzheimersdisease/alzheimersdisease.htm - US
National Institute of Neurological Disorders and Stroke: Alzheimer’s disease
information sheet.
Amnesia
J F Kihlstrom, University of California, Berkeley, CA, USA
E L Glisky, University of Arizona, Tucson, AZ, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Alcoholic blackout Amnesia without loss of consciousness,
in which the intoxicated person retains the ability to
perform certain 'automatized' behaviors, without any
subsequent memory for the episode.
Amnesia A special case of forgetting in which the memory
loss is greater than would be expected under ordinary
circumstances. Anterograde amnesia affects memory for
events occurring after the instigating event; retrograde
amnesia affects memory for events occurring before the
instigating event.
Amnesic syndrome A profound deficit in learning and
memory usually associated with bilateral damage to the
diencephalon or the medial portions of the temporal lobe. It
always involves an anterograde amnesia, and may involve a
retrograde amnesia as well.
Functional amnesia A significant loss of memory
attributable to an instigating event, usually stressful, that
does not result in insult, injury, or disease affecting brain
tissue. Its most common forms are psychogenic amnesia,
psychogenic fugue, and multiple personality disorder.
Implicit memory In contrast to explicit memory, which
entails conscious recall or recognition of past events,
implicit memory refers to any effect of a past experience on
subsequent experience, thought, or action - for example,
priming effects.
Infantile and childhood amnesia An amnesia observed in
adults, affecting memory for personal experiences occurring
in the first 5-7 years of life. Infantile amnesia commonly
covers the period before language and speech develop.
Posthypnotic amnesia A retrograde amnesia induced by
means of hypnotic suggestion; it may be canceled by a
prearranged reversibility cue.
Priming A phenomenon where an event facilitates (positive
priming) or impairs (negative priming) performance on
another, later task.
Source amnesia A phenomenon where a subject retains
access to new information acquired during a learning
experience, but forgets the learning experience itself; it is
commonly reflected in cryptomnesia, or unconscious
plagiarism.
Transient global amnesia A benign and temporary amnesia
characterized by sudden onset, apparently caused by
momentary vascular insufficiencies affecting brain tissue.
Trauma-memory argument The notion, rooted in the
nineteenth century ideas of Sigmund Freud and Pierre Janet,
that psychological trauma causes amnesia by virtue of a
process of repression or dissociation.
Traumatic retrograde amnesia A retrograde amnesia
resulting from a concussive blow to the head; most of the
affected memories are eventually recovered, except for a
'final RA' affecting the accident itself.
Amnesia
Amnesia may be defined as a special case of forgetting, where
the loss of memory is greater than would be expected under
ordinary circumstances. A head-injured patient is no longer
able to learn things that he was once able to master easily;
a patient with psychogenic fugue loses her personal identity
as well as her fund of autobiographical memories. Amnesia
includes frank pathologies encountered in neurological and
psychiatric clinics, such as the amnesic syndrome, Alzheimer's
disease (AD), traumatic retrograde amnesia (RA), and the
interpersonality amnesia characteristic of multiple personality
disorder (MPD). But it also includes abnormalities of memory
observed ubiquitously, such as infantile and childhood amne-
sia, the exaggerated forgetfulness associated with normal aging,
and the memory failures associated with sleep and general
anesthesia. These naturally occurring failures of memory have
their counterparts in amnestic states induced in otherwise
normal, intact individuals by means of experimental techni-
ques, such as electroconvulsive shock (ECS) in laboratory rats
and posthypnotic amnesia in college sophomores.
Experimental research on memory began with the publica-
tion of Ebbinghaus' Uberdas Gedachtniss ( On Memory) in 1885,
but the clinical description of amnesia dates from even earlier.
Korsakoff first described the amnesic syndrome that bears
his name in 1854. And in 1882, Ribot published Les Maladies
de la memoire ( Diseases of Memory), with a detailed description
of the consequences for memory of brain insult, injury, and
disease, as well as a unified theory of memory and amnesia.
On the basis of his observations, and Hughlings Jackson's
principle that ontogeny recapitulates phytogeny, Ribot con-
cluded that brain disorder produces a progressive toss of memory
that affects memories in the reverse order of their development.
Thus, in traumatic RA, memories for events occurring immedi-
ately before the accident are most likely to be lost. This principle,
now known as Ribot's law, does not always hold, but it was an
important first step in the journey from clinical description to
scientific theory.
For reasons that are not completely clear, clinical and exper-
imental study of amnesia languished for the first half of the
twentieth century, but was revived by observations of a patient,
H.M. (full name: Henry Molaison), who became amnesic fol-
lowing surgical resection of portions of his medial-temporal
lobes, including the hippocampus, in a desperate attempt to
relieve intractable epileptic seizures. From the time of his oper-
ation in 1953, at the age of 27, until his death in 2009 at the
116
Amnesia 117
age of 82, he had no conscious recollection of any episode in
his life. In addition, George Talland's 1965 pioneering mono-
graph, Deranged Memory, reported an extensive psychometric
and experimental study of amnesic patients with Korsakoff
syndrome. These studies ushered in a 'golden age' of amnesia
research in which clinicians and experimentalists joined forces
under the banner of cognitive neuropsychology (later, renamed
cognitive neuroscience) - a discipline that attempts to integrate
evidence obtained from the intensive study of brain-damaged
patients with theories of normal cognitive function.
The Amnesic Syndrome
The amnesic syndrome represents a profound deficit in learning
and memory; it is by far the most commonly studied pathology
of memory. Its most characteristic feature is a gross anterograde
amnesia (AA), meaning that the person cannot remember events
that have occurred since the time of the brain damage. Short-
term memory (as measured by digit span, for example) is
unimpaired; but after even a few moments' distraction, these
patients cannot remember what they have said or done, or what
has been said or done to them, just recently. In the classic case,
the patient's cognitive deficits are specific to long-term mem-
ory: general intelligence, perception, reasoning, and language
functions are spared. But this AA is associated with several
different etiologies, and these disparate origins are associated
with somewhat different patterns of memory and cognitive
deficit. For more details, see 'Amnesia and the Brain' by Race
and Verfaellie, following this article.
One form of the amnesic syndrome, now known as dience-
phalic amnesia, was first described by Sergei Korsakoff in asso-
ciation with alcoholism. These patients typically show RA as
well as AA, meaning that they also have difficulties in remem-
bering events from their premorbid life, especially those from
the years immediately preceding their disease. Remote mem-
ory, such as for childhood events, is apparently preserved. Note
that such a pattern conforms to Ribot' s law. Chronic alcoholics
often suffer from a deficiency of vitamin Bj (thiamine) which
results in bilateral damage to structures of the diencephalon,
including the upper portion of the brainstem, the mammillary
bodies, the dorsomedial nucleus of the thalamus, and the
mammillothalamic tract. The acute phase of illness, known as
Wernicke's encephalopathy, is characterized by confusion, dis-
orders of visual function, and ataxia; Korsakoffs syndrome
emerges as the chronic phase (the entire course of illness is
sometimes known as Wernicke-Korsakoff syndrome) . Although
this disease is now typically prevented by the introduction of
vitamin-enriched commercial foods, other etiologies, including
vascular insufficiencies, tumors, and frontotemporal brain
damage can have similar effects.
Another form of amnesic syndrome, known as temporal lobe
amnesia, stems from bilateral lesions in the medial portion
of the temporal lobe, and especially the hippocampus, ento-
rhinal cortex, and surrounding structures (again, there are also
material-specific amnesias resulting from unilateral damage to
these structures). The most famous case is patient H.M., but
there are many other famous cases. Other cases have been caused
by brain tumors, ischemic episodes, head trauma, and herpes
encephalitis. Temporal-lobe amnesics show AA by definition.
The occurrence of RA in these patients is somewhat controver-
sial, and may vary if damage extends beyond the hippocampus
itself. H.M. showed little RA; another patient, R.B., showed an RA
covering only about 1 or 2 years prior to his surgery; yet another,
K.C., displayed an RA covering his entire premorbid life.
There is also frontal-lobe amnesia, which is qualitatively
different from the amnesic syndrome. Frontal-lobe patients
are not globally amnesic, but they frequently show deficits on
tasks requiring memory for temporal order, as well as mem-
ory for the source of newly acquired knowledge. They also
lack metamemory capabilities - that is, they have little appreci-
ation of the contents stored in their own memories, or in the
availability of appropriate memory strategies. Patients who
have frontal damage in addition to diencephalic or medial-
temporal lobe damage experience their greatest difficulties on
memory tasks requiring strategic planning and organization,
suggesting that frontal damage may impair executive func-
tion, rather than memory per se.
Finally, transient global amnesia is a temporary (typically
lasting several hours) condition characterized by sudden
onset. It closely resembles the permanent diencephalic and
medial-temporal lobe amnesias, in that it involves both AA
and RA, but - as its name implies - it is brief. The condition,
while frightening, is benign: after remission, there are no signs
of permanent brain damage (and little risk of another episode
in the future). Transient global amnesia appears to be caused
by temporary vascular insufficiency affecting brain tissue; inter-
estingly, many cases appear in association with physical exer-
tion or mental stress.
In the absence of permanent brain damage, something akin
to the amnesic syndrome may be observed in cases of alcoholic
blackout. Blackout involves amnesia without loss of conscious-
ness. The intoxicated individual may engage in conversation or
perform other actions normally; but after regaining sobriety,
he or she will have no memory for the episode. Blackouts are
most commonly observed in chronic alcoholics, though they
do occur to nonalcoholics who are severely intoxicated. In any
case, blackout is most likely to occur when the person ingests
large quantities of alcohol rapidly, especially when fatigued
or hungry. Alcohol folklore suggests that the amnesia is an
instance of state-dependent retrieval - that is, that the mem-
ories return when the person resumes drinking. However, lab-
oratory research clearly indicates that the memories covered by
blackout are unrecoverable, and thus that the amnesia reflects
an encoding deficit. Sedative drugs, such as barbiturates and
benzodiazepines, also produce irreversible AA.
The different patterns of task performance offer clues about
the nature of the memory deficit in the amnesic syndrome. In
principle, any instance of forgetting may be attributed to a
failure at one or more of three stages of memory processing:
encoding (the creation of a memory trace of a new experience),
storage (the retention of trace information over time), and
retrieval (the recovery of trace information for use in ongoing
experience, thought, and action). Logically, a syndrome that
affects memory for postmorbid but not premorbid events is
most likely due to encoding failure. And, in fact, it has been
suggested on the basis of laboratory experiments with lesioned
rats and monkeys that the hippocampus and other structures
in the medial-temporal lobe mediate the consolidation and
storage of new memories. An alternative formulation assumes
1 1 8 Amnesia
that representations of the various elements of an event are
distributed widely in the cortex, and that the hippocampus
creates a kind of index and binds the elements together. In
either case, the occurrence of AA means that the hippocampus
is crucial for memory formation, even though the memories
themselves are not stored there.
What about the RA? Some degree of RA is usually, but not
always, observed in the amnesic syndrome. Logically, damage
to a structure that consolidates and organizes new memories
should have no effect on old ones. In some cases, RA may
reflect the disruption of premorbid memories that were incom-
pletely consolidated at the time of disease onset; this would
produce a temporal gradient, but the extent of RA observed
would seem to imply that proper consolidation requires weeks,
months, or years instead of seconds, minutes, or hours. On the
other hand, if the hippocampus serves a binding and indexing
function, its destruction will create an RA by effectively pre-
venting the retrieval of memories that remain available in
storage; this would produce an amnesia for remote as well as
recent memories and would not necessarily produce a tempo-
ral gradient. In some cases, what appears as RA may in fact be
AA, reflecting the slow onset of an insidious disease process -
and producing the appearance of a temporal gradient; this
suggestion is particularly plausible in the case of diencephalic
amnesia associated with chronic alcohol abuse but cannot
account for amnesias of sudden onset. It has also been sug-
gested that the RA may reflect the fact that, once activated and
retrieved, premorbid memories are then re-encoded: as a result,
the same encoding failure that produces the AA for postmorbid
events results in a progressive loss of premorbid memories as
well, and emergence of RA.
Even conclusions about encoding deficits must be qualified
to some extent. At first glance, the AA observed in the amnesic
syndrome appears to be a complete inability to acquire new
information. However, closer examination indicates that cer-
tain aspects of learning and memory are spared even in the
densest cases of amnesia. Thus, the patient H.M. learned to
solve the Tower of Hanoi puzzle, but he did not recognize
the puzzle as familiar. Amnesic patients who study the word
ELATED do not remember it just minutes later; but when pre-
sented the stem ELA and asked to complete it with the first word
that comes to mind, they are more likely to produce ELATED (as
opposed to ELASTIC or ELABORATE ) than would be expected
by chance. The ability of amnesic patients to acquire cognitive
and motor skills, and to show priming effects in word-stem
completion and other tasks, shows that they are able to acquire
new information through experience - although, somewhat
paradoxically, they do not remember these experiences and
may not be aware that they possess this knowledge.
The limits of such learning are still being studied, but
already they have motivated a distinction between two expres-
sions of episodic memory, explicit and implicit. Explicit memory
(EM) refers to the conscious recollection of a previous episode,
as in recall or recognition. By contrast, implicit memory (IM)
refers to any change in experience, thought, or action that is
attributable to such an episode, such as skill learning or
priming. The dissociation between EM and IM in amnesic
patients indicates that some forms of learning and memory
are preserved. According to one view, amnesics suffer from
a specific inability to encode declarative knowledge about
specific events, but retain an ability to acquire procedural
knowledge. This would account for their ability to acquire
new cognitive and motor skills. Preserved priming has been
attributed to the automatic activation of declarative knowledge
structures that were stored prior to the brain damage, or to
the encoding of new episodic representations in a primitive
perceptual memory system that lacks the kinds of information
(e.g., about the meaning of an event, or its spatiotemporal
context) that would support EM.
Traumatic RA
Another form of amnesia occurs as a consequence of head
trauma. A very severe blow to the head can bruise gray matter
and shear white matter, producing both cortical and subcortical
damage that may result in AA and RA similar to that observed in
the amnesic syndrome. Even in the absence of such damage, a
blow to the head can result in a concussion, or temporary
cessation of electrical activity in the cortex and loss of conscious-
ness. The recovery of consciousness begins with the return of
simple reflexes, then the gradual return of purposeful move-
ment, and then speech (another pattern predicted by Ribot's
law). After the patient appears fully oriented, he or she will
display an AA for some time, as well as an RA for the accident
itself and the events leading up to the accident. Typically, the AA
is immediate, that is, it will start at the time of the trauma. But if
the loss of consciousness is delayed, the onset of the AA will be
delayed as well. Such lucid intervals suggest that the AA is a result
of vascular complications that may take some time to develop.
The RA in traumatic amnesia is characterized by a temporal
gradient, meaning that it is densest for events nearest the time
of the accident - yet another example of Ribot's law at work.
However, the gradient is broken by islands of memory consisting
of isolated events, not necessarily personally important, that
are remembered relatively well. The extent of the RA is corre-
lated with the extent of the AA. Although the memories covered
by the AA are permanently lost, apparently reflecting an encod-
ing deficit, the RA gradually remits. It was once thought that
this recovery began with the earliest memories and proceeded
forward, which again would be predicted by Ribot's law.
Although the most recent events are generally recovered last,
more careful studies show that the shrinkage of amnesia is
accomplished by filling in the gaps that surround the islands
of memory, leaving a final RA covering the accident itself and
the moments or minutes leading up to it, and perhaps a few
islands of amnesia. The shrinkage of amnesia clearly indicates
that traumatic RA is a disorder of retrieval, and that the islands
of memory act as anchors to support the recovery process.
However, the final RA may reflect either a loss of memory
from storage or, more likely, a disruption of consolidation.
A nontraumatic form of RA is observed in psychiatric
patients who are administered electroconvulsive therapy (ECT)
for acute affective disorder. In ECT, electrical stimulation (e.g.,
100 V, 500 mA for 500 ms), delivered from surface electrodes
applied over the temporal lobe, induces a convulsive, tonic-
clonic seizure not unlike those of grand mal epilepsy; after
a short series of such treatments (e.g., 6-10 sessions over
2-3 weeks), patients often experience a rapid return to their
normal mood state (ECT is not a cure, as episodes of
Amnesia 1 1 9
depression or mania may recur). Because they are anesthetized
when the treatment is delivered, patients experience no pain or
distress from the convulsions themselves; because they receive
muscle relaxants, the convulsions do not result in bone
trauma. However, the seizure does produce both AA and RA
as adventitious consequences (i.e., unrelated to treatment
success). The RA shows the same sort of temporal gradient
observed following concussive blows to the head. Because
there is less memory impairment (though no difference in
treatment outcome) with unilateral than with bilateral elec-
trode placement, ECT is usually delivered to the nondominant
hemisphere. The RA gradually clears up (except for the
moments before ECT is actually delivered), but memories
affected by the AA cannot be recovered.
The amnesia induced by ECT shows a dissociation between
EM and IM similar to that observed in the amnesic syndrome.
In one experiment, patients who studied a list of words within
90 min following administration of ECT showed a deficit in
recognition, but no deficit in priming on a word-stem comple-
tion test. In another study, patients who read word strings
presented in mirror-reversed fashion before delivery of ECT
later showed an advantage in reading those words, even
though they failed to recognize these words as familiar.
What about the memories covered by the final RA?
Although ECS may disrupt encoding processes, it does not
appear to remove the memory traces from storage. The relevant
evidence comes from studies of the effects on memory of ECS
administered to animals. A common research paradigm is
called one-trial, step-down, passive avoidance learning. A rat
is placed on a shelf above an electrified floor. If the animal
steps down, it receives a footshock, and jumps back up on the
shelf. Under ordinary circumstances, the animal will not return
to the floor: it learns in one trial to avoid the shock by doing
nothing. But if the animal received a dose of ECS, after it
recovers, it steps down onto the floor after being placed on
the shelf. It is as if it has forgotten all about the shock.
ECS-induced amnesia shows a temporal gradient similar to
that observed in other forms of traumatic RA. If the ECS is
delayed from the time of the original learning experience, there
is less amnesia than if it is administered immediately after-
ward. But the extent of amnesia also depends on how memory
is measured. The same time at which the amnesic animal steps
down (as if the footshock never happened), it shows a marked
increase in heart rate (as if it is afraid). Moreover, if the animal
receives reminder treatments such as tail shock in another envi-
ronment or immersion in circulating ice water, it will remain
on the shelf and avoid the floor. The desynchrony between
behavioral and psychophysiological indices of fear is analo-
gous to the dissociation between EM and IM observed in
human amnesic patients; and the effectiveness of reminder
treatments shows that at least some aspects of the forgotten
event have been preserved. Memories covered by the final RA
may never be accessible to conscious recollection, but they may
nonetheless be expressed as implicit memories.
Functional Amnesia
Clinically significant amnesias are not confined to cases of
organic brain syndrome. Psychiatrists and clinical psychologists
sometimes encounter forms of functional amnesia in a group of
mental illnesses known as the dissociative disorders. Functional
amnesia may be defined as a loss of memory that is attributable
to an instigating event that does not result in insult, injury, or
disease to the brain. Because there is no evidence of head
injury, such memory failures are also labeled as psychogenic
amnesia. In the classic formulation, which owes much to Sig-
mund Freud and Pierre Janet, traumatic stress causes amnesia
via a psychological process variously labeled as repression or
dissociation. Several forms of functional amnesia are listed in
the Diagnostic and Statistical Manual of Mental Disorders under
the category of the dissociative disorders, all of which entail a
disruption of conscious memory and identity.
In the twentieth century, the trauma-memory argument
became an important fixture of popular culture as well as
clinical folklore, but empirical evidence in its favor has been
surprisingly lacking. Careful prospective analyses indicate
that the vast majority of victims of documented trauma,
including childhood sexual abuse, remember their experiences
perfectly well. Any forgetting that is observed appears most
likely due to infantile and childhood amnesia (see later para-
graphs), the mere passage of time, or a failure to appreciate
the nature of the experience at the time it occurred (meaning
that the 'trauma' was not experienced as such). Moreover,
both human and animal studies indicate that high levels of
physiological arousal, such as would accompany traumatic
stress, enhance rather than impair memory - probably due
to the activation of (1-adrenergic receptors. The problem with
patients suffering from posttraumatic stress disorder is not
that they cannot remember their trauma; it is that they
cannot forget it.
In dissociative amnesia, the patient cannot remember specific
events, usually covering a continuous period of time, resem-
bling RA; there is no AA. As in the amnesic syndrome and
traumatic RA, the patient's fund of semantic and procedural
knowledge remain intact. Compared to traumatic RA, dissocia-
tive amnesia appears to be more extensive and longer lasting.
Clinical lore holds that dissociative amnesia can be reversed by
hypnosis or barbiturate sedation, but evidence for the reliabil-
ity of recollections produced by these techniques is largely
lacking. Because of concerns about the inherently suggestive
nature of hypnosis, most American legal jurisdictions forbid
witnesses from offering testimony based solely on hypnotically
'refreshed' memories.
Dissociative fugue entails a more extensive loss of autobio-
graphical memory, covering the whole of the person's life, a
loss and/or change in identity, and sometimes physical reloca-
tion (from which symptom the syndrome derives its name).
Such cases often come to the attention of police and health
providers when a person cannot identify himself; or when she
comes to herself in a strange place and does not know how
she got there. Interestingly, fugue patients lose self-knowledge
and autobiographical memory, but they do not seem to lose
their fund of semantic memory, or their repertoire of pro-
cedural knowledge. Upon recovery the patient is left with an
amnesia covering the events of the fugue state itself, and
retains no knowledge of whatever identity he or she may
have adopted in that state. Examination of such cases after
they are resolved sometimes reveals an instigating episode of
psychological stress.
1 20 Amnesia
In dissociative identity disorder (DID; formerly known as
MPD), two or more personalities appear to inhabit a single
body, alternating control over experience and action. One of
these personalities is often 'primary,' in that it is the one that
has been manifest the longest, and known by most other people.
Most important in the present context, the various 'alter egos'
appear to be separated by an amnesic barrier that prevents one
alter ego from gaining access to the memories of another.
In many cases, the amnesia is asymmetrical, in that Personality
A may be aware of Personality B, but not the reverse. The amne-
sia largely affects identity and autobiographical memory; as a
rule, the various personalities share semantic memory and pro-
cedural knowledge. The most widely accepted theory of DID
holds that it develops in defense against abuse, trauma, or depri-
vation in early childhood - but again, actual evidence for a causal
link between trauma and any form of amnesia is very weak.
Reports of DID were relatively common in the clinical
literature before 1920, and then virtually disappeared. There
was a resurgence of DID, bordering on epidemic, beginning in
the 1970s, but it is not clear how many of these cases - and
there were hundreds if not thousands of them tabulated in the
literature - were iatrogenic in nature, or simply misdiagnosed.
Where the alternate personalities were initially elicited through
hypnosis or other special techniques, or when an amnesic
barrier was absent, the case is especially suspect.
DID is sometimes offered as an insanity defense, claiming
that a second personality is actually responsible for crimes of
which the first personality is accused. DID does raise interesting
issues of criminal law: in principle, the actions of one personal-
ity may be outside another personality's ability to control; inter-
personality amnesia may prevent the accused from assisting in
his or her defense; and techniques intended to elicit testimony
from a personality may violate constitutional safeguards against
self-incrimination. However, DID has rarely proved successful
as a defense against criminal charges.
Several experimental studies confirm the existence of inter-
personality amnesia in DID. Thus, for example, one alter ego is
often unable to recall or recognize a list of items studied by
another. Interestingly, there is some evidence that IM may be
spared in these cases. Thus, one alter ego may show savings in
relearning, interference, transfer of training, or priming effects
involving a list studied exclusively by another one. Although
the available research is somewhat ambiguous, in general it
seems that the amnesic barrier is permeable in the case of
implicit memories.
Just as the amnesic syndrome finds its experimental analog
in drug-induced amnesia, and traumatic RA in ECT and ECS,
the functional amnesias seen in the dissociative disorders have
their laboratory parallel in posthypnotic amnesia. Following
appropriate suggestions and the termination of hypnosis,
many subjects cannot remember the events that transpired
while they were hypnotized. After the hypnotist administers a
prearranged cue, the critical memories become accessible again;
the fact of reversibility marks posthypnotic amnesia as a dis-
ruption of memory retrieval. The amnesia does not occur unless
it has been suggested (explicitly or implicitly), and memory is
not reinstated merely by the reinduction of hypnosis; thus,
posthypnotic amnesia is not an instance of state-dependent
memory. Response to the amnesia suggestion is highly corre-
lated with individual differences in hypnotizability: while
hypnotic 'virtuosos' typically show a very dense amnesia, their
insusceptible counterparts show little or no forgetting.
Like the organic amnesias, posthypnotic amnesia selectively
affects episodic as opposed to semantic or procedural memory.
The subject may forget which words appeared on a study
list, but retains the ability to use these words in speech and
writing. Skills acquired in hypnosis transfer to the posthyp-
notic state, and suggestions for amnesia have no impact on
practice effects. Subjects who learn new factual information
while being hypnotized may retain it despite suggestions for
amnesia, but these same subjects may well forget the circum-
stances in which this knowledge was acquired - a phenomenon
of source amnesia that has also been observed in the amnesic
syndrome and elsewhere. Finally, there is good evidence that
priming effects are preserved in posthypnotic amnesia. That is,
subjects who cannot remember words from a study list are more
likely to use those words as free associations or category in-
stances than would be expected by chance. Thus, posthypnotic
amnesia shows the familiar dissociation between EM and IM.
Because functional amnesia occurs in the absence of brain
damage, and because posthypnotic amnesia occurs in response
to suggestion, questions inevitably arise about malingering,
simulation, and behavioral compliance. Unfortunately, it is
difficult to distinguish between genuine and simulated amne-
sia in either clinical or experimental situations. Claims of
amnesia are readily accepted when there is palpable evidence
of brain damage. It should be understood, however, that evi-
dence of a significant interpersonal or sociocultural compo-
nent does not necessarily mean that functional amnesia is
faked. Rather, it means that functional amnesia is complex.
Hypnosis may be a state of altered consciousness, but it is
also a social interaction; thus, it should not be surprising to
discover that the subject's response to amnesia suggestions
will be influenced by the precise wording of the suggestion,
the discourse context in which it is embedded, the subject's
interpretation of the hypnotist's words, and perceived social
demands. The social context may be important in the organic
amnesias, but its role is magnified in their functional
counterparts.
Amnesia Through the Life Span
Some forms of amnesia occur naturally in the course of psy-
chological development. For example, adults rarely remember
much from early childhood: the earliest memory is typically
dated between the third and fourth birthdays, and is limited to
a relatively small number of isolated fragments until about 5 or
7 years of age. The appearance of childhood amnesia is not
merely an artifact of the long retention interval between child-
hood encoding and adult retrieval: something special seems to
happen to memories for childhood events. Infantile and child-
hood amnesia affects only memories for personal experiences.
Children acquire a vast fund of information, and a consider-
able repertoire of cognitive and motor skills, which they carry
into adulthood. Whether this selectivity reflects merely the
effects of constant rehearsal, or the dissociation between EM
and IM similar to that observed in source amnesia, is not clear.
Infantile amnesia, covering the first year or two of life, may
be attributed at least in part to the lack of language and to the
Amnesia 121
immaturity of the neocortex and other critical brain structures.
However, the exact mechanism for childhood amnesia, cover-
ing the years after the second birthday, remains uncertain.
The classic explanation for childhood amnesia was proposed
by Freud. In his view, during the phallic stage of psychosexual
development, the child resolves the Oedipus complex by
repressing infantile sexual and aggressive impulses, as well as
any thoughts, images, and memories that might be related to
them. Since (according to the theory) the young child's entire
mental life is concerned with these topics, all memories of early
childhood are repressed - except a couple of banal screen
memories that aid repression by giving the person something
to remember. Recall that the major goal of psychoanalysis is to
lift the repressive barrier, so that patients can acknowledge and
cope realistically with their primitive instinctual urges. Other
theories emphasize the relationship between cognitive pro-
cesses employed at encoding and retrieval. For example, Ernst
Schachtel proposed that memories encoded by preoedipal,
'primary-process' modes of thought cannot be retrieved by
postoedipal, 'secondary-process' schemata. A similar account
can be offered from Piaget's perspective, emphasizing the
incompatibility between sensory-motor and preoperational
encodings, and the retrieval processes characteristic of concrete
and formal operations. Note that all these theories predict that
memories of childhood experience should be accessible to
young children, who have not undergone the 'five-to-seven
shift' (so named because of the major cognitive change occur-
ring between these ages) between preoperational thought and
concrete operations. In contrast, some theorists have argued
that young children simply do not possess the information-
processing capacity - specifically, the ability to pay attention to
two things at once, like an event and its episodic context -
required to encode retrievable memories. In this case, the
prediction is that children will know little more about their
childhood histories than adults do.
Although infantile and childhood amnesia are often attrib-
uted to autochthonous aspects of cognitive and neural devel-
opment, it is clear that the child's interactions with other
people are extremely important determinants of whether he
or she will remember some past event. After all, as Ulric Neisser
has pointed out, the 'five-to-seven year shift' is not simply
a matter of moving from the preoperational period to con-
crete operations (or, for that matter, from before to after
the acquisition of a theory of mind). It is also when the
child first goes to school, and moves into an environment
that is more structured with respect to time and place - thus
affording the child an opportunity to distinguish one event
from another. Even before the child enters school, research by
Katherine Nelson, Robyn Fivush, Judith Hudson, and others
underscores the important role played by joint reminiscence
between child and parent in shaping the child's appreciation
of narrative structure, including the causal as well as temporal
relations among events, thus strengthening individual mem-
ories and connecting them both with each other and with
the present.
At the other end of the life cycle, it appears that even the
healthy aged have difficulty learning new information and
remembering recent events. Normal aging has little effect on
primary or short-term memory, as reflected in digit span or the
recency component of the serial-position curve; but it has
substantial effects on secondary or long-term memory, espe-
cially after moderately long retention intervals. Again, the defi-
cit primarily affects episodic memory: the elderly do not lose
their fund of semantic information (although they may
become slower on such semantic-memory tasks as word-
finding); and their repertoire of procedural knowledge remains
intact, provided that they have been able to maintain these
skills through practice.
At the same time, it should be noted that episodic-semantic
comparisons almost inevitably confound the type of memory
with retention interval. Memories of recent experiences have, by
definition, been encoded recently; most semantic knowledge
was acquired while the individual was relatively young. Sur-
prisingly, little is known about the ability of older individuals
to learn new vocabulary or acquire new world knowledge.
The aged do show impairment in episodic memory for remote
events, but it is not clear whether this reflects age differences
in retrieval processes, or simply the effects of the retention
interval and opportunities for proactive and retroactive
interference.
A relatively recent topic in research on aging memory com-
pares EM and IM. Compared to the young, the aged show
definite impairments on EM (especially free recall, less so on
recognition); but they show less deficit, or none at all, on IM
tasks such as stem completion. Part of the reason for their
problems with EM may lie in the difficulty that the elderly
have in processing contextual information. Spatial context,
temporal context, and source are necessary for distinguishing
one event from another, and thus crucial to conscious recollec-
tion. Whether this difficulty is specific to contextual features of
events, or merely a reflection of a more general limitation on
cognitive resources, is unclear.
Memory problems are confounded in the dementing illnesses
often associated with aging - for example, AD. The severe
memory problems associated with AD are likely related to the
increase of neuritic plaques and neurofibrillary tangles, partic-
ularly in medial-temporal regions of the brain. These changes,
as well as neuronal loss and depletion of neurotransmitters in
other cortical and subcortical areas, especially the hippocam-
pus and other medial-temporal lobe structures, contribute
to the extensiveness of the disease process. Both AA and RA
emerge early in the course of these diseases, and progressively
worsen. In contrast to the amnesic syndrome, however, the
memory deficit in dementia affects 'short-term' as well as
'long-term' memory, and forms part of a larger cluster of def-
icits affecting a broad swath of cognitive and emotional life,
including impairments in semantic and procedural memory as
well as episodic memory. In the latter stages of their illness,
demented patients may show anosognosia, or a lack of aware-
ness of their deficits.
Does the abnormal forgetting observed in aging and
dementia extend to IM as well as EM? Research on this ques-
tion is still at a very early stage, but already it seems fairly clear
that IM is relatively spared in normal aging. Thus, elderly
subjects fail to recognize studied words, but show priming
effects on word-fragment completion. With respect to AD and
other forms of dementia, however, some controversy remains.
There is some evidence of intact motor-skill learning in AD
patients, but there is also evidence of impaired performance on
priming tasks. The issue is complicated by the fact that AD is
1 22 Amnesia
a progressive illness. Although impairments in EM may be
observed quite early in the course of the disease, deterioration
of IM may wait until later stages.
Amnesias of Everyday Experience
Amnesia is a symptom of neurological or psychiatric disorder,
but it is also something that occurs in the ordinary course of
everyday living. The most familiar example is sleep. A great deal
transpires while we are asleep, including events in the external
environment and endogenous activity such as dreams, night-
mares, and (in some cases) episodes of somnambulism (sleep-
walking) and somniloquy (sleeptalking), but virtually none of
this is remembered in the morning. In fact, our inability to
remember what has been happening is often the phenomeno-
logical basis for inferring that we have been asleep. Similarly,
attempts at sleep learning have been almost uniformly unsuc-
cessful, leading investigators to conclude that we are able to
learn during sleep only to the extent that we stay awake.
A paradox here is that a large body of evidence now indicates
that sleep plays an important role in the consolidation of
presleep memories.
Most investigators explain sleep-induced amnesia in terms
of an encoding deficit or consolidation failure. According to
this view, the low levels of cortical arousal characteristic of
sleep effectively impair complex information-processing func-
tions. Thus, events in the environment are not noticed, relevant
information in memory is not retrieved, and traces of new
experiences are not encoded in retrievable form. Some evi-
dence favoring this view comes from studies of memory for
dreams. Sleepers who are awakened during REM sleep almost
invariably report a dream, apparently by virtue of retrieval
from primary memory; but dreams are rarely reported upon
awakening in the morning, which requires access to trace
information in secondary memory. However, subjects will
remember a dream in the morning if they awaken directly
out of REM sleep. And dreams reported during REM awaken-
ings will be accessible in the morning, provided that the
dreamer has remained awake long enough to rehearse the
dream before returning to sleep.
Most evidence of sleep-induced amnesia comes from stud-
ies of EM, leading to speculation that evidence of memory for
sleep events, including successful sleep learning, might be
obtained with measures of IM. Research on this topic has
only just begun, but the available evidence is negative. When
care is taken to insure that there is no evidence of cortical
arousal indicative of awakening, subjects show neither EM
nor IM for events that occurred while they were sleeping.
Even if positive evidence for sleep learning were obtained,
it would almost certainly not be as efficient as learning in the
normal waking state.
Amnesia is also an important component of general anes-
thesia induced in surgical patients. Clinically, the success
of general anesthesia is indicated by the patient's lack of
response to instructions, suppression of autonomic and skele-
tal responses to incisions and other surgical stimuli, and
absence of retrospective awareness of pain and other events
occurring during surgery. Thus, by definition, amnesia is a
consequence of adequate general anesthesia. But, as with
sleep, the amnesia is always assessed in terms of EM, leaving
open the possibility that even adequately anesthetized patients
might show IM for surgical events. Some anecdotal evidence
favoring this proposition is provided by occasional cases in
which patients awaken from surgery with an inexplicable dis-
like of their surgeon - an attitudinal change which is plausibly
traced to unkind remarks made about the patient by members
of the surgical team.
In recent years, this question has been the object of consid-
erable investigation, and in fact research employing paradigms
derived from studies of the amnesic syndrome has sometimes,
but not always, provided evidence of spared IM. Thus, patients
who are presented a list of words during surgery, sometimes
show significant priming effects. Such effects are not always
obtained, however; and even when they are obtained, they are
relatively small. Certainly the scope of information processing
during general anesthesia cannot compare to what is possible
when the patient is awake and properly oriented; for example,
IM after anesthesia may well be limited to the processing of
the physical properties of stimuli, but not their meaning.
What accounts for the different outcomes across the available
research is not clear. Perhaps some anesthetic agents impair EM
but spare IM, while others impair both. Such a result might
yield interesting insights about the biological foundations of
memory.
Theoretical and Practical Implications
Research on amnesia is intrinsically interesting, but it also has
theoretical and pragmatic implications. At the theoretical
level, amnesia engages our attention because it seems to
carve nature at its joints. Amnesia is selective, and the differ-
ence between those aspects of memory that are impaired in
some form of amnesia, and those that are spared, promises to
provide information about the processes underlying memory
functioning and the organization of memory into different
systems. Such conclusions are based on the logic of dissociation.
In single dissociations, Variable A affects performance on Task
Y but not Task Z; in double dissociations, Variable A affects Y
but not Z, while Variable B affects Z but not Y; in reversed
associations, changes in A increase Y and decrease Z, while
changes in B decrease Y and increase Z; in stochastic indepen-
dence, performance on Task Y is uncorrelated with perfor-
mance on Task Z. All other things being equal, differences
such as these suggest that the tasks in question differ in
qualitative terms. If they were only quantitatively different,
they would be correlated with each other, and influenced by
the same variables.
Such dissociations are commonly observed in amnesia. For
example, the fact that the amnesic syndrome affects the recency
portion of the serial-position curve, but not the primacy com-
ponent, has been cited as evidence that primary (short-term)
and secondary (long-term) memory are qualitatively different
memory systems, perhaps with different biological substrates
(one affected by the brain lesion, the other not). Evidence
from amnesia also has been used to support other structural
distinctions: between declarative and procedural knowledge
and - within the domain of declarative knowledge - between
episodic and semantic memory. Thus, amnesic patients have
Amnesia 1 23
difficulty learning new factual information, but retain an ability
to acquire new cognitive and motor skills; and if they do retain
new factual knowledge, they display amnesia for the circum-
stances in which this information was acquired. Logic and expe-
rience tell us that when something breaks, it does so along
natural boundaries, which form lines of least resistance. When
a disorder of memory separates past memory from new learning,
procedural and declarative knowledge, or episodic and semantic
memory, it tells us that these distinctions, conjured in the
minds of theorists, actually mean something in the real world.
The fact that these kinds of dissociations are observed in all
sorts of amnesia - not just the amnesic syndrome, but in trau-
matic RA, psychogenic amnesia, and posthypnotic amnesia
as well - strengthens the conclusion that the theoretical distinc-
tions are psychologically and biologically valid.
Of particular interest in recent theory are the various dis-
sociations between explicit and implicit expressions of episodic
memory. To date, three broad classes of theories have been
proposed to explain these dissociations; each has several exem-
plars. According to the activation view, the activation, by a
current event, of preexisting knowledge representations is suf-
ficient for IM; but EM requires elaborative activity, in which
individually activated structures are related to each other.
According to the processing view, IM is an automatic conse-
quence of environmental stimulation, while EM occurs by
virtue of controlled processes that are limited by attentional
resources. According to the memory systems view, IM reflects the
activity of a perceptual representation system that holds infor-
mation about the form and structure of the objects of percep-
tion, and EM reflects the activity of an episodic memory system
that represents knowledge about the meaning of events and the
context in which they occur.
Research on the amnesic syndrome, including studies of
both human patients and animal models, indicates that the
medial-temporal lobe, including the hippocampus, entorhi-
nal cortex, and perirhinal and parahippocampal cortex, forms
the biological substrate of EM. But the diencephalic form of
amnesic syndrome seems to indicate that the mammillary
bodies and the dorsomedial nucleus of the thalamus are also
critical for memory. And, of course, the hippocampus itself is
a complex structure, with many elements that may each play a
special role in memory processing. As research continues,
investigation of amnesia will make a unique and valuable con-
tribution to understanding the relation between EM and IM,
and the biological foundations of each.
At the same time, evidence of preserved memory func-
tioning offers new insights concerning amelioration and
rehabilitation in cases of amnesia. Loss of EM has debilitat-
ing consequences for afflicted individuals in everyday life.
Amnesic patients are often unable to keep track of events,
remember appointments or schedules, engage in educational
or vocational pursuits, or manage home activities. Attempts
at rehabilitation have frequently focused on restoration of
damaged EM processes either through the use of repetitive
drills or by teaching patients mnemonic strategies such
as visual imagery or verbal elaboration. These retraining
attempts have met with limited success: there is no evidence
that exercising damaged neural or cognitive mechanisms
leads to positive outcomes; and although patients have
sometimes been able to acquire a few pieces of information
by using mnemonic techniques, they do not use the strate-
gies spontaneously in everyday life.
On the other hand, rehabilitation strategies that have
focused on providing compensatory devices designed to
bypass problems in daily life have been somewhat more
promising. External aids such as notebooks, diaries, alarm
watches, and environmental labels have enabled some amnesic
patients to function somewhat more independently, although
use of such devices often requires considerable amounts of
training and practice. The hand-held computer, potentially
a powerful prosthetic for people with memory impairments,
has yet to be extensively used for this purpose probably
because of the problems in teaching amnesic individuals how
to use such a device.
The finding that IM and procedural knowledge often
remain intact even in cases of severe amnesia has recently
prompted researchers to begin to explore ways in which
these preserved processes might be exploited beneficially for
rehabilitation purposes. Cuing techniques, which take advan-
tage of amnesic patients' ability to respond normally to word
stem or fragment cues, have been used successfully to teach
individuals new factual information such as vocabulary as
well as procedural tasks such as data entry and word proces-
sing. Continued research in this direction, paralleling more
theoretically based research concerning preserved memory
functions in amnesia, should enable further progress toward
improving the ability of amnesic individuals to function
effectively in their everyday lives.
See also: Aging and Cognition; Alzheimer’s Disease; Amnesia and the
Brain; Dissociative Disorders; Episodic Memory; Hippocampal
Formation; Hypnosis; Memory; Memory, Neural Substrates;
Posttraumatic Stress Disorder.
Further Reading
Balso J V and Shimamura AP (2002) Frontal lobes and memory. In: Baddeley AD,
Kopelman MD, and Wilson BA (eds.) Handbook of Memory Disorders. London:
Wiley.
Brand M and Markowitsch HJ (2006) Amnesia I: Clinical and anatomical issues.
In: Farah MJ and Feinberg TE (eds.) Patient-Based Approaches to Cognitive
Neuroscience, 2nd edn., pp. 289-301. Cambridge, MA: MIT Press.
Cermak LS (ed.) (2000) In: Memory and Its Disorders, 2nd edn., vol. 2. Amsterdam:
Elsevier.
Fujii T, Moscovitch M, and Nadel L (2000) Memory consolidation, retrogrde
amnesia, and the temporal lobe. In: Cermak LS (ed.) Memory and Its
Disorders. Handbook of Neuropsychology, 3rd edn., vol. 2, pp. 223-250.
Amsterdam: Elsevier.
Glisky EL (2007) Changes in cognitive function in human aging. In: Riddle DR (ed.)
Brain Aging: Models, Methods, and Mechanisms, pp. 3-20. Boca Raton, FL:
CRC Press.
Glisky EL and Glisky ML (2008) Memory rehabilitation in older adults. In: Stuss DT,
Winocur G, and Robertson IH (eds.) Cognitive Neurorehabilitation:
Evidence and Application, 2nd edn., pp. 541-561 . New York: Cambridge
University Press.
Keane MM and Verfaellie M (2006) Amnesia II: Cognitive issues. In: Farah MJ and
Feinberg TE (eds.) Patient-Based Approaches to Cognitive Neuroscience, 2nd edn.,
pp. 303-314. Cambridge: MIT Press.
Kihlstrom JF (2006) Trauma and memory revisited. In: Uttl B, Ohta N, and
Siegenthaler AL (eds.) Memory and Emotions: Interdisciplinary Perspectives,
pp. 259-291. New York: Blackwell
1 24 Amnesia
Kihlstrom JF and Cork RC (2007) Anesthesia. In: Velmans M and Schneider S (eds.)
A Companion to Consciousness, pp. 628-639. Oxford: Blackwell.
Kihlstrom JF and Schaefer DL (2000) Functional amnesia. In: Boiler F and Grafman J
(eds ) Handbook of Neuropsychology, 2nd edn., vol. 2, pp. 409-427. Amsterdam:
Elsevier.
McGaugh JL (2004) The amygdala modulates the consolidation of memories of
emotionally arousing experiences. Annual Review of Neuroscience 27: 1-28.
Walker MP and Stickgold R (2006) Sleep, memory, and plasticity. Annual Review of
Psychology 57: 139-166.
Relevant Websites
http://www.nia.nih.gov/Alzheimers/- Alzheimer’s Disease Education and Research (NIH).
http://www.memory-disorders.org/ - Memory Disorders Research Society.
http://www.memorylossonline.com/ - Memory Loss and the Brain (Newsletter).
http://www.nia.nih.gov/Alzheimers/ - Memory Loss at the Movies.
http://www.ninds.nih.gov/disorders/tbi/tbi.htm - Traumatic Brain Injury Information
Page (NINDS).
Amnesia and the Brain
E Race and M Verfaeliie, Boston University School of Medicine, Boston, MA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Amnesia Pathological memory loss resulting from brain
injury, illness, or psychological disturbance.
Anterograde amnesia Impaired memory for facts or events
encountered after the onset of amnesia (postmorbid
memory loss).
Confabulation Falsification of memory without the intent
to mislead.
Consolidation Process by which information is
assimilated and stabilized into memory. Consolidation
includes both fast processes operating at the cellular
level and slow processes operating at the systems
level.
Declarative memory A form of memory characterized by
the ability to consciously bring to mind facts and events.
Declarative memory is dependent on the integrity of the
limbic system. Episodic memory and semantic memory are
both forms of declarative memory.
Episodic memory Memory for personally experienced
events or 'episodes' that are specific in time and place.
Limbic system System of brain structures associated with
memory function, including the medial temporal lobes,
diencephalon, mammillary bodies, and fornix.
Medial temporal lobes Medial region of the temporal lobes
associated with memory function. Subregions include the
hippocampus, parahippocampal gyrus, entorhinal cortex,
and perirhinal cortex.
Nondeclarative memory Memory expressed through
behavioral performance without conscious memory
content. Includes classical conditioning, skill learning, and
repetition priming.
Retrograde amnesia Impaired memory for information
acquired prior to the onset of amnesia (premorbid
memory loss).
Semantic memory Memory for general knowledge (facts or
concepts) that is impersonal and context-free.
The Amnesic Syndrome
Amnesia can be defined as a special case of memory loss
that is distinct from ordinary forgetting. The amnesic syn-
drome is characterized by a profound disorder of memory in
the context of preserved general intelligence and other cogni-
tive abilities such as language, perception, and attention. Per-
manent amnesia can be differentiated from more common
forms of memory loss, such as memory loss associated with
aging or depression. In particular, amnesia is characterized by
anterograde memory loss, in which patients cannot remember
events that occur after the onset of amnesia. Retrograde amne-
sia, in which patients cannot remember experiences or informa-
tion acquired before the onset of the illness, is also commonly
observed, though the extent of retrograde amnesia is more
variable. Occasionally patients present with focal retrograde
amnesia, but this presentation has been argued to be psycho-
genic in nature.
The most widely studied clinical case of amnesia is patient
H.M. In 1953, H.M. underwent surgery in which his medial
temporal lobes (MTLs) were bilaterally resected to relieve
medically intractable epileptic seizures. Though the surgery
succeeded in alleviating his epileptic symptoms, H.M. immedi-
ately showed profound anterograde amnesia that persisted until
his death in 2008. In the 55 years following his surgery, H.M.
demonstrated a remarkable inability to learn and intentionally
retrieve new information. His memory impairment was evident
in many types of memory tests (e.g., cued recall, free recall, and
recognition) across multiple domains of information (includ-
ing verbal, visual, auditory, and somatosensory). In everyday
life, H.M. was unable to recall conversations that he had with a
visitor 30 min ago, or the fact that he had a visitor at all.
He could not acquire new information about salient life
events occurring after his surgery, such as the death of his
father, and he also had difficulty recalling premorbid mem-
ories, with most of his personal memories deriving from the
age of 1 6 or earlier. Interestingly, despite these severe memory
problems, H.M.'s IQ was above average (118 on the Wechsler
Adult Intelligence Scale) and he performed normally on stan-
dard tests of perception, short-term memory, and language
comprehension. The specificity of H.M.'s impairment to long-
term memory provided strong evidence that the MTL is critical
for normal human memory function.
While amnesia can also have a psychogenic origin, the focus of
this article is on organic amnesia caused by nonprogressive brain
damage to the MTL and related limbic system structures. The
laterality of damage to the limbic system results in modality-
specific memory impairments, with damage to left-lateralized
limbic structures producing primarily impaired verbal memory
(such as story recall and list learning) while damage to right-
lateralized limbic structures produces impaired visual memory
(such as memory for figures or designs) and sometimes, spatial
memory. While left-sided lesions seem to produce more func-
tionally significant memory problems in humans, likely due to
the importance of verbal memory and language to human
behavior, bilateral lesions are required for permanent global
amnesia. A range of vascular, infectious, and traumatic pro-
cesses can cause damage to the limbic system. The following
section describes the most common etiologies of permanent
amnesia.
125
1 26 Amnesia and the Brain
Etiologies of Amnesia
Herpes Simplex Encephalitis
Herpes simplex encephalitis (HSE) is a viral infection that
causes hemorrhagic lesions in the brain. Neural damage due
to HSE typically occurs in MTL regions, including the hippo-
campus, entorhinal cortex, perirhinal cortex and parahippo-
campal cortex. Damage can also extend into the insular cortex,
basal forebrain, and anterolateral and inferior temporal lobe
regions. Anterograde amnesia is a prominent feature of HSE
and can be accompanied by aphasia and agnosia when neural
damage includes the anterolateral and inferior temporal lobes.
In addition, lesions extending into lateral temporal regions can
result in extensive retrograde amnesia.
Patient S.S. provides a striking demonstration of the mem-
ory impairments following HSE. Patient S.S. became densely
amnesic in 1971 after developing HSE that caused bilateral
lesions in anterolateral and medial portions of his temporal
lobes, as well as lesions in his insula and putamen. S.S. has not
been able to form any new declarative memories since the time
of his illness. He cannot retain information regarding signifi-
cant family matters or recent public events, and he cannot
acquire new general knowledge such as vocabulary introduced
into the English language since the onset of his illness. S.S. also
has extensive retrograde amnesia for autobiographical and per-
sonal semantic information that includes most of his adult life.
Despite these severe memory impairments, S.S. is of superior
intelligence and performs well on tasks of short-term memory,
language, executive function, and deductive reasoning skills.
Anoxia
Anoxic injury results from a loss of oxygen to the brain and can
be caused by a variety of conditions such as cardiac arrest,
respiratory distress, or carbon monoxide poisoning. Certain
brain regions, such as the hippocampus, are particularly vul-
nerable to anoxic injury due to their physical location and
biochemical makeup. Selective memory impairment following
anoxia can vary from mild to severe, depending on the extent
of MTL damage, and can include both anterograde and retro-
grade amnesia. Anoxic brain damage that extends outside the
MTL to regions such as the basal ganglia, thalamus, white
matter, or other neocortical regions can affect additional cog-
nitive and perceptual abilities. In addition, anoxic injuries
occurring shortly after birth can cause bilateral hippocampal
atrophy leading to developmental amnesia that presents as
impaired learning and retention in the face of preserved atten-
tion, reasoning, and visuospatial skills.
Stroke
Stroke can produce amnesia by blocking blood supply to vari-
ous components of the limbic system, including limbic system
white matter. Stroke involving the bilateral posterior cerebral
artery (PCA) can cause lesions in the posterior hippocampus,
parahippocampal gyrus, and occipitotemporal gyrus and can
result in a dense amnesic syndrome. Anterograde amnesia
following bilateral PCA stroke can include impairment in ver-
bal recall and recognition as well as in visuospatial memory.
Patients can also present with retrograde amnesia. Memory
problems can also occur following left-lateralized PCA lesions,
and can affect verbal and visual information either transiently
or permanently.
Stroke can also cause amnesia by damaging tissue in the
thalamus. Strokes affecting blood supply to the anterior or
medial regions of the thalamus most commonly occur due to
small vessel occlusive disease or hypertensive hemorrhages and
can damage the anterior thalamic nuclei as well as thalamo-
frontal connections. Executive dysfunction frequently accom-
panies memory impairments, and patients often display worse
performance on recall than on recognition tests. The extent and
persistence of retrograde amnesia is variable.
Several features of stroke make it an important source of
information about the nature of memory impairments in
amnesia. First, there are usually no pre-illness neurological
issues to complicate interpretation of the effects of stroke.
Second, the abrupt onset of stroke avoids concerns about
adaptive changes in the brain before initial assessment. Finally,
stroke generates relatively discrete lesion boundaries without
co-occurring diffuse injury, enabling regional effects to be
mapped with greater specificity.
Wernicke-Korsakoff Syndrome
Wernicke-Korsakoff syndrome (WKS) is caused by thiamine
deficiency and is most commonly observed as an effect of
alcohol abuse in populations over 40 years of age. Because
thiamine helps to produce energy needed for proper neuronal
function, insufficient thiamine can cause neuronal damage or
death. The thalamus and mammillary bodies are particularly
sensitive to damage due to thiamine deficiency.
Diagnosis of WKS is given when an individual with a his-
tory of chronic, heavy drinking or malnutrition presents with
anterograde amnesia. This diagnosis can be supported by neu-
roimaging or autopsy findings showing degeneration of the
thalamus and mammillary bodies and loss of brain volume
around the fourth ventricle. In the acute phase, patients exhibit
a confused state, eye movement disturbances, and ataxia. In the
chronic phase, patients exhibit dense anterograde amnesia that
is attributed to neural damage in the anterior and dorsomedial
nuclei of the thalamus and mammillothalamic tract. Confabu-
lation and source memory errors can also occur in WKS and are
thought to reflect additional frontal dysfunction. In addition to
anterograde amnesia, many WKS patients also have profound
retrograde amnesia that is characterized by a temporal gradient
for autobiographical information (i.e., greater impairment for
recent compared to remote personal memories). However,
interpretation of retrograde amnesia is complicated by premor-
bid lifestyle characteristics and frontal dysfunction that may
interfere with memory encoding and retrieval.
Anterior Communicating Artery Aneurysm
Rupture of an anterior communicating artery (ACoA) aneu-
rysm can cause lesions in the basal forebrain, which are the
origin of the cholinergic pathways to the hippocampus, the
striatum, and the frontal lobes. Amnesia with ACoA aneurysm
is likely due to disruption of hippocampal function, but con-
troversy remains about the specific role of other lesioned
regions. Lesions in the basal forebrain are usually bilateral
Amnesia and the Brain 1 27
and if damage includes the basal forebrain, striatum, and
frontal lobes, the resultant amnesia is severe and persistent.
Approximately 1 8% of ACoA aneurysm survivors will have
some form of cognitive deficit, of which severe amnesia is the
most common. Anterograde amnesia is characterized by
delayed recall deficits for verbal and nonverbal material and
memory recall is worse than recognition. Severe retrograde
amnesia is also common, but is not universally observed. The
clinical presentation of patients suffering from ACoA aneurysm
can be complicated by significant executive deficits and promi-
nent confabulation. Of all the etiologies of amnesia, patients
with ACoA aneurysm are the most likely to be unaware of their
memory deficits, which can cause safety and behavioral
problems.
Information-Processing Descriptions of Amnesia
Though functional heterogeneity exists amongst the different
etiologies of amnesia, comparison across cases with different
lesion locations and associated deficits has enabled identifica-
tion of the core information-processing deficits in amnesia. The
following sections describe the core information-processing
characteristics of anterograde and retrograde amnesia.
Anterograde Amnesia
Anterograde amnesia includes impaired memory for person-
ally experienced events (episodic memory) as well as imper-
sonal facts and concepts (semantic memory). Together, these
forms of memory are categorized as declarative (explicit) mem-
ory and can be distinguished from nondeclarative (implicit)
forms of memory that are largely preserved in amnesia. To
demonstrate the distinction between impaired declarative
and preserved nondeclarative memory in amnesia, consider
the example of learning to ride a bicycle. While amnesic
patients may not be able to recall specific instances when
they have ridden a bicycle in the past (impaired declarative
memory), they can nonetheless demonstrate the motoric skill
of riding a bicycle (preserved nondeclarative memory). This
example reflects one form of nondeclarative memory, proce-
dural memory, that is intact in amnesia: amnesic patients are
able to retrieve premorbidly learned motor skills and are able
to acquire new ones, even if they are unable to retain declara-
tive knowledge of having learned these skills. Other forms of
nondeclarative memory that are preserved in amnesia include
conditioning (provided that the conditioned stimulus and the
unconditioned stimulus are overlapping) and perceptual and
conceptual repetition priming. Repetition priming is the
behavioral facilitation that occurs when processing a repeated
compared to a novel stimulus, even without explicit awareness
of the repetition. Perceptual repetition priming refers to facili-
tated processing of visual information, such as stimulus form,
when visual information is repeated after an initial occurrence.
Conceptual repetition priming refers to facilitated processing
of stimulus meaning, independent from stimulus form, when
analysis of stimulus meaning is repeated. The preservation of
nondeclarative memory functions in amnesia is thought to
reflect intact neural function in regions outside the limbic
system, such as the basal ganglia and premotor/motor cortices
(motor learning), cerebellum (conditioning), and inferior and
lateral temporal cortex (perceptual and conceptual priming).
In the case of perceptual and conceptual priming, neocortical
regions in inferior and lateral temporal cortex that store
perceptual and conceptual representations are thought to be
tuned with repetition such that previously accessed informa-
tion can be more effectively activated and retrieved.
In contrast to largely preserved nondeclarative memory
function, amnesic patients demonstrate profound declarative
memory impairments on explicit memory tests requiring con-
scious retrieval of recent experiences. Explicit memory tests
can take the form of recall or recognition. Although both recall
and recognition tests require conscious retrieval of mnemonic
information, they differ with respect to their processing
demands and can be differentially impaired in amnesia.
According to dual-process models, recognition memory
relies on two distinct processes: recollection and familiarity.
Recollection refers to the effortful and intentional remember-
ing of a prior encounter with its contextual source, whereas
familiarity refers to the subjective feeling of knowing that an
item was previously encountered without any knowledge of
the context in which the item was acquired. Recollection and
familiarity can be differentially impaired in amnesia depend-
ing on the locus of MTL damage. Amnesic patients with
damage limited to the hippocampus have been described as
having impaired recollection but preserved familiarity. Consis-
tent with this theory, these patients perform poorly on free
recall tasks but demonstrate intact performance on item recog-
nition tasks. Recall deficits with hippocampal damage are
thought to reflect impaired binding of contextual information
into the memory trace, a process that is normally mediated by
the hippocampus and makes recollection possible.
In contrast, when lesions extend into extrahippocampal
regions of MTL, such as perirhinal cortex, patients demonstrate
more severe declarative memory impairments that include
both recollection and familiarity. These impairments are evi-
dent in both recall and recognition tests, though impairments
in recall may be more severe than impairments in recognition.
Extrahippocampal regions of MTL are thought to support
memory for individual items without their accompanying
episodic context.
While neuroimaging and neurophysiological studies pro-
vide convergent evidence for a division of labor in the MTL
according to regions that support familiarity and those that
support recollection, functional dissociations along these lines
have not been consistently observed. An alternative proposal is
that the hippocampus supports both recollection and familiar-
ity (supporting both item recognition and recall) and that MTL
subregions functionally dissociate according to whether they
support stronger memories (hippocampus) or weaker mem-
ories (extrahippocampal regions of MTL). While the precise
contributions of different MTL lesions to functional impair-
ments in amnesia remain to be specified, episodic memory
deficits can be considered to be a core characteristic of amnesia.
In addition to pervasive deficits in acquiring episodic infor-
mation, new semantic learning can also be impaired in amne-
sia when MTL lesions are extensive. In such cases, patients
demonstrate minimal ability to acquire new facts or con-
cepts, particularly when this new information is difficult to
incorporate into existing knowledge structures. For example,
1 28 Amnesia and the Brain
patient H.M. was unable to acquire the meaning of new words
that had entered into the vocabulary since the onset of his
amnesia in 1953 (e.g., words such as 'granola' or 'jacuzzi').
The magnitude of semantic learning deficits in amnesia has
been found to correspond to the amount of MTL damage.
Patients with lesions limited to the hippocampus can acquire
some degree of new semantic knowledge, including vocabulary
words, indicating that new semantic learning can be supported
by regions outside the hippocampus proper (e.g., perirhinal,
parahippocampal, or lateral temporal cortex). Patients with
larger MTL lesions extending into lateral temporal lobe have
more severe impairments in semantic learning. Though these
patients can acquire small amounts of new semantic knowl-
edge with very extensive repetition, likely reflecting the abil-
ity of intact lateral temporal regions to gradually acquire
new semantic knowledge after a great number of encoding
episodes, such learning is typically very inflexible.
While the extent and location of lesions within the MTL
greatly impacts the nature of anterograde memory impair-
ments in amnesia, even more striking are the cognitive symp-
toms that occur when frontal deficits are superimposed on the
core amnesia. Frontal lesions impair executive processes that
contribute to memory, such as the ability to mentally manipu-
late or organize information to be encoded into memory and
the ability to initiate and evaluate memory search. Executive
deficits can lead to impairments in temporal or contextual
aspects of memory and can lead to high levels of intrusions
in recall tests or false endorsements in recognition tests. Rather
than exhibiting a qualitatively different form of amnesia,
patients with additional frontal lesions display a pattern of
deficits consistent with a combination of a classic amnesic
syndrome and additional problems associated with a frontal
dysexecutive syndrome.
Retrograde Amnesia
Though many premorbid memories are preserved in amnesia,
some degree of retrograde memory loss is often observed. Retro-
grade amnesia for events immediately preceding the onset of
the brain lesion is particularly pronounced. According to Ribot' s
law, retrograde amnesia is characterized by a time gradient in
which the memory impairment is inversely related to the age of
the to-be-recalled event. By this view, memory is worse for more
recent compared to more remote memories. While patients with
MTL damage reliably demonstrate such a temporal gradient of
retrograde memory loss for semantic information, with relative
sparing of remote memory compared to recent memory, MTL
damage does not spare remote memories in all cases, and pat-
terns of retrograde amnesia appear to depend on the type of
memory under investigation. In particular, when personal epi-
sodic (autobiographical) memories from different past time
periods are probed (e.g., memory for events occurring during
childhood, during early adulthood, and within the past 5 years),
memory impairments can extend for decades and in some cases
can include a patient's entire lifespan.
Temporally graded retrograde amnesia has been interpreted
in terms of standard models of consolidation. In these models,
memories are thought to depend on the MTL, and particularly
the hippocampus, until they can be transferred to more stable
cortical storage systems. The hippocampus plays a time-limited
role in memory and is crucial only for the initial consolidation
of declarative memories. Thus, while MTL mechanisms are
critical for the retrieval of recent memories that have yet to be
consolidated, the MTL is no longer needed to retain and
recover memories once consolidation is complete.
Observations of extensive temporal gradients of autobio-
graphical memory loss are difficult to incorporate into consol-
idation theory, as the loss of such memories extends well
beyond the time frame during which biological consolidation
is likely to operate. An alternative theory, multiple trace theory,
suggests that the MTL is always necessary for the retrieval of
episodic memories, including both recent and remote autobio-
graphical memories. According to multiple trace theory, older
memories become more resistant to disruption because each
time a memory is retrieved, a new memory trace is established
that is linked to the older memory traces. Thus, older mem-
ories are represented by more and stronger traces and are more
resistant to partial lesions of the MTL whereas newer memories
are represented by fewer and weaker traces and are more vul-
nerable to damage. By this view, the extent of retrograde amne-
sia is determined by the amount of MTL damage as well as the
number of traces by which a memory is represented. As such,
very extensive MTL damage can affect autobiographical mem-
ories for all time periods.
Yet another pattern of remote memory loss is seen in
patients whose lesions extend beyond the MTL into the ante-
rolateral temporal lobes. In these cases, retrograde amnesia is
characterized not only by extensive autobiographical memory
loss, but also by severe semantic memory loss. The latter is
thought to reflect damage to the neocortical sites where seman-
tic information is stored. Impairments in semantic memory
are often more prominent after lesions to the left temporal
lobe, likely reflecting the role of the left hemisphere in verbal
lexical-semantic processing.
Patients with frontal lobe damage can also present with
extensive retrograde amnesia that may encompass both semantic
and episodic memory. These retrograde memory impairments
are particularly evident under high demands of self-initiated
retrieval and likely reflect deficits in frontal executive processing.
Confabulatory responses may also be prominent with frontal
lobe damage, reflecting impaired monitoring of information
retrieved from memory.
Theoretical and Clinical Implications
While memory deficits are the hallmark of amnesia, this article
has described the important point that not all forms of mem-
ory are similarly impaired in amnesia. Memory impairments
can include both retrograde and anterograde components
but differ according to episodic and semantic demands and
depend on lesion location and extent. Though structurally and
functionally heterogeneous, amnesia provides an important
window onto the multiple forms of memory supported by
the human brain. The patterns of impaired and preserved
function in amnesia clearly demonstrate that rather than
being a unitary system, long-term memory comprises function-
ally and structurally distinct memory subsystems.
In addition to expanding our knowledge about long-term
memory systems, studies of patients with amnesia have also
Amnesia and the Brain 1 29
influenced our understanding of cognitive functions outside
the memory domain. Emerging evidence indicates that amne-
sic patients have difficulty envisioning the future and are
impaired at generating imaginary scenes. These results suggest
that the brain regions involved in remembering the past over-
lap with those involved in thinking about novel future scenar-
ios. It has been proposed that richly envisioning future
scenarios requires flexible retrieval and recombination of
details from the past, drawing on the same memory processes
that are impaired in amnesia. The integrity of long-term mem-
ory mechanisms may also impact other related functions in
amnesia, such as planning and problem solving.
While the theoretical advances provided by the study of
amnesic patients are significant, caution must be taken when
inferring strict one-to-one mappings of neural structure to
cognitive function. In particular, the behavioral measurements
indexed by standard neuropsychological tests may reflect con-
tributions from multiple underlying cognitive processes or
neural systems. Indeed, recent evidence raises questions
about the independence between memory systems that has
been inferred from the performance of amnesic patients on
standard neuropsychological tests (such as implicit memory vs.
explicit memory and short-term memory vs. long-term mem-
ory). For example, while standard neuropsychological mea-
sures suggest that memory impairments in amnesia are
selective to long-term memory, a growing body of evidence
suggests that short-term memory may also be impaired. In
particular, recent behavioral studies of amnesic patients with
damage to the MTL have demonstrated impairments in retain-
ing relational information over delays as brief as one second.
Together with recent functional neuroimaging data in humans
that demonstrates delay-period activity in the MTL during
short-term memory tasks, these results challenge traditional
neuroanatomical dissociations between short-term memory
and long-term memory and suggest that the mnemonic role
of the MTL may extend beyond the long-term memory
domain.
Though the nature of memory impairments in amnesia
must be interpreted carefully, research into this topic will con-
tinue to enable greater clarification of how memory represen-
tations are formed in the human brain and how they contribute
to behavior. In turn, such insights may serve as a foundation for
improved diagnosis and more targeted treatment of memory
disorders. In addition, improved understanding of how the
brain supports memory can inform new rehabilitation techni-
ques. Because pharmacological treatments for organic amnesia
are not widely tested or available, behavioral techniques cur-
rently offer the most promising method for helping amnesic
patients manage and remediate their memory problems. Reha-
bilitative techniques that take advantage of learning mechan-
isms that are preserved in amnesia, such as repetitive training
and implicit learning, can be effective in teaching amnesic
patients new information. With such techniques, patients can
be trained to use external memory aids such as personal digital
assistants (PDAs) or paging devices. Such memory aids can
provide useful reminders about important events, such as the
need to take medications, providing functional support in
everyday life.
See also: Amnesia and the Brain; Episodic Memory; Memory;
Semantic Memory.
Further Reading
Corkin S (2002) What’s new with the amnesic patient H.M.? Nature Reviews
Neuroscience 3: 153-160.
Eichenbaum H, Yonelinas AR, and Ranganath C (2007) The medial temporal lobe and
recognition memory. Annual Review of Neuroscience 30: 123-152.
Giovanello KS and Verfaellie M (2001) The relationship between recall and recognition
in amnesia: Effects of matching recognition between amnesic patients and controls.
Neuropsychology 15: 444-451.
Kensinger EA and Giovanello KS (2005) The status of semantic and episodic memory
in amnesia. In: Chen FJ (ed.) Progress in Neuropsychology Research: Brain
Mapping and Language, 1st edn., pp. 1-14. Hauppauge, NY: Nova Science.
Lim C and Alexander MP (2009) Stroke and episodic memory disorders.
Neuropsychologia 47: 3045-3058.
Moscovitch M, Nadel L, Winocur G, Gilboa A, and Rosenbaum RS (2006) The cognitive
neuroscience of remote episodic, semantic and spatial memory. Current Opinion
in Neurobiology 1 6: 1 79-1 90.
Nadel L and Moscovitch M (1997) Memory consolidation, retrograde amnesia and
the hippocampal complex. Current Opinion in Neurobiology 7: 217-227.
Reed JM and Sguire LR (1998) Retrograde amnesia for facts and events: Findings from
four new cases. Journal of Neuroscience 18: 3943-3954.
Rosenbaum RS, Winocur G, and Moscovitch M (2001) New views on old memories:
Re-evaluating the role of the hippocampal complex. Behavioral Brain Research
127: 183-197.
Schmolck H, Kensinger EA, Corkin S, and Squire LR (2002) Semantic knowledge in
patient H.M. and other patients with bilateral and medial and lateral temporal
lobe lesions. Hippocampus 12: 520-533.
Scoville WB and Milner B (1957) Loss of recent memory after bilateral hippocampal
lesions. Journal of Neurology, Neurosurgery & Psychiatry 20: 11-21.
Squire LR and Zola SM (1998) Episodic memory, semantic memory, and amnesia.
Hippocampus 8: 205-211 .
Tulving E and Markowitsch HJ (1998) Episodic and declarative memory: Role of
the hippocampus. Hippocampus 8: 198-204.
Analogical Reasoning
D Gentner and L Smith, Northwestern University, Evanston, IL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Analog Situation or domain involved in analogical
mapping; either the base (source) or target.
Base (or source) Analog from which inferences and
explanatory structure are drawn; typically, the more
familiar or concrete domain: for example, in the analogy
'An electric circuit is like a plumbing system,' the base
is a plumbing system.
Candidate inference Fact posited about the target analog
based on completing the common relational structure
between base and target: for example, in the above analogy,
a possible inference is 'Higher voltage leads to greater current
in an electric circuit, just as higher pressure leads to greater
water flow in a plumbing system.'
Conceptual metaphor System of words and phrases used
conventionally to talk about one domain by analogy with
another: for example, 'Love is a journey: The road is
sometimes steep; You have to take the rough with the
smooth, etc.'
Literal similarity Likeness based on overall similarity;
applies when two situations are similar in their objects and
entities and also in their relational structure: for example,
one dishwasher tends to be literally similar to another
(alike in both appearance and causal structure).
Relational similarity Likeness based on relations
common to both domains or situations (whether or not the
objects in the two systems resemble each other):
for example, an electrical circuit and a plumbing system
can be relationally similar if they have a common
causal structure.
Source (see base ) structural alignment Identifying
correspondences between two analogs, based on their
common relational structure.
Structural consistency The property of having a
clear set of matches between the two analogs. In a
structurally consistent alignment, the parts of the two
analogs are in one-to-one correspondence, and the analogy's
inferences are clear.
Structural similarity (see Relational Similarity )
Surface similarity Likeness based only on similar objects
and background context between two domains/situations,
without a common relational structure: for example, a
dishwasher may look like a clothes dryer, but their
mechanical and causal relational structures are quite
different.
Systematicity Preference for matching deep
systems of connected relations, rather than smaller
relational sets.
Target Analog one is drawing inferences about; typically
the less familiar or more abstract domain: for example,
in the analogy 'An electric circuit is like a plumbing system,'
the target is electric circuit.
Introduction
Analogical reasoning - the ability to perceive and use relational
similarity between two situations or events - is a fundamental
aspect of human cognition. Indeed, some researchers suggest
that it is the crucial cognitive mechanism that most distin-
guishes human cognition from that of other intelligent species.
It is a core process in scientific discovery and problem-solving,
as well as in categorization and decision-making.
Reasoning by analogy involves identifying a common rela-
tional system between two situations and generating further
inferences driven by these commonalities. The commonalities
may also include concrete property matches between the situa-
tions, but this is not necessary for analogy; what is necessary is
overlap in relational structure. Although this may sound like
a highly complex process, people routinely use analogy in every-
day life. For example, people readily apply proverbs to situa-
tions based on purely relational matches. When you hear 'That's
locking the barn door after the horse has gone,' you don't look
for a barn; rather, you apply the relational pattern - a precaution
taken after the damage is done - to some current situation. This
kind of relational mapping is the essence of analogy.
In the most typical case of analogy, a familiar domain (the
base or source ) serves as a model by which one can comprehend
and draw new inferences about a less familiar domain (the
target). Consider a rather timely example, used by Sterman and
his colleagues at MIT to explain the behavior of atmospheric
carbon dioxide; they describe the balance of carbon dioxide
(C0 2 ) in the atmosphere by analogy with a bathtub:
The amount of water in a bathtub is determined by the rates of water
flowing into the tub and water flowing out through the drain. As
long as the inflow of water into the tub exceeds the outflow, the
bathtub will continue to fill.
Likewise, the amount of carbon dioxide (CO„) in the atmo-
sphere is determined by the rates of CO2 emissions and CO2
removal.
In this analogy, the bathtub corresponds to the atmo-
sphere. Water inflow corresponds to C0 2 emissions into the
atmosphere and water outflow to C0 2 removal. This analogy
invites the (correct) inference that as long as C0 2 emissions
exceed removal, C0 2 levels in the atmosphere will continue
to rise. This illustrates a typical feature of analogy: A process
that cannot be seen becomes easier to grasp by virtue of
an analogy with a familiar situation. This example also reveals
the power of analogy to highlight a common relational pat-
tern. Often, such a common pattern is given a linguistic
label - in this case, a stock-and-flow system - to facilitate
130
Analogical Reasoning 131
remembering the abstraction and applying it to other situa-
tions. A student who grasps this analogy will find that this
abstraction is useful in reasoning about other arenas, such as
cash flow. Analogy is often the most effective way for people
to learn a new relational abstraction; this makes it highly
valuable in education.
Analogy is also frequently used in argumentation, where it
allows the speaker to guide his or her audience toward a
particular framing and set of inferences. For example, when
a US district court judge in December 2002 ordered Microsoft
to include Sun Microsystems' version of Java with the Win-
dows operating system, a lively web discussion ensued, often
involving analogy. One writer wrote: 'Please explain to me
why Microsoft should be forced to include third party soft-
ware in their OS? Every time I buy a six pack of coke, should
a can of Pepsi be included?' and another wrote 'That would
be like (my attorney) being forced to refer clients to his
competition, since they didn't have as much business as
him.' Many of these analogies were picked up and either
extended or reversed. Defending the decision, another writer
wrote 'If Ford had a monopoly on cars, they certainly would
not be allowed to sell their cars with only Ford brand radios
and tires
As demonstrated in the above examples, analogies vary
widely in their appearance, content, and usage. But they can
all be characterized by a set of processes common to analogical
reasoning of all types. These processes are:
• Retrieval : Given some current topic in working memory, a
person may be reminded of a prior analogous situation in
long-term memory.
• Mapping: Given two cases present in working memory
(either through analogical retrieval or simply through
encountering two cases together), mapping involves a pro-
cess of aligning the representations and projecting infer-
ences from one analog to the other.
• Evaluation: Once an analogical mapping has been done, the
analogy and its inferences are judged.
We begin with mapping, the core process in analogical
reasoning, and its subprocesses, reserving retrieval for later.
The rationale for this is that, while analogical reasoning invari-
ably involves a mapping process, it does not always require
finding a second analog in memory. For example, when ana-
logies are used in argumentation, both analogs are typically
presented to the reasoner, who must then carry out a mapping
and evaluate the analogy.
Mapping
Mapping is the core process of analogy, and has therefore been
the main focus of analogy research. At a first level, the mapping
process consists of finding how two situations are similar, and
then bringing across further inferences from the better-known
situation (the base, or source) to the less familiar one (the target).
What distinguishes analogy from other kinds of similarity
is that for two situations to be analogical, they must be similar
in their relational structure. Analogy research has largely agreed
on a set of principles laid out by Dedre Gentner in 1983, in
a theory called structure mapping. According to structure-
mapping theory, analogical mapping requires aligning the two
situations based on their commonalities - particularly their
common relational structure - and projecting inferences from
the base to the target, according to this alignment.
Alignment
In an analogy, the two situations being compared can be
aligned on the basis of common relational structure. For exam-
ple, consider the simple perceptual analogy in Figure 1. These
two scenes are analogous: they can be aligned - that is, the
elements can be placed in one-to-one correspondence such
that the same relational structure (Circle 1 > Circle 2 > Circle
3, which could be labeled steadily decreasing size) holds in both
scenes. There is something satisfying about noticing structural
alignments like this one. For example, in this case, despite the
obvious object similarity between the two circles indicated by
black arrows (both circles are the same size), when people are
led to compare these two scenes (i.e., to engage in analogical
mapping), they will match objects that occupy the same rela-
tional role in their respective figures: for example, the smallest,
rightmost circles.
This same kind of alignment process occurs with a complex
analogy like the bathtub example. Here too, people will align
two domains based on their common relations. However, for
scientific analogies, the matching relational structure will gener-
ally be governed by causal relations rather than spatial relations.
Even though most of us are not aware of how we process
analogies, research suggests that there are some implicit prin-
ciples that people follow during analogical mapping. First, we
like our alignments to be structurally consistent. Two things are
required for an alignment to be structurally consistent. First,
each object in the base should match to one and only one
thing in the target; this is known as one-to-one correspondence.
Object Relational
match match
1 1
OOo
.OOo
Figure 1 Perceptual analogy depicting decreasing size. The smallest circle in (a) could match either with the middle circle in (b) (an object match)
or with the rightmost circle (the relational match). Adapted from Markman AB and Gentner D (1993) Structural alignment during similarity
comparisons. Cognitive Psychology 25: 431-467.
132 Analogical Reasoning
For example, for the pair in Figure 1, people either map the
smallest circle in 1A with the smallest circle in IB (the rela-
tional match) or else with the middle circle (the object match)
but not both. Second, if two relations are matched with one
another, then their arguments must also be matched. We can
see the principle of structural consistency at work in scientific
analogies just as in simple spatial analogy above. For example,
one-to-one correspondence holds in the bathtub analogy:
water draining from the tub cannot correspond to both C0 2
emissions and C0 2 removal.
Further, when we align two situations, we do not simply
find one pair of matching relations and stop there: rather, we
prefer to match large, deep-connected systems. This preference
is known as the systematicity principle : people prefer to align two
domains based on large, connected relational systems, rather
than just a single common relation. In the bathtub analogy,
people generally prefer to align the entire stock-and-flow sys-
tem that characterizes water flow and C0 2 flow, rather than
simply noting that both involve the single relation of one thing
flowing into another. Our desire for systematicity reflects an
implicit preference for analogies that are highly informative
and have inferential power.
Figure 2(a) provides a schematic depiction of structural
alignment. Notice that this depiction shows a one-to-one
correspondence between elements of the two domains - each
element is mapped to (at most) one element in the other domain.
Also, not only are relations matched but their corresponding
arguments are matched as well. Finally, a large, inferentially
rich relational pattern is matched, illustrating systematicity.
Inference
Analogies permit us to draw new inferences about the target.
Indeed, one major reason we use analogy is to learn something
new about the target domain by recruiting our knowledge of
a relationally similar base domain. But this brings up a key
question: How do we avoid drawing the many wrong (or even
ridiculous) inferences that we might make if we simply
mapped across whatever we know about the base to the target?
Clearly, analogical reasoning would be useless if we had to
spend time rejecting inferences such as pouring Mr. Bubble into
the atmosphere can make for an enjoyable evening, which could be
derived from the bathtub analogy. One key finding in analogy
research is that people are highly selective in the inferences
they make from analogies - we do not simply bring over
everything we know about the base to the target.
According to the structure-mapping view, inference hap-
pens as a natural outcome of the structural alignment process.
Once the base and target have been aligned and their common
relational structure found, if there are additional parts of the
relational pattern in the base that are not present in the target,
then this missing pattern will be brought over as a candidate
inference (Figure 2(b)). Thus, one way to think about infer-
ence generation is as a process of relational pattern completion.
The requirement that candidate inferences be connected to the
common relational pattern effectively filters which inferences
will be considered. For example, in the bathtub analogy, pour-
ing in some Mr. Bubble makes for an enjoyable evening is not
connected to the common relational structure (as amplified
below), so this inference would not normally be made.
(a)
Figure 2 Analogy as structure mapping, (a) Initial alignment of
common relational structure. Relations are matched between domains,
and their arguments are also matched, (b) A frequent outcome of making
an analogy is that candidate inferences are generated by completing the
missing relational pattern in the target, (c) A possible outcome of
structural alignment is abstraction of the common relational pattern.
Because inference and structural alignment are so tightly
linked, perhaps it is not surprising that many of the constraints
people impose on alignment they also use in inference. People
prefer inferences that are consistent with the rest of the match-
ing structure between the base and target. In addition, people
prefer systematicity in inference: that is, people are more likely
to project inferences that are connected to large relational
patterns, rather than to project isolated parts of the base or
inferences involving only a subset of the matching relational
pattern. Clement and Gentner in a 1991 study found support
for systematicity in inference: people were more likely to
import a fact from the base to the target when it was connected
to other facts shared with the target. In analogical matching
and inference, people are not interested in isolated coinci-
dental matches; rather, there is a tacit preference for deeply
interconnected relational patterns.
Here is how you might generate an inference in the bathtub
analogy. First, you align the known fact that the amount of water
entering and leaving the tub determines the total amount of water in
the tub with the known fact that the amount ofC0 2 entering and
leaving the atmosphere determines the total amount of C0 2 in the
Analogical Reasoning 1 33
atmosphere. You can then draw new inferences. For example, let
us assume you know that the amount of water in the tub will
decrease only if the amount of water draining is greater than
the amount of water entering. You can carry this fact over to the
target as an analogical inference: 'The amount of C0 2 in the
atmosphere will decrease only if C0 2 removal exceeds C0 2
emissions.' This inference is warranted by its connection to the
aligned relational structure.
As just discussed, one way in which analogy is useful is that it
helps us learn new information about the target by suggesting
inferences. Another benefit of analogy is abstraction : that is, we
may derive a more general understanding based on abstracting
the common relational pattern (Figure 2(c)). For example, on
the basis of the bathtub analogy, a student might extract an
abstract schema of a stock-and-flow system: as long as inflow
exceeds outflow, the stock will increase; and as long as outflow
exceeds inflow the stock will decrease. Once formed, this
abstraction can serve as a general schema for other stock-and-
flow systems, such as those that occur in economics or biology.
In addition to highlighting potential abstractions, analogies
can also call attention to certain differences between the ana-
logs. For example, a salient difference between the C0 2 system
and the bathtub system is that the amount of water in a
bathtub can change rather quickly by adjusting the inflow or
outflow, whereas if you adjust C0 2 emissions and removal, it
takes several decades to see a corresponding change in C0 2
levels in the atmosphere.
Evaluation
Once the common alignment and the inferences have been
generated, the analogy and its inferences are evaluated.
The criteria for evaluation can be grouped into three classes.
The first factor is the factual correctness of the inferences gener-
ated by the analogy. If the analogy yields inferences that are
untme, the inferences and the analogy will in general be
rejected, or at least revised. Of course, in some cases one cannot
immediately identify whether an inference is true or not, as
when making predictions about a future event, or when pre-
dicting a scientific outcome by analogy with another domain.
A related factor in evaluating inferences is adaptability, how
easy it is to modify a fact from the base to fit the target. People
accept inferences that are highly adaptable to the target more
readily than those that are less adaptable.
A second factor that people use in the evaluation of inferences
is goal relevance. Goal relevance has been explored as a major
factor in analogical reasoning in the theories of Keith Holyoak
and colleagues. They emphasize that inferences that are relevant
to the current goals of the reasoner are more likely to be projected
during analogical inference, and are more important in evaluat-
ing the analogy. This constraint is particularly germane in prob-
lem-solving situations. During problem-solving, even if an
analogy yields a reasonable inference, it is unlikely to be retained
if it does not bear on the problem at hand. Spellman and
Holyoak showed in a 1996 study that when two possible map-
pings are available for a given analogy, people will select the
mapping whose inferences are relevant to their goals.
A third factor that may influence evaluation is how much
new knowledge the analogy and its inferences can potentially
provide. The idea is that inferences that potentially yield a
significant gain in new knowledge may be desirable (even if
somewhat risky), especially when brainstorming or dealing
with unfamiliar domains. The evaluation of inferences and
of the whole analogy can mutually influence one another.
Evaluation of particular inferences contributes to the larger
evaluation of the analogy. If an analogy generates false infer-
ences, we will generally reject or at least revise the analogy.
Analogical Retrieval
So far, we have been discussing a scenario in which two ana-
logs are already present in working memory. However, some-
times only one analog is currently present, and we experience
a reminding to something that may be similar or analogous
(e.g., a previously solved problem). Thus, understanding the
use of analogy in reasoning requires some account of how
potential analogs are accessed in long-term memory - what
leads people to think of analogies? While relational similarity
exerts a strong influence on analogical mapping, it has much
weaker effects on retrieval from memory. People often fail to
retrieve potentially useful analogs, even if they share relational
structure. In a classic 1980 study by Mary Gick and Keith
Holyoak, participants were given a very difficult insight prob-
lem. When people simply read the problem and tried to solve
it, only 10% succeeded. When another group was given a story
with an analogous solution (but with different specific con-
tent) before receiving the insight problem, about 30% solved
it - three times as many as without the analogy. Yet despite this
impressive gain, the majority still typically failed to solve the
problem. Although one might have suspected that they had
simply forgotten or never encoded the analogous story, this
was not the case. When people were given a hint to try using
the prior story, the proportion solving the problem rose to
around 90%. Thus, the solution rate tripled if people had
heard an analogous story; and it tripled again if people were
reminded of the prior analog. People's failure to access the
prior analogous problem resulted not from forgetting it but
from failure to be reminded of it by the current problem. This
is an example of what Alfred North Whitehead called 'inert
knowledge' - knowledge that is not accessed when needed.
Remindings to potential analogs are typically driven by
surface similarities, such as similar objects and contexts, rather
than by similarities in relational structure. Of course, it is
important to bear in mind that in all these studies, some
individuals show genuine relational transfer. It is not the case
that relational remindings never occur; it's just that they are
much rarer than surface remindings and overall similarity
remindings. Gentner and colleagues have proposed two inter-
related explanations for this. First, people often encode experi-
ences in a content-specific manner, so that later remindings
occur only for highly surface-similar experiences. Second, there
is evidence that people's representations of relations are more
context-specific than those of objects and entities. For example,
Brian Ross gave people mathematical word problems to study,
and later gave them new word problems. Most of their later
remindings were to examples that were similar on the surface
(e.g., both problems talked about mechanics), irrespective of
whether the underlying mathematical principles matched.
1 34 Analogical Reasoning
Experts in a domain are more likely than novices to retrieve
relationally similar examples, but even experts retrieve some
examples that are similar only on the surface. However, as
demonstrated by Laura Novick in 1988, experts reject mislead-
ing surface remindings more quickly than do novices. Thus,
especially for novices, there is an unfortunate dissociation:
while accuracy of transfer depends critically on the degree of
relational pattern matching, memory retrieval depends largely
on surface similarity between domains.
Factors that Influence Mapping
People's fluency in carrying out analogical mappings is influ-
enced by three broad kinds of factors. First are factors internal
to the analogical mapping itself, such as systematicity -
whether the common relational system possesses a deeply
connected structure - and transparency - the degree to which
corresponding elements are similar. The second category
includes characteristics of the reasoner, such as age and exper-
tise. The third includes task factors such as processing load,
time pressure, and context.
We have already discussed the importance of structural
consistency and systematicity in analogical mapping, so we
turn directly to transparency. Transparency depends chiefly on
the degree of similarity between corresponding objects. A high-
transparency analogy is one in which the objects that play the
same roles in the common relational structure are highly simi-
lar (or identical), and the objects that play different roles are
quite dissimilar. Such an analogy is generally both obvious
and easy to align correctly. The most pronounced case of
high transparency is literal similarity, in which both relations
and objects match. As discussed above, literal-similarity
matches are more reliably retrieved from memory than are
purely relational analogies. To this, we can add that even in
online processing, literal-similarity matches are processed faster
than purely relational matches (in which the corresponding
objects lack similarity). This fits with our intuitions: for exam-
ple, it is easier to see how one tiger is like another than to see
how a tiger is like an eagle (both are carnivores that hunt alone) .
However, although high-transparency matches are natural and
easy to process, many useful explanatory analogies are of rela-
tively low transparency - that is, the corresponding objects are
not at all similar. The bathtub analogy discussed earlier is an
example of a low-transparency analogy: the corresponding
objects are very dissimilar, for example, water looks very differ-
ent from C0 2 (and in any case, we never directly observe C0 2
molecules). Finally, as noted above, achieving a relational
alignment is more difficult when noncorresponding objects
are similar. Thus, the worst case of low transparency is cross-
mapped analogy, in which similar objects play different roles
within the relational structure, as exemplified below.
Both transparency and systematicity interact with individ-
ual characteristics of the reasoner, notably age and experience.
For example, Gentner and Toupin gave children a simple story
and asked them to reenact the story with new characters. Both
6- and 9-year-olds performed best when the corresponding
characters were highly similar between the two stories (the
literal-similarity condition - high transparency). They per-
formed less well when corresponding characters were different
(medium transparency), and they performed worst when simi-
lar characters played different roles across the two stories (the
cross-mapped condition - low transparency). Thus, both age
groups were sensitive to the transparency of the correspon-
dences. In addition, older children (but not younger children)
benefited strongly from systematicity: when they were given a
summary statement that provided the structure for the plot,
their performance stayed high regardless of transparency. They
were able to use the relational system provided by the plot
description to maintain relational correspondences despite the
tempting object matches.
Further studies have corroborated this finding that when
relational similarity is pitted against object similarity as in the
cross-mapped conditions mentioned above, younger children
are highly influenced by object matches and less able to attend
to relational matches than are older participants. This shift
from a focus on objects to a focus on relations has been termed
the relational shift. Although there is widespread agreement
that such a shift occurs, developmental researchers differ on
why. Gentner and colleagues have argued that the relational
shift is driven primarily by gains in relational knowledge. An
alternative view, proposed by Graeme Halford and colleagues,
is that the shift results from a developmental increase in pro-
cessing capacity; according to this view, processing relational
matches requires more processing capacity than processing
simple object matches. A third view, championed by Lindsey
Richland and colleagues, explains the relational shift as stem-
ming from maturational increases in inhibitory control, which
permits the child to suppress object matches in favor of rela-
tional matches.
The third class of factors affecting analogical processing
concerns task variables such as time pressure, processing
load, and immediate context. One generalization that emerges
from several studies is that making relational matches requires
more time and processing resources than making object attri-
bute matches. For example, a study by Robert Goldstone and
Doug Medin found that when people are forced to answer
quickly, they are strongly influenced by object matches (such
as a black wing with a black wing), even in cases where they
would choose a relational match (such as both wings same
color with each analog) if given sufficient time.
Adult performance in mapping tasks is also influenced by
immediately preceding experiences. For example, in the one-
shot mapping task of Markman and Gentner, subjects are
shown a pair of cross-mapped pictures, such as a truck towing
a car and a car towing a boat (Figure 3). The experimenter
points to the car in the first picture, and the subject indicates
which object in the second picture 'goes with' it. Subjects often
choose the object match (e.g., the other car). However, if they
have previously rated the similarity of the pair, they are likely
to choose the relational match (the boat). These findings sug-
gest that carrying out a similarity comparison encourages
structural alignment.
This one-shot mapping task has also been used to test
whether processing load influences analogical processing. The
experimenter pointed to the cross-mapped object in the first
picture (the car), and subjects were instructed to point to the
relational correspondence (the boat) in the second picture.
Subjects made more object-mapping errors when given an
extra processing load, such as having to count backward.
Analogical Reasoning 1 35
Figure 3 Sample cross-mapped pictures from the Markman and
Gentner study. Carrying out a prior comparison increases the likelihood
that adults will match the car in (a) with the boat in (b) based on their
common relational role (both are the things being towed), rather than
simply matching the two cars.
Neuropsychology of Analogical Reasoning
Recent studies have begun investigating the neural correlates
of analogical processing. The studies so far converge on areas
within the left prefrontal cortex as important in analogical
reasoning. Of course, much remains to be discovered. We do
not yet know the degree to which these prefrontal areas are
specialized for analogy as compared to other higher order
reasoning processes. We also do not yet know whether and to
what degree other areas of the brain are involved in analogy,
and whether this differs according to the type of analogy.
Further studies exploring a greater range of analogical tasks
and materials should give us a more complete picture.
Analogy in Naturalistic Settings
Analogy researchers note that analogy is a ubiquitous com-
ponent of human thinking. However, much of the research
focuses on analogy use in one very specific setting: the experi-
mental psychology laboratory. Many studies of analogical
reasoning in the laboratory have adopted an approach of
explicitly providing participants with an analogy and eliciting
particular responses from them (e.g., inference preference rat-
ings, identifying correspondences). This method enables psy-
chologists to closely observe phenomena tied to analogical
mapping. However, this state of affairs suggests two questions:
(1) To what degree do people spontaneously use analogies in
real-world contexts? and (2) Is this analogical reasoning
guided by the same preferences (e.g., relational priority, sys-
tematicity) identified in the laboratory? For this discussion, we
move out of the laboratory to review work on analogy in real-
world settings, or analogy 'in the wild.'
Dunbar and his colleagues have investigated the use of
analogy in naturalistic environments in a variety of contexts.
In one project, he studied the day-to-day processes of scientists
in microbiology laboratories. Dunbar found that analogical
thinking was a key component of all aspects of scientific
reasoning, ranging from hypothesis generation to experimental
design, data interpretation and explanation. Analogy is also a
key component of the way scientists reason about unexpected
findings. Interestingly, Dunbar observed that many of the ana-
logies scientists made were of high overall similarity, sharing
not only causal structure but also many superficial features. For
example, a scientist working on a novel type of bacterium might
hypothesize that its genetic sequence is like that of a highly
similar species. These studies suggest that people frequently
make analogies that reveal deep relational similarities, but
that superficial similarity between domains aids in the noticing
of these analogies, as many theories of analogy would predict.
However, scientists do sometimes use the kind of dramatic
analogy that constitutes a true leap in understanding. For
example, Robert Boyle likened molecular motion to leaves
being swirled around by the wind; and Johannes Kepler lik-
ened the course of the planets around the sun to the course of a
boat steering in a current (the gravitational attraction). Histor-
ical analyses have documented many cases like these, in which
scientists have used analogies based on relational patterns that
share little or no superficial similarity. These types of analogies
often occur in connection with larger discoveries and shifts of
paradigms, such as Rutherford's analogy likening the structure
of an atom to that of the solar system, which displaced the
then-dominant plum-pudding model.
One interesting case of everyday analogy is the use of
conceptual metaphors, as discussed by George Lakoff, Mark
lohnson, and colleagues. For example, statements like 'Their
marriage is going over some serious bumps in the road' or
'They're just coasting along without putting in energy' would
be part of the 'love is a journey' conceptual metaphor. Often,
the same base term can be conventionally used with many
different targets; for example, we can liken progress in a career,
in college, or even in a specific project to a journey, as in 'she's
zipping along on her paper, way ahead of the other students.'
Many of these metaphoric systems behave like extended ana-
logies, as illustrated by the journey example. Another fre-
quently used example is the conventional mapping from
space to time, as in 'The holidays will soon be here' and
'Exams will come after the holidays this year.' We also find
extended analogical metaphors in the introduction of new
technical concepts, such as (computer) virus, and accompanying
ideas such as antibodies against such viruses.
Another everyday use of analogy is in humor. For example,
Benjamin Franklin stated, "A countryman between two lawyers
is like a fish between two cats. " A more elaborate example comes
from Louis Menand, writing in the New Yorker of February 23,
2009: "Postmodernism is the Swiss Army knife of critical con-
cepts. It's definitely overloaded, and it can do almost any job you
need done." Research by Jeff Loewenstein and Chip Heath in
136 Analogical Reasoning
2009 has shown that many jokes (as well as many children's
stories and advertisements) follow a three-step repetition-break
plot structure : two closely similar stories are given, followed by the
break - a sudden change from the parallel plot structure. This
kind of structure will be familiar to anyone who has heard a
'three guys walk into a bar' joke; the first two, closely similar,
stories set up an aligned structure, and the humor comes from
the surprise when the third story 'breaks' that structure.
Analogical Reasoning Without Awareness
Analogical reasoning has typically been considered a high-level
reasoning process; for this reason, analogy has traditionally
been thought of as a deliberate, conscious activity. Much of
the research on analogy accords with this assumption: experi-
mental work tends to focus on the deliberate use of analogy,
where learners either discover or are given an analogy, use it to
derive new inferences, and accept only those inferences that
they consider structurally sound, plausible, and goal-relevant.
However, research in the past decade has demonstrated that not
all analogical reasoning is deliberate. Isabelle Blanchette and
Kevin Dunbar first demonstrated the analogical insertion effect,
in which analogical inferences are integrated unknowingly into
mental representations of the target domain. In their studies,
participants read descriptions of a target issue (e.g., legalizing
marijuana) and then were given an analogy to another situation
(e.g., ending Prohibition). On a subsequent recognition test,
these participants often misidentified analogical inferences as
facts actually presented about the target: that is, they mistakenly
'recognized' assertions about marijuana repeal as part of the
passage, when what they had actually read was the analogous
assertions about alcohol and Prohibition.
The studies just described demonstrate that analogical infer-
ences can be drawn without the reasoner's full awareness. How-
ever, in both of these studies, the participants were explicitly
told about the analogy between the two domains. This invites
the question of whether analogical inference can occur without
explicit awareness of the analogy. Recent research by Samuel
Day and Dedre Gentner shows that the answer is yes. In their
study, people read a series of brief passages and then answered
questions about one of the later passages. Unbeknownst to the
participants, the later passage was analogous to a prior passage,
of which there were two versions. The results revealed that
participants spontaneously made analogical inferences from
whichever version of the early passage they had received, with-
out recognition of having done so. These results show that
information from a single analogous instance can influence
the way in which another situation is understood or remem-
bered without an individual's awareness. This suggests the
intriguing possibility that analogy may be a process by which
people implicitly understand and structure everyday experi-
ences, and form abstract schemas over similar experiences.
Conclusion
Analogy is at the core of higher-order cognition. As Douglas
Hofstadter puts it, "Analogy is the engine of cognition.'' Analog-
ical thinking enters into creative discovery, problem-solving,
categorization, and learning and transfer. This realization has
led to a surge in research activity over the last three decades,
resulting in great gains in our understanding of analogical pro-
cessing. But there remain many open questions. We need a
better understanding of how analogy operates in everyday
learning and reasoning. What determines when people will
spontaneously compare things, and how much they will profit
from the comparison? Are spontaneous, nonaware analogies
common, and if so what are their effects? When and how do
people filter out bad analogies? Another area that is being
actively explored is the neural underpinnings of analogical pro-
cessing, as discussed above. There has also been recent interest
in further specifying the development of analogical ability in
children, and also in understanding the role of analogy in
children's everyday learning across a variety of domains, from
language acquisition to category formation. Finally, we are just
beginning to explore analogical processing in other species.
Cross-species comparisons will help to delineate the cognitive
components of analogical ability.
See also: Creative and Imaginative Thinking; Human Intelligence;
Reasoning.
Further Reading
Blanchette I and Dunbar K (2002) Representational change and analogy: How analogical
inferences alter representations. Journal of Experimental Psychology. Learning,
Memory, and Cognition 28: 672-685.
Day S and Gentner D (2007) Nonintentional analogical inference in text comprehension.
Memory & Cognition 35: 39-49.
Dunbar K (1995) How scientists really reason: Scientific reasoning in real-world
laboratories. In: Sternberg RJ and Davidson J (eds.) Mechanisms of Insight,
pp. 365-395. Cambridge, MA: MIT Press.
French RM (1995) The Subtlety of Sameness: A Theory and Computer Model of
Analogy-Making. Cambridge, MA: MIT Press.
Gentner D (1983) Structure-mapping: A theoretical framework for analogy. Cognitive
Science 7:155-170.
Gentner D, Holyoak KJ, and Kokinov B (eds.) (2001) The Analogical Mind: Perspectives
from Cognitive Science. Cambridge, MA: MIT Press.
Gentner D and Markman AB (1997) Structure-mapping In analogy and similarity.
American Psychologist 52: 45-56.
Gick ML and Holyoak KJ (1980) Analogical problem solving. Cognitive Psychology
12: 306-355.
Hesse MB (1966) Models and Analogies in Science. Notre Dame, IN: University of Notre
Dame Press.
Hofstadter DR (2001) Analogy as the core of cognition. In: Gentner D, Holyoak KJ, and
Kokinov BN (eds.) The Analogical Mind: Perspectives from Cognitive Science,
pp. 499-538. Cambridge, MA: MIT Press.
Holyoak KJ and Thagard P (1 996) Mental Leaps: Analogy in Creative Thought.
Cambridge, MA: MIT Press.
Lakoff G and Johnson M (1980) Metaphors We Live By. Chicago, IL: University of
Chicago Press.
Loewenstein J and Heath C (2009) The repetition-break plot structure: A cognitive
influence on selection in the marketplace of ideas. Cognitive Science 33: 1-19.
Markman AB and Gentner D (1993) Structural alignment during similarity comparisons.
Cognitive Psychology 25: 431-467.
Thagard P (1992) Conceptual Revolutions. Princeton, NJ: Princeton University Press.
Relevant Websites
http://www.gdufs.edu.cn/professor/pro-gui/myhomepage/metaphor.htm - Center for
the Cognitive Science of Metaphor Online.
http://groups.psych.northwestern.edu/gentner/ - Dedre Gentner’s website.
http://www.psych.ucla.edu/faculty/faculty_page?id1/446&area1/43 - Keith Holyoak’s
website.
Anger
J R Averill, University of Massachusetts, Amherst, MA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Aggression The intentional infliction of harm on others,
against their wishes and not for their own good.
Anger An emotional state that involves both an attribution
of blame for some perceived wrong and an impulse to
correct the wrong or prevent its recurrence.
Category mistake The logical fallacy of assuming that what
is true of one category (e.g., aggression) is also true of a
related category (e.g., anger); category mistakes are
particularly common when the relation between the
categories is one of class inclusion, part-whole, or
cause-effect.
Catharsis A purgation ('cleansing') or clarification of
emotion; used to explain the relief sometimes associated
with vicarious emotional experiences, for example, during
theatrical performances (Aristotle) and psychotherapy
(Freud).
Crime of passion Legal terminology for homicide
committed while in an emotional state, typically anger; also
termed voluntary manslaughter.
Emotional state A short-term disposition to respond
based on an evaluative judgment (e.g., that something is
good or bad, challenging or threatening, just or unjust);
as traditionally conceived, emotional states differ from
other short-term dispositions in that the relevant
evaluation is nondeliberate and the ensuing response
seems to be beyond personal control (a passion rather
than an action).
Hate A long-term disposition to respond with hostility.
While anger is typically accompanied by physiological
arousal and precipitous action, hate is primarily a way of
thinking. Hate can foster anger; conversely, unresolved
anger can culminate in hate. But hate can also serve as an
antidote to one's own anger for, under normal conditions,
those whom we hate cannot hurt us.
Social constructionism The view that most emotions, anger
included, are not an intrinsic part of human nature
(i.e., genetically determined) but, rather, are constructed
from elementary processes in conformance with social
norms (beliefs and rules).
Anger and aggression are often discussed under one heading, a
fact that has distorted discussions of both. Anger need not be,
and typically is not, manifested in aggression; conversely,
aggression is frequently manifested for reasons that have little
to do with anger. This article focuses on anger as an emotion;
the orientation is social-psychological. Psychologically, anger
is an accusation of wrongdoing, an attribution of blame, and a
determination to correct the perceived wrong; socially, anger
functions as a kind of informal judiciary that helps regulate
interpersonal relationships.
Generic and Specific Uses of the Concept of Anger
The concept of anger is often used in two distinct ways: first,
as a generic term to cover a wide range of related emotional
responses, such as envy, jealousy, fury, frustration, annoyance,
contempt, and the like; and second, as a specific emotion on a
par with other emotions in the same general category (e.g.,
anger vs. envy). Linguistically, when a single term is used to
refer to phenomena at two different levels of generality, the
term so used is known as a synecdoche. 'Anger' is often used as
a synecdoche, both in everyday speech and psychological
theory.
It is important to distinguish the generic and specific uses of
the concept of anger, for what is tme of a category at one level
of generality need not be true at another (higher or lower)
level. To conflate levels of generality is to commit a 'category
mistake.' Category mistakes are particularly common when
synecdoches are involved, for the use of the same term makes
it easy to traverse between levels of generality without recog-
nizing the shift in meaning.
Consider, first, the meaning of anger as a generic category.
What do members of the category (e.g., envy, jealousy, and
fury, as well as anger as a specific emotion) have in common?
On some occasions, at least, all are associated with aggressive
behavior. In its generic sense, then, anger may be used to refer
to almost any aggressive emotional response. Angry aggression
in this sense is typically contrasted with instrumental aggres-
sion, that is, aggression deliberately used as a means for achiev-
ing some extrinsic reward (as in a robbery).
Now, consider anger as a specific emotion. How does anger
in a narrow sense differ from other members of the general
category of anger-like emotions, such as envy, jealousy, and
fury? As is explained more fully below, anger as a specific
emotion involves an attribution of blame for wrongdoing
and a desire for reparation. Only rarely is anger in this sense
accompanied by aggression, at least by physical aggression. The
main objective of anger is not to harm the instigator, but to
correct the perceived wrong and prevent recurrence.
Category mistakes can proceed from the top-down (generic
to specific categories) or from the bottom-up (specific to
generic categories). An example of a top-down mistake is the
common assumption that because anger (in the generic sense)
connotes aggression, then anger (in the specific sense) must
also involve a tendency toward aggression, even if that ten-
dency is repressed, bottled up, disguised, or otherwise made
unobservable. An example of a bottom-up mistake is the
137
138 Anger
inclination on the part of perpetrators to excuse aggression of
all sorts, from child abuse to urban riots, with the plea: 'It
wasn't my fault; he/she/they made me angry.'
The following discussion focuses on anger as a specific
emotion. No attempt is made to review the vast amount of
research related to aggression in general. Considerable ad-
vances have been made in understanding the biological (evo-
lutionary) and physiological bases of aggressive behavior; as
important as that understanding may be, it is only tangentially
related to anger as a social-psychological phenomenon. On
the other hand, data relevant to anger come from a wide range
of sources, including historical (ethical) teachings, legal pro-
ceedings, surveys of everyday experience, experimental studies,
cross-cultural research, and clinical practice.
Anger in Historical Perspective
Historically, and today, anger appears on nearly every list of
'basic' emotions. Often, this is taken to mean that anger is
somehow prior to society, for example, that anger is a relic of
biological evolution. Anger, however, does not arise from the
depths of some presocial self. It embodies a history of social
meanings. From early infancy, we are emotional apprentices,
first to our parents, and then to teachers, ministers, artists,
entertainers, and other formal agents of cultural transmission.
The apprenticeship continues on a less formal basis through-
out life as we interact with family, friends, and acquaintances
in everyday, face-to-face encounters. In the process, we learn
the beliefs and rules that help constitute our emotions, and the
skills to enact them properly. Subsequently, we may refine and
transform what we have acquired, but we can never completely
escape the vestiges of our cultural heritage.
Western culture has been greatly influenced by the conflu-
ence of two main streams of thought, Greco-Roman philosophy
and Judeo-Christian theology. Each stream has contributed its
share to current conceptions of anger. A little later, we will
consider briefly the views of three early representatives of these
traditions, namely, Aristotle, Seneca, and Lactantius. But first, a
bit of etymology may be helpful.
Anger is an emotion. On that, everyone would agree. But
what is an emotion? On that, few psychologists agree. For most
of Western history, from the time of the ancient Greeks until
about the middle of the eighteenth century, what we now call
emotions were commonly referred to as passions. As tradition-
ally conceived, a passion is anything a person (or physical
object) 'suffers'; this would include, in the case of humans,
both emotions and diseases. Hence, from the same root
(the Greek, pathe, and the Latin, pati, passiones), we get such
emotion-related terms as passion, pathos, and sympathy and
such disease-related terms as patient, pathogen, and pathology.
Aristotle considered that the passions formed one of ten
fundamental categories of being. Actions - things a person
does deliberately and with forethought - formed another dis-
tinct category. This distinction between passions and actions is
embedded in our ordinary language, as when we say of a
person that he was 'gripped,' 'seized,' and 'overcome' by emo-
tion. Anger, in particular, is often depicted as a 'wild beast'
that must be 'tamed,' or as an 'inner force' that will 'explode' if
not adequately 'vented.' Such conventional ways of speaking
help constitute the way we think and feel. But linguistic con-
ventions can mislead as well as inform.
When addressing the practical concerns of rhetoricians, as
opposed to the theoretical concerns of metaphysicians, Aristotle
defined anger "as an impulse, accompanied by pain, to a con-
spicuous revenge for a conspicuous slight directed without
justification towards what concerns oneself or towards what
concerns one's friends" ( Rhetoric , 1378a30). Two aspects of this
definition are especially noteworthy. First, the 'slight' (by which
Aristotle meant any show of contempt, spite, or insolence, espe-
cially by a friend or inferior) had to be 'without justification.' A
reprove that is deserved is not an adequate provocation to anger,
no matter how upsetting it might be. Second, the revenge must
be 'conspicuous'; otherwise the offender will not realize that he
or she is being punished for the slight. To these two points, we
should add a third important feature of Aristotle's analysis,
namely, the response must be appropriate to the provocation
and the situation, being neither too weak nor too strong.
Aristotle's analysis contains a paradox. If anger is a passion,
how can it be so finely tuned to the social context, in both
initiation and expression? One way to resolve this paradox is
to deny that anger, at least under ordinary circumstances, is
beyond personal control: to assert, in effect, that the classifica-
tion of anger as a passion is based more on metaphor than on
fact. More will be said about this possibility later. Another
possible resolution is to regard anger as a primitive reaction
to painful stimuli, an impulse that is regulated but not con-
stituted by social norms. This is the position of many current
psychological theories of anger and aggression. It was also the
position taken by Aristotle, who believed that the physiological
changes that accompany anger (e.g., a 'boiling of blood about
the heart') can impel a person to act in a nondeliberate and
irrational manner.
Aristotle's answer, however, hardly seems adequate. The
physiological changes that accompany a brisk walk, say, exceed
those that accompany most episodes of anger, and yet a brisk
walk does not lead a person to act irrationally. Conversely,
some of the worst deeds committed in the name of anger are
done in a cold and calculated manner, with no more than
usual physiological involvement.
The Roman philosopher and statesman Seneca (4 BC-AD
65) wrote the first full work devoted entirely to anger. As a
follower of Stoicism, Seneca defined passion as a form of false
judgment, the mental analogue of a physical disease. (After all,
who would deliberately, 'in his right mind,' make a false judg-
ment?) This definition of passion led Seneca to reject the
notion that anger is necessarily tied to one's bodily condition
or that it could be anything but a human emotion. A false
judgment, no less than a true judgment, is born of reason, no
matter how ill-begotten the false judgment might be. Seneca
also rejected the Aristotelian notion that anger, properly
directed and in moderation, can be considered good. If on
occasion a false judgment produces a beneficial outcome, it is
a fortunate happenstance. Whatever good is done in anger,
according to Seneca, can be done even better and with greater
surety, following rational deliberation. In a manner prescient
of current cognitive therapies, Seneca offered advice on how to
reinterpret events so that they are no longer judged offensive,
or, when that is not possible because the offense is too egre-
gious, to correct the wrong in a rational and deliberate manner.
Anger 1 39
Stoicism was the dominant philosophy of the Roman
Empire, and views such as those expressed by Seneca were
very influential - in theory, if not in practice. And although it
might at first seem like a relatively minor footnote to the
history of thought, Seneca's treatment of anger also posed a
major challenge to the ascendancy of Christianity. Jehovah of
the Old Testament is often depicted as wrathful, vengeful, and
punishing. The biblical Christ, too, was not without anger, as
when he drove the moneychangers from the temple. How
could an all-knowing, all-powerful, and all-loving God be
susceptible to anger, if anger is indeed a passion (whether
conceived as a form of false judgment or a physiologically
based impulse)?
This question was addressed by Lactantius, a Roman con-
vert to Christianity and confidant to the Emperor Constantine
(fourth century AD). Lactantius criticized the notion that anger
involves a desire for revenge. Rather, anger in the strict sense
(i.e., righteous or just anger) is a mental act motivated to
restrain offenses. Such anger is proper to God and necessary
to humankind, according to Lactantius, for no one who is good
and just can fail to be moved at the sight of evil. We therefore
rise to punishment, not for the sake of revenge, but in order
that morals be preserved and license suppressed. Of course,
Lactantius recognized that angry people sometimes do more
harm than good. That, however, is a feature of 'unjust' anger,
a debasement of true anger that (in a manner similar to
Aristotle) he attributed to interference from bodily reactions.
The views of Aristotle, Seneca, and Lactantius, although
presented here in the most cursory manner, highlight some
of the issues and controversies that still divide contemporary
theories of anger.
Is anger a passion or an action? In terms of the language
we use to describe anger in everyday discourse, it is clearly a
passion. On this point, Aristotle, Seneca, and Lactantius would
agree; however, the reasons they give for this classification are
very different, a fact that might call into question the nature of
the classification itself. Indeed, Lactantius comes close to deny-
ing that anger is a passion, at least when he speaks of divine or
(in humans) righteous anger. Perhaps the closest modern ana-
log of Lactantius's divine anger (a fully rational, disembodied
anger) can be found in the area of Artificial Intelligence. Com-
puters can be programmed to simulate righteous anger, that is,
to evaluate 'offenses,' interrupt ongoing activity, and take cor-
rective measures. We need not postulate special circuitry, no
less some primitive anger 'instinct' or 'drive' on the part of the
computer. All that is required is the programming of appro-
priate rules.
What is the role of physiological arousal in anger? To the
extent that anger leads to vigorous action, it may be accompa-
nied by noticeable bodily changes. There is also evidence that
the physiological changes associated with anger are somewhat
different from those typically associated, say, with fear. (Anger
shows a norepinephrine-like pattern; fear, an epinephrine-like
pattern.) Moreover, research suggests that when a person is
physiologically aroused, for whatever reason, and regardless
of the patterning of the arousal, he or she may be more prone
to anger or to any other emotion appropriate to the situation
than would otherwise be the case. Thus, the potential impor-
tance of physiological arousal during anger should not be
underestimated, but neither should it be overestimated.
Physiological arousal is neither a necessary nor a sufficient
condition for anger, fundamentally, anger is to be understood
primarily in cognitive rather than physiological terms. This
fact was recognized by Aristotle and even more unequivocally
by Seneca and Lactantius.
What is the function of anger? When appropriately experi-
enced and expressed, anger serves an important social func-
tion, namely, the restraint of offense (Aristotle, Lactantius);
indeed, when faced with adequate provocation, anger is not
only appropriate, but it is also a God-given duty (Lactantius).
Needless to say, anger can be, and too often is, used for selfish
and even antisocial ends (Seneca). But a phenomenon must
be understood in terms of its normative application, not its
misapplication, although the latter may be more dramatic and
in need of attention.
To conclude this section, we might ask with regard to
historical teachings in general: Does anger exist outside of a
particular sociohistorical context, or is it a product of that
context? Theorists who adopt the first alternative are some-
times referred to as 'naturalists,' for they believe that anger is
an inherent part of human nature. Those who favor the second
alternative are known as 'social constructionists,' for they
believe that anger is constituted, not just regulated, by social
beliefs and rules. Lor social constructionists, views such as
those presented by Aristotle, Seneca, and Lactantius are of
more than historical interest; they help form the cultural
matrix within which anger is constituted.
Anger in Courts of Law
In contemporary society, the law of homicide both embodies
and promotes social norms with respect to anger. In Anglo-
American common law, two grades of criminal homicide are
generally recognized, murder and manslaughter, and each of
these may be divided into two subcategories, first and second
degree murder, and voluntary and involuntary manslaughter.
Murder is homicide committed with 'malice aforethought,'
that is, deliberately and with the intent of gain (cf. instru-
mental aggression). Second degree murder differs from first
degree murder in the presence of mitigating circumstances.
Manslaughter is a lesser crime than murder. Voluntary man-
slaughter is the technical name for a 'crime of passion,' that is,
homicide committed during emotion, typically anger. It is
'voluntary' because the angry person wants to attack the victim.
Involuntary manslaughter is accidental, but nevertheless cul-
pable homicide, for example, due to negligence.
As just noted, anger serves to mitigate a charge of homicide
from murder to voluntary manslaughter. This is no trivial
matter. A conviction for murder can mean a life sentence or
even the death penalty. By contrast, voluntary manslaughter
carries a much lighter sentence - typically little more than a
few years in prison, and often probation. Why should this
disparity exist? And how does a jury decide whether a defen-
dant was truly angry at the time of the crime?
The law stipulates four major criteria for deciding whether
a person was truly angry at the time of a killing: adequacy of
provocation, heat of passion, insufficient cooling time, and a
causal connection between the provocation and the crime.
'Heat of passion' is a vague reference to the behavior of the
140 Anger
individual at the time of the homicide; and 'causal connection'
refers to the fact that the aggression must be a direct response
to the provocation. The other two criteria - the adequacy of
provocation and insufficient cooling time - are the most inter-
esting from an analytical point of view, for they are judged
by the so-called 'reasonable-man test.' This test stipulates that
the provocation be sufficient to arouse an ordinary member
of the community to anger so intense that it might lead to
homicide and, again according to community standards, that
there be insufficient time for the anger to dissipate before the
homicide is committed.
The reasonable-man test provides ostensibly objective cri-
teria (community standards) against which the feelings and
reactions of the defendant can be compared. If there is a match,
the defendant may be considered to have been in an angry
state; if there is no match, the defendant is not judged to have
been angry, regardless of his or her feelings at the time. Put
differently, the reasonable-man test objectifies anger; no longer
does anger refer simply to a state of mind of the defendant, but
to an objectively existing state of affairs to which the individual
may or may not have attained. The so-called 'reasonable-man'
is, after all, no actually existing person, but is, rather, an idea-
lized embodiment of the norms of the community.
The above considerations suggest that the attribution of
anger in courts of law is a kind of social subterfuge. The
crime of passion is still a crime, but one that is to some degree
excusable. The angry person was, after all, doing his duty by
correcting some wrong as defined by community standards. He
is therefore given a two-way excuse. First, the victim is, in a
sense, put on trial along with the defendant; and if the victim
(instigator) was sufficiently guilty (had committed a provoca-
tion egregious enough to be judged 'adequate'), then he or
she must share the blame for having provoked the incident.
Second, as anger is classified as a passion, the defendant cannot
be held fully responsible for behavior that was presumably
beyond personal control.
Anger in Everyday Affairs
Crimes of passion present a misleading picture of anger in
one very important respect - they are crimes. The presumed
'reasonable-man' has proven himself to be unreasonable in
his resort to excessive violence. To what extent can we general-
ize from crimes of passion to anger in everyday affairs?
Anger is a very common emotion. When asked to keep diary
records, most people report becoming mildly angry once or
twice a day, and seriously angry once or twice a week. The most
frequent targets of anger are friends and loved ones; the most
common instigations involve some perceived misconduct, for
example, an intentional wrongdoing or avoidable accident;
and the motivation of the angry person is to correct or prevent
recurrence of the 'wrong,' not to hurt the instigator.
Anger may be expressed in a great variety of ways, depend-
ing on the person and situation. Common responses include
talking the incident over with the instigator, verbal repri-
mands, withdrawal of privileges, brooding, and so on. At
higher levels of intensity, an angry episode may include such
behaviors as shouting, slamming doors, and stomping out.
Only rarely does anger result in direct physical aggression
against the instigator, at least among adults. In fact, being
unusually kind to or solicitous of the instigator - a 'contrary
reaction' - is more common than direct physical aggression.
Needless to say, it is easy to think of episodes of anger that
do not conform to the above pattern. For example, on occasion
we all have become angry at an inanimate object, as when our
car stalls in the middle of an intersection; in such instances,
however, we typically imbue the object with human qualities
(the car had no 'right' to stall). Overwhelmingly, anger is an
interpersonal emotion; it is directed at persons or other entities
(the self, human institutions) that in some way can be held
accountable for their actions.
Few people find pleasure in their own anger; to be the target
of another person's anger is even more unpleasant. Neverthe-
less, surveys indicate that the consequences of most everyday
episodes of anger are evaluated positively. Even instigators (the
'wrongdoers') report that, in the majority of cases, they gain
from being the target of anger, for example, by coming to
realize their own faults or by gaining a better understanding
of the angry person's point of view.
An important caveat must be added to this last observation:
The more aggressive the response, the less likely it is that anger
will have a positive outcome. Rather than being a typical
manifestation of anger, aggression reflects a failure of anger
to achieve its objective, which is to correct the wrong, not to
hurt the target.
Rules of Anger
On the basis of historical teachings, legal procedures, and
everyday experience, it is possible to infer the norms that
help determine the way anger is experienced and expressed.
Some general rules of anger are presented in Table 1.
All theorists recognize the importance of rules for the expe-
rience and expression of emotion. There is a cmcial difference
among theorists, however, in the way the rules are presumed to
work. Naturalists assume that anger is an inherent part of
human nature and that rules function only to regulate how
the emotion is experienced. Social constructionists, by con-
trast, believe that the rules and associated beliefs not only
regulate but also help constitute anger as a distinct emotion.
There are other differences among these positions as well. For
example, naturalists search for the origins of anger in biological
evolution, whereas social constructionists look toward history;
naturalists see an essential link between anger and aggression,
whereas social constructionists believe that aggression is only
one of many ways in which anger may be expressed; and
naturalists regard anger as universal, whereas social construc-
tionists consider anger to be culturally specific.
The Utku Eskimos supposedly do not have a name for, nor
do they express an emotion equivalent to, anger. Social con-
structionists take such data as support for their contention that
anger is not a biologically based, universal phenomenon; nat-
uralists, on the other hand, argue that the Utku simply sup-
press their anger, or express it in subtle and indirect ways, in
order to preserve harmonious social relationships. There is no
easy way to resolve these divergent interpretations, for the issue
depends partly on how anger is defined, for example, as a
general category or as a specific emotion (cf. the earlier
Anger 141
Table 1 Some rules of anger as inferred from historical teachings, legal procedures, and self-reports of everyday experiences
1 . A person has the right (duty) to become angry at intentional wrongdoing or at unintentional misdeeds if those misdeeds are correctable (e.g., due to
negligence, carelessness, or oversight)
2. Anger should be directed only at persons and, by extension, other entities (one’s self, human institutions) that can be held responsible for their actions
3. Anger should not be displaced on an innocent third party, nor should it be directed at the target for reasons other than the instigation
4. The aim of anger should be to correct the situation, restore equity, and/or prevent recurrence, not to inflict injury or pain on the target, or to achieve
selfish ends through intimidation
5. The angry response should be proportional to the instigation; that is, it should not exceed what is necessary to correct the situation, restore equity, or
prevent the instigation from happening again
6. Anger should follow closely upon the provocation and not endure longer than is needed to correct the situation (typically a few hours or days, at most)
7. Anger should involve commitment and resolve; that is, a person should not become angry unless appropriate follow-through is intended,
circumstances permitting
Source: Averill JR (1993) Illusions of anger. In: Felson RB and Tedeschi JT (eds.) Aggression and Violence: A Social Interactionist Perspective. Washington, DC: American
Psychological Association, with permission from American Psychological Association.
discussion of anger as a synecdoche). However, this much can
be said: as cross-cultural research continues to reveal the great
diversity of ways that human societies have developed to deal
with potential frustrations and interpersonal conflicts, the case
for the universality of anger as a specific emotion becomes
increasingly problematic.
Consider liget, a fundamental emotion among the Ilongot,
a headhunting people indigenous to the Philippines. Like
anger, liget can be occasioned by insults, slights, and other
affronts to the self, and the violation of social customs. Unlike
anger, the most dramatic manifestation of liget is cutting off
the head of another person; the identity of the victim is largely
irrelevant - a man, woman, child, anyone will do, but prefera-
bly a stranger. The taking of a head helps establish one's place
as an equal and honored member of Ilongot society, and hence
may be an occasion for celebration. Liget also finds expression
in a variety of ways other than headhunting. For example, male
liget is implicated in both courtship and childbirth; 'concen-
trated' in the sperm, it helps make babies. Liget also stimulates
work and provides the strength and courage to overcome
obstacles. From these few remarks, it is evident that liget is
constituted by different rules and beliefs than is anger. A crazed
American might cut off the head of another, but only a well-
socialized Ilongot could experience liget. Moreover, if a crazed
American did cut off the head of another, we would not
attribute the response to anger - neither in a court of law
(cf. previous discussion of crimes of passion) nor in everyday
affairs.
In short, the rules presented in Table 1 may seem self-
evident, but that is not because they are universally held.
Rather, they are culturally based norms that, when interna-
lized, become part of our 'second nature.' Other cultures have
different norms, different emotions, different 'second natures.'
Anger and the Social Order
We question the sincerity and even the character (or mental
health) of a person who claims to be angry but who declines to
take action, circumstances permitting (see Rule #7, Table 1).
This link between anger and action is not due simply to the fact
that anger is a reaction to unpleasant events. Of greater impor-
tance is the fact that anger involves a commitment to the
shared values that make social life possible. The demand for
action is thus a social as well as a psychological imperative.
In traditional moral teachings, the failure to become angry
under appropriate circumstances was regarded to be as sinful
as excessive or unjustified anger; in the words of the seven-
teenth century divine, Thomas Fuller, "anger is one of the
sinews of the soul; he that wants it hath a maimed mind."
A modern extension of this traditional teaching can be found
in the popularity of 'assertiveness training.'
In most everyday episodes, the transgressions that provoke
anger are relatively minor - breaking a promise, ignoring a
responsibility, being inconsiderate and the like. This fact tends
to obscure the role that anger plays in maintaining the social
order. No matter how trifling our everyday promises, respon-
sibilities, and considerations may seem to be, they are the
threads from which the fabric of society is woven. If they are
too frequently broken without mend, society unravels. In a
very real sense, then, the many small mendings that are the
everyday experiences of anger, each minor in its own right,
help sustain a way of life.
Anger can be a tool for social change as well as mainte-
nance. There may come a time when we commit ourselves to
values other than those embraced by the larger society. In a
fight for civil rights and equality, for example, we may turn our
anger on people and institutions to whom we previously held
fealty. But the new norms we now seek to establish are, typi-
cally, old norms applied to new circumstances or extended
to new groups (e.g., women, minorities). Even as a tool for
radical social change, anger tends to be conservative of the
norms it embodies.
Anger in Health and Disease
Anger has implications for the health of the individual as well
as society. Evidence for an etiological role of anger is best
documented in the case of coronary heart disease and stroke.
Chronically experienced anger that goes unexpressed, or that is
expressed too vehemently, appears to be especially damaging
to the cardiovascular system, increasing the risk of both coro-
nary heart disease and stroke.
Unexpressed or repressed anger has also been postulated as
a cause of depression. The concept of repression owes much to
Freud and psychoanalysis, although it is not tied to any one
theoretical position. The notion of repressed anger has even
142 Anger
passed into the legal system (thus extending the 'insufficient
cooling time' criterion for crimes of passion). The basic idea is
that, due to situational or personal constraints, anger is too
threatening for the individual to even admit experiencing.
Banned from consciousness, the anger nevertheless finds outlet
in a disguised and maladaptive fashion.
Depression can have many causes, both physiological and
psychological. It need not, and typically does not, have any-
thing to do with repressed anger. There are occasions, however,
when depressive-like symptoms can take on some of the qua-
lities of an angry response. Consider the case of a woman who
is so fatigued that she finds it difficult to get out of bed in the
morning, feels despondent through most of the day, and is
haunted by frequent thoughts of suicide. Life becomes a bur-
den both to her and to those around her. Now add the follow-
ing considerations: the woman's husband treats her with
neglect; she sincerely believes that she has no right to be
angry with her husband, for his neglect of her is somehow
deserved; nevertheless, in subtle ways her behavior causes con-
siderable inconvenience to her husband; and, finally, she
derives barely concealed satisfaction from her husband's dis-
comfort. In such a case, we might well speak of the woman's
depression as being a manifestation of 'repressed' anger. That,
however, is a metaphorical way of speaking, for the woman's
behavior follows only some, but not all, of the standard norms
for anger. A more parsimonious explanation might be that the
woman has acquired, in this situation, at least, a set of idiosyn-
cratic beliefs and rules that help constitute an emotion that
resembles both anger (retaliation for a perceived wrong) and
depression (fatigue, despondency, self-reproach).
Recall the emotion of liget described earlier. It bears some
semblance to anger within our own culture, but it is consti-
tuted by different norms and hence represents a different emo-
tion. Norms of emotion can vary within, as well as between,
cultures. There exist subcultures of increasingly smaller scope,
down to the level of the family and ultimately the individual.
An emotion that is the product of personal (idiosyncratic)
norms is particularly subject to misinterpretation because
usual forms of consensual validation are lacking. But nothing
is gained by way of clarity, and much may be lost, if we
label the nonconforming emotion by some common name
('anger,' say, or 'depression') and attribute its idiosyncrasy to
repression.
Less often recognized than its pathogenic effects, anger can
also have ameliorative effects on the course of disease. For
example, patients suffering from serious diseases such as cancer
have a better prognosis if they become angry at their condition
than if they become merely frightened or depressed. Although
the underlying mechanisms are not clear, a possible explana-
tion is that the angry person is more likely to 'fight' the illness,
for example, by taking an active role in his or her treatment.
Catharsis
The idea of repression is closely tied to that of catharsis. Dur-
ing psychoanalytic therapy (what Freud originally called
the 'cathartic method'), presumably repressed impulses are
brought to consciousness and expressed in an adaptive fash-
ion. LIsing a rather graphic metaphor, Freud compared cathar-
sis to the draining of pus from an infected wound. Continuing
with the example of depression, let us suppose the woman
described earlier turns to psychotherapy for help. During treat-
ment, she may come to recognize on a conscious level the
unfairness of her treatment, and, blaming her husband rather
than herself, she may resolve to do something to change the
situation. From a psychoanalytic perspective, it might be said
that her depression 'lifts' as her repressed anger is 'released.'
An alternative explanation is that one kind of emotional
response (an ill-defined anger/depression combination) has
been replaced by a more adaptive response (normative anger).
A good deal of experimental research has been devoted to
the possible cathartic effects of emotional expression, espe-
cially with respect to anger and aggression. On the whole, the
results have not been kind to the catharsis hypothesis. More
often than not, the expression of anger on one occasion facil-
itates, and, on subsequent occasions, encourages its expres-
sion. This is opposite to the mitigating effect predicted by the
catharsis hypothesis. In the case of anger and aggression, as in
so many other areas, practice makes perfect or, at least, easier.
When Anger Turns to Hate
In popular conception, anger and hate are closely allied con-
ditions. In actuality, the matter is not so simple. As mentioned
earlier, anger is more likely to be directed at friends and loved
ones than at strangers and enemies. One reason is opportunity:
close contact with friends and loved ones offers frequent occa-
sions for anger. Another reason is that friends and loved ones
can hurt us in ways that strangers and enemies cannot; a breach
of trust, for example, can only occur where tmst exists. Most
importantly, if we want a relationship to continue, perceived
wrongs must be set right, and anger is one way of achieving
that end. It is primarily when anger proves ineffective, or is
pursued for selfish ends, that it may turn to hatred on the part
of the target as well as the angry person.
Put differently, unresolved anger may cool and solidify into
an enduring animosity, which predisposes an individual to a
variety of different emotions depending on the fortunes of the
hated person; thus, we may feel sorrow at the success of our
enemies and joy ( schadenfreude ) at their failures, as well as
anger at their presumed transgressions. We may even hate
persons or entire groups who have never done us any wrong,
as in prejudice. While normative anger is focused on the mis-
deeds of others, real or imagined, and attempts at repair, hatred
has an inner dynamic that is more purely destructive of the self
as well as of others. In his novel Demian, Hermann Hesse
observed: "If you hate a person, you hate something in him
that is part of yourself. What isn't part of ourselves doesn't
disturb us." That is both an overgeneralization (not all hatreds
fit the pattern) and an understatement (something similar
could be said of other sentiments; e.g., if you love a person,
you love something in him that is part of yourself). Keeping
these qualifications in mind, Hesse's observation contains a
truth about hatred that is worth remembering.
Gender Differences
A common stereotype is that women are less capable of anger
than are men. This stereotype is also sometimes used to explain
Anger 143
the greater frequency of depression diagnosed among women
than men (cf. previous discussion of repressed anger). How-
ever, surveys of the everyday experience of anger suggest that
women become angry as often, as intensely, and as effectively
as men. Gender differences in anger, to the extent that they
exist, must be sought on a more subtle plane than suggested by
the stereotype.
In the conduct of interpersonal relations, men tend to place
greater emphasis on competition and individual rights, whereas
women tend to place greater emphasis on cooperation and
affiliation. Men also tend to be physically more aggressive than
women. Such differences in orientation and response style can-
not but have reverberations on the way anger is experienced and
expressed. Some of the things that a man finds offensive may
not offend a woman, and vice versa; and when angry, a man
may do and say things that a woman typically would not (e.g.,
men are more likely than women to engage in physical violence
when angry, and women are more likely than men to cry) . But as
important as such differences may be, they should not be inter-
preted to mean that anger is less appropriate to women than to
men, or that women are less capable of anger than are men.
'Different' does not necessarily imply 'more' or 'less.' Women are
as sensitive as men to unfair treatment; and in response to
provocations that they deem adequate, women are as likely as
men to make their feelings known clearly and effectively.
Wrongful Anger
The rules presented in Table 1 suggest that anger, properly
constituted, serves to correct socially recognized wrongs, and
that the anger should not exceed what is necessary to achieve
that end. And, as described earlier, the majority of episodes
fit that description. However, the rules of anger are easily
broken, either unintentionally or willfully, with consequences
that are harmful to the self and to interpersonal relationships.
When the rules are violated, we may speak of wrongful
anger (provided the violation is not so egregious as to preclude
the response from being classified as anger altogether). Three
types of wrongful anger have already been discussed in passing,
namely, anger that goes unexpressed, resulting in continued
provocation and unrelieved tension; anger that is expressed
indirectly or in accordance with idiosyncratic rules; and anger
that is expressed intemperately, resulting in unwarranted vio-
lence. To these, we could add anger that is unduly prolonged, is
displaced on innocent third parties, is oft-repeated without
adequate follow-through, and so forth.
The rules of anger may be violated for many reasons, some
of which antedate the angry episode - for example, inadequate
socialization, neurological dysfunction, and inordinate stress
or fatigue. Extraneous stimuli in the immediate situation can
also lead to inappropriate anger. According to Berkowitz's
neoassociationistic theory of anger and aggression, aversive
stimuli of any kind (which would include aspects of most
provocations to anger) tend to automatically elicit aggressive
thoughts, feelings, and behavioral inclinations, either innately
or through prior experience. These tendencies are especially
likely to find expression if other aggressive cues are present
(e.g., a weapon) or customary restraints are inhibited (e.g.,
through the use of alcohol). By contrast, the present article
has focused on higher order sociocognitive mechanisms by
which anger may serve as a rationale for aggression, before,
during, and after the fact, regardless of what other precipitating
factors may be involved. The two types of mechanisms (socio-
cognitive and neoassociationistic) are not mutually exclusive;
indeed, they tend to complement one another. For a more
complete discussion of how anger can go awry, the reader
should consult the article on aggression in this Encyclopedia.
Regardless of the precipitating factors, with a better under-
standing of what constitutes an adequate provocation and with
the acquisition of effective coping skills, aggressive tendencies
can be typically curbed. But the outcome of anger - whether
constructive or destructive - does not depend exclusively on
the behavior of the angry person. Anger is more like a dialog
than a soliloquy; put differently, there are 'rules of engagement'
for the target as well as for the angry person. The most benefi-
cial response for a target (assuming the anger is justified) is an
explanation to correct any misunderstanding, an apology if no
misunderstanding exists, and a resolve to prevent recurrence.
Too often, unfortunately, anger provokes anger, conflict esca-
lates, attitudes harden, and resentments build; eventually, all
goodwill may be squeezed from the relationship (see earlier
discussion of when anger turns to hate).
When goodwill is absent, the rules of anger may be violated
'wantonly,' so to speak. That is, anger may be used deliberately
(if not consciously) to achieve ends that are only tangentially
related to the instigation. The concept of anger, it will be recalled,
has exculpatory implications, and this could be in two ways:
firstly, as anger is commonly interpreted as a passion rather
than an action, the angry person presumably cannot control
his or her own behavior; secondly, if the anger is accepted as
justified, the blame for initiating the incident is shifted from the
angry person to the instigator. Not surprisingly, perpetrators and
their apologists often seek to excuse vicious behavior - child
abuse, spouse battering, mugging and robbery, gang warfare,
urban riots, etc. - by attributing it to anger. And in a multitude
of more minor ways, we all sometimes resort to anger to intimi-
date and coerce others in the pursuit of selfish ends.
Wrongful anger is unwittingly fostered by the misguided
advice, often found in popular psychology books, that it is
healthy to 'let your anger out.' If a response is inappropriate
to begin with, letting it out in the name of anger is doubly
inappropriate, for a response cannot be labeled as anger with-
out endorsing, no matter how unwittingly, its exculpatory
implications. It would be far better to heed the warning of
Aristotle: "Anyone can get angry - that is easy; . . . but to do
this to the right person, to the right extent, at the right time,
with the right motive, and in the right way, that is not for
everyone nor is it easy; wherefore goodness is both rare and
laudable and noble" ( Nichomachean Ethics, 1109a25).
See also: Aggression; Depression; Homicide.
Further Reading
Averi 1 1 JR (1 982) Anger and Aggression: An Essay on Emotion. New York:
Springer-Verlag.
144 Anger
Averill JR (1993) Illusions of anger. In: Felson RB and Tedeschi JT (eds.) Aggression
and Violence: A Social Interactionisl Perspective. Washington, DC: American
Psychological Association.
Berkowitz L (2003) Affect, aggression, and antisocial behavior.
In: Davidson RJ, Scherer KR, and Goldsmith HH (eds.) Handbook of
Affective Sciences, pp. 804-823. New York, NY: Oxford University
Press.
Briggs JL [WO) Never in Anger: Portrait of an Eskimo Family. Cambridge, MA:
Harvard University Press.
Finkel NJ and Parrott WG (2006) Emotions and Culpability. Washington, DC: American
Psychological Association.
Friedman HS (ed.) (1991) Hostility, Coping, and Health. Washington, DC: American
Psychological Association.
Geen RG (1990) Human Aggression. Pacific Grove, CA: Brooks/Cole.
Rosaldo MZ (1980) Knowledge and Passion: llongot Notions of Self and Social Life.
Cambridge, England: Cambridge University Press.
Rosenwein BH (ed.) (WD&) Anger's Past: The Social Uses of an Emotion in the Middle
Ages. Ithaca, NY: Cornell University Press.
Tavris C {\%9) Anger: The Misunderstood Emotion, rev edn. New York: Simon and
Schuster.
Thomas SP (ed.) (1993) Women and Anger. New York: Springer.
Wyer RS and Srull TK (eds.) (1 993) Advances in Social Cognition, Vol. 6: Perspectives
on Anger and Aggression. Hillsdale, NJ: LEA.
Zillmann D (1983) Transfer of excitation in emotional behavior. In: Cacioppo JT and
Petty RE (eds.) Social Psychophysiology: A Sourcebook, pp. 215-240. New York:
Guilford Press.
Animal Cognition
L Castro and E A Wasserman University of Iowa, Iowa City, IA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Associative learning Making the psychological connection
between contiguous events in the environment.
Categorization Placing different objects or events into
separate classes or categories.
Conceptualization The process of forming an abstract or
generic idea from particular instances.
Discrimination Responding differently to two or more
stimuli differing in one or more respects.
Generalization Transferring a learned response from one
stimulus to another.
Matching-to-sample Experimental procedure for studying
memory in which reward is given for responding to a
stimulus if it is the same as a prior sample stimulus.
Memory Control of current behavior by past stimulation.
Metacognition Thinking about one's own cognitive states
and processes.
Numerical processing Set of abilities related to
understanding and manipulating numbers, such as: using
symbols to denote quantities; discriminating, ordering, and
comparing different quantities; and, combining quantities
in order to perform arithmetic operations.
Reinforcer Any consequence of a behavior that increases
the probability that the organism will repeat that behavior.
Time-out A brief period of time during which there is no
stimulation; time-out is a form of punishment that is used to
decrease the probability that the organism will repeat a
behavior.
The Science of Animal Cognition
For centuries, it was believed that a clear discontinuity held
between the cognitive capabilities of humans and animals.
Philosophers and other thinkers tended to be quite skeptical
of animals' mental abilities. For example, the great French
philosopher Rene Descartes deemed animals to be mere
machines whose actions could be explained without invoking
any cognitive processes whatsoever. In his Letter to the Marquess
of Newcastle, Descartes proposed that "the reason why animals
do not speak as we do is not that they lack the organs but that
they have no thoughts. " Approximately 5 0 years later, the great
British philosopher John Locke, in An Essay Concerning Human
Understanding, joined Descartes in disparaging the cognitive
capacities of animals: "the having of general ideas is that
which puts a perfect distinction betwixt man and brutes, and
is an excellency which the faculties of brutes do by no means
attain to."
All that changed some 200 years later when Charles Darwin
advanced his revolutionary ideas about the evolution of spe-
cies. In The Descent of Man, Darwin went beyond suggesting the
evolution of physical traits to proposing the evolution of the
mind as well. For Darwin, there was "no fundamental differ-
ence between man and the higher mammals in their mental
faculties" (p. 66). Human and animal intelligence differ only
in degree, not in kind, Darwin argued; so, there should be no
sharp schism between human and animal mind. The human
mind might be the final step in the evolution of intellectual
functions, but the roots of human mental processes should be
observable in animals as well.
Darwin's unsettling ideas had actually been presaged
by other philosophers and psychologists. In his Principles of
Psychology, Herbert Spencer suggested that the mind could be
understood only by examining how it had evolved. According
to Spencer, mental capacities could be represented along a
continuum with no great gaps: a continuous progression
from simple associative learning to complex forms of abstrac-
tion and reasoning.
The ideas of Darwin and Spencer on the evolution of the
mind were staunchly defended by Thomas Huxley, who wrote
in 1874 that: "the doctrine of continuity is too well established
for it to be permissible to suppose that any complex natural
phenomenon comes into existence suddenly and without being
preceded by simpler modifications; and very strong arguments
would be needed to show that such complex phenomena as
those of consciousness first made their appearance in man."
These principles of evolutionary biology clearly clashed
with Cartesian philosophy. The gradual evolution of cogni-
tion was a direct consequence of this new biological vantage
point. Critically, within that new paradigm, it was entirely
plausible to believe that mental capabilities could be observed
in organisms other than humans.
Once the hypothesis of mental continuity was enunciated,
evidence to support it was needed. Darwin himself amassed a
large collection of supportive anecdotes: stories told by natur-
alists, zookeepers, and pet owners attesting to how smart ani-
mals really were. For example:
Dr. Hayes, in his work on The Open Polar Sea, repeatedly remarks
that his dogs, instead of continuing to draw the sledges in a compact
body, diverged and separated when they came to thin ice, so that
their weight might be more evenly distributed. This was often the
first warning which the travelers received that the ice was becoming
thin and dangerous.
Darwin was alert to the hazards of relying on anecdotal
evidence alone; such observations lacked the information
needed to pinpoint the mechanisms responsible for the ani-
mals' behavior. Relevant to the prior example, Darwin won-
dered about the origins of the dogs' behavior: "now, did the
dogs act thus from the experience of each individual, or from
the example of the older and wiser dogs, or from an inherited
habit, that is from instinct?"
145
146 Animal Cognition
Any bona fide scientific analysis of animal cognition
requires careful recording of the relevant behaviors as well as
precise knowledge and control of the variables influencing
or determining that behavior. Darwin's disciple, George
Romanes, advocated for an objective analysis of the mind in
his book Animal Intelligence: "We can only infer the existence
and the nature of thoughts and feelings from the activities of
the organisms which appear to exhibit them." (p. 1). Yet,
despite his advocacy for and his interest in documenting men-
tal continuity across species, Romanes' work too was simply a
compilation, albeit extensive and thorough, of still more infor-
mal observations and anecdotes.
The true transition to the scientific study of animal cogni-
tion began with C. Lloyd Morgan. In his Introduction to Com-
parative Psychology, Morgan emphasized the importance of
distinguishing between an animal's behavior and the interpre-
tation of that behavior. In order to properly interpret an organ-
ism's behavior, high standards of objectivity are needed. As
well, systematic experimental studies are required in order
to reach incisive conclusions as to the origins of behavior,
which he himself undertook with baby chicks whose develop-
mental history could be known and carefully controlled.
Finally, circumspection was to be one of Morgan's enduring
contributions. Morgan's famous canon urged researchers to be
cautious in advocating advanced cognitive interpretations
when the behavior in question might be more parsimoniously
explained by simpler behavioral processes.
These then were the forebears of the science of comparative
cognition. Following the methods of natural science, research
in comparative cognition today seeks to discover the extent to
which our mental capacities are similar to those of other ani-
mals and to answer many longstanding questions. Are animals
intelligent? Is human intelligence superior to animal intelli-
gence? What are the cognitive and neurobiological mechan-
isms of intelligent action?
In this article, we will consider a few particularly promi-
nent and promising areas of research in animal cognition:
numerical processing, memory and planning, conceptualiza-
tion, and metacognition. This sampling of topics, issues,
methods, and findings should provide readers with a good
taste of the current state of the science of animal cognition.
A broader compilation of animal cognition studies can be
found in Comparative Cognition edited by Wasserman and
Zentall.
Numerical Processing
Number is an important property of objects in the external
world. Critically, numerical processing can be considered to be
mediated by an abstract cognitive process, because behavioral
control by number must be made irrespective of the specific
physical features of the items to be evaluated; 2, 5, or 8 items
are 2, 5, or 8 regardless of whether the items are apples,
pebbles, or geckos. Traditionally, it has been believed that
linguistic competence is necessary for numerical processing.
Although it is the case that a verbal system does mediate several
numerical and mathematical operations, considerable evi-
dence has been obtained documenting numerical processing
by animals and infants lacking language; this evidence suggests
that common preverbal processes underlie numerical and
mathematical competencies in humans and animals.
Basic numerical abilities comprise: learning and using sym-
bols to denote specific quantities; discriminating, ordering,
and comparing different quantities; and, combining quantities
so that basic arithmetic operations, such as addition and sub-
traction, can be accomplished.
The Japanese primatologist Tetsuro Matsuzawa taught a
5-year-old chimpanzee, Ai, to use Arabic numerals to denote
the number of items in a display. Ai was first presented with
either 1 or 2 items in a display window and then required
to choose either the 'I' or the '2' key on a numeric keyboard.
As Ai mastered the lower numbers, higher numbers of items
were progressively introduced, up to a maximum of 6. Regard-
less of the items being colors, objects, or symbols, Ai's accuracy
exceeded 90%. Nevertheless, Ai had greater difficulty reporting
the numbers when they were close to one another (e.g.,
it was easier for Ai to distinguish 2 from 5 than to distinguish
2 from 3) and when the numbers were at the higher end of
the scale (e.g., Ai was more accurate at identifying 2 and 3
than at identifying 5 and 6, even when the disparity between
them was in both cases 1 item). It seems that, as quantities
increase, the discrepancies between them become less obvious
and, in turn, the discrimination between them becomes more
challenging.
Interestingly, Ai's behavior is not an isolated curiosity.
Humans too find it easier to distinguish between 2 and 14
items than between 4 and 5 items. This finding is called the
numerical distance effect. As well, humans can rapidly distin-
guish between 2 and 3 items, but we cannot so rapidly say if
there are 14 or 15 items, in this case showing what is called the
numerical magnitude effect. The numerical magnitude effect
has also been found in monkeys, rats, and pigeons.
Recent research has found that rhesus monkeys can learn to
choose the smaller of two simultaneously presented sets of
items when such nonnumerical factors as area and density
are controlled. Monkeys can also choose the larger of two
quantities even when the items are sequentially presented, as
when the items fall one by one into an opaque container.
In addition, Brannon and her colleagues have found that
monkeys' accuracy rises as the disparity between the amounts
to be evaluated increases. Indeed, the monkeys' performance
is affected by the ratio of the values being compared. As pre-
dicted by Weber's law, the larger the amounts, the greater
the difference between them must be in order to be distin-
guishable. This relationship was true for human college
students as well.
These and other observations have led researchers to pro-
pose that humans and animals share a system that calculates
inexact, but approximate representations of quantity. For
example, when having to choose among different waiting
lines, we tend to select the shortest. We rarely do so by precisely
counting the people in line, but rather do so by a process of
rough estimation; we use an approximate preverbal number
system. It is not difficult to imagine that animals in the wild
benefit from the ability to distinguish between trees containing
5 and 15 fruits or predators attacking in groups of 2 or 4. Small
or large quantities can surely make a difference and numerosity
judgments in these situations seem to be similarly performed
by verbal and nonverbal organisms.
Animal Cognition 147
Estimating and representing numerical values appears to be
necessary for the next step in numerical reasoning: interrelating
quantities so that operations like adding, subtracting, multi-
plying, and dividing can be accomplished. Can animals per-
form these basic arithmetic operations? Studies using a
preferential looking paradigm have suggested that preverbal
human infants and monkeys can understand simple arithmet-
ical operations, such as adding and subtracting a small number
of visually presented objects.
For example, when monkeys watched as two eggplants were
placed behind a screen, they looked longer when the screen
was removed and only one eggplant was present than when
two eggplants were present. Monkeys may have spent more
time looking at the incorrect outcome because it surprisingly
violated the rules of arithmetic. But, the results of such studies
may be explained in terms of the infants' and monkeys' under-
standing of object permanence: organisms can keep track of
occluded objects and when one object unexpectedly appears
or disappears, they are surprised; no real arithmetic ability is
required in that case.
More compelling evidence of addition in monkeys has been
found by Brannon and Cantlon. They presented animals with
two sets of dots on two sequential screens. Then, on a third
screen, the monkeys had to choose from two sets: one contain-
ing the number of dots equal to the sum of the two sets and a
second containing a different number of dots (see Figure 1).
Monkeys learned to choose the array that roughly corre-
sponded to the arithmetic sum of the two sets of dots and
they readily transferred this behavior to novel combinations.
Some problems proved to be more difficult than others.
For example, monkeys were faster and more accurate when,
after being presented with a set of 2 dots and a second set of
6 dots, they had to choose between 8 and 14 dots compared to
when they had to choose between 8 and 1 0 dots. The closer the
two possible answers, the more difficult the task.
When humans were given the same nonverbal task, their
pattern of performance was strikingly similar. Although
humans were generally more proficient than monkeys, both
Figure 1 A monkey performing the addition task designed by Brannon
and Cantlon. The monkey had to choose from two sets of dots: one set
containing the same number of dots as the sum of two previously
presented sets and a second set containing a different number of dots.
Photograph courtesy of Elizabeth Brannon, Duke University.
species' reaction times and accuracies were affected by the ratio
between the numerical values of the choice arrays: the larger
the numerical difference between the correct and the incorrect
choices, the better the performance. These results further sug-
gest that humans and animals share an approximate calcula-
tion system that does not involve verbal processing.
Research in numerical processing thus shows that animals
possess the evolutionary precursors to advanced mathematical
abilities: animals can identify and name quantities, compare
and discriminate different quantities, and even manipulate
those quantities to perform simple additive operations.
Memory
There is no question that animals can learn new information.
Pavlov's classical conditioning experiments showed that, after
presenting dogs with a sound followed by food, the mere
presentation of the sound alone would make the dogs salivate,
a response that was initially elicited only by food. We can infer
that the dogs learned that the sound predicted food. And, if
animals can learn, then that implies they remember the events
they experienced in the past. So, the fact that the dogs came to
salivate to the sound suggests that they remembered the pair-
ing of sounds with food.
Unlike humans, animals cannot verbally express their recol-
lections; but, there are other ways we can study their memory.
One popular procedure is delayed matching-to-sample. In the
basic task, a sample stimulus (a colored light or a picture) is
presented for a few seconds. Then, the sample is removed and a
delay ensues. After this delay (which can range from seconds to
minutes), the sample is again presented along with one or
more comparison stimuli. The animal's task is to pick the
choice stimulus that matches the sample. Monkeys and
pigeons successfully perform this task. Moreover, their accu-
racy generally decreases as the delay increases, which is true for
humans as well. Further, animals can perform this matching
task even when new sample and comparison stimuli are pre-
sented to them, showing effective transfer to a novel situation.
Human memory studies have ttaditionally involved
learning and retrieval of lists of items. People are given lists
of items and they must later recall as many items as possible. In
this situation, the first and the last items in the list tend to be
better recalled - primacy and recency, respectively. Animals too
show these serial position effects.
The memory of pigeons, monkeys, and humans for lists of
visual items has been directly compared after different delays
between the last item and the recognition test. All three species
show the same pattern: strong recency, but no primacy at the
shortest delays; the classic U-shaped serial position function at
intermediate delays; and, strong primacy, but no recency at the
longest delays. Thus, the same mechanisms may mediate the
serial memory performance of all three species.
Although similarities between species abound, there are
disparities too. Sometimes, animals can be even more profi-
cient than humans! One noteworthy example comes from
Tetsuro Matsuzawa's laboratory in Inuyama, Japan. Chimpan-
zees were first taught to contact each of the numerals from 1 to
9 in order when they appeared in random locations on a
computer touchscreen. Even when the sequence lacked some
148 Animal Cognition
Figure 2 A chimpanzee performing the memory task designed by Matsuzawa and his colleagues. On the left, the chimpanzee is touching in sequential
order the numerals from 1 to 9 located randomly on the screen. On the right, after brief presentation, the numerals were covered by white squares;
here too, the chimpanzee has to touch each of the squares in the correct sequence. This procedure is detailed in Inoue S and Matsuzawa T (2007)
Working memory of numerals in chimpanzees. Current Biology 17: R1004-R1005. Photographs courtesy of Tetsuro Matsuzawa, Primate
Research Institute, Kyoto University.
of the numbers (e.g., 1-3-4-6-8-9), the chimpanzees could still
respond in sequential order from the smallest to the largest.
Humans could also do this, of course. But, a critical variation
was later introduced. The numerals were presented for a very
short duration (from 650 to 210 ms) and then they were
replaced by white squares (see Figure 2). Now, subjects had
to remember which numeral appeared in which location and
then touch each of them in the correct sequence. Humans'
performance plummeted, but chimpanzees' performance
remained high. The chimpanzees could retain an accurate
image of the very brief visual scene, an ability that surpasses
our own perceptual memory capacity. This eidetic (or photo-
graphic) memory can sometimes be seen in young children,
but very rarely in adults. This striking developmental disparity
was also seen in chimpanzees; the young animals were the best
performers in this task.
Despite these remarkable illustrations of animal memory,
some memory capacities have been deemed to be uniquely
human. Episodic memory refers to the ability to remember
specific events in time and space, thereby involving: the con-
tent of an experience (what), its place of occurrence (where),
and its time of occurrence (when).
Many mammals and birds are food-hoarding animals; they
cache food in specific places for future use. Food hoarding
implies that: (1) food has to be stored and concealed and (2)
actual consumption is deferred for hours, days, weeks, or
months. Successful hording appears to require animals to use
past spatial and temporal information to retrieve food later.
In a series of experiments with scrub jays, Nicola Clayton
and her colleagues allowed animals to store and recover worms
and peanuts. Fresh worms are the scrub jays' preferred food.
But, the worms do not last very long; after a few days, they
degrade and become unpalatable. Peanuts are less appetizing,
but they are nonperishable; so, they can be consumed at any
time. The jays were trained in the laboratory to cache worms on
one side of a distinctive tray and to cache peanuts on the other
side (Figure 3 shows a scrub jay caching food from a tray).
Critically, the jays were allowed to retrieve their caches either 4
or 124 h later. After 4 h, the birds tended to inspect the side of
the tray where the worms should be; but, after 124 h, the birds
tended to inspect the side where the peanuts should be. No
olfactory or visual food cues were available; the jays had to rely
on their memory at the time of cache recovery.
Figure 3 A scrub jay caching worms and nuts in the compartments of a
sand-filled ice-cube tray. Photograph courtesy of Nicola S. Clayton,
University of Cambridge.
Evidently, the jays can remember: (1) whether peanuts or
worms had been cached - what; (2) on which side of the tray
each of the food items had been stored - where; and (3),
whether a short or a long time had elapsed since the food
items had been cached - when. Thus, the cache and retrieval
behaviors of the scrub jays meet the criteria for episodic
memory.
Many researchers believe that such successful and elaborate
food caching suggests that animals plan for the future. How-
ever, in order to speak of planning, current behavior must be
based on its future consequences and must be independent of
the animal's prevailing motivation. Clayton and her colleagues
asked if scrub jays could anticipate their future need states and
act accordingly. The researchers placed the jays into two differ-
ent compartments on alternate mornings for 6 days. In one
compartment, food was always there in the morning; in the
other compartment, the bowl was empty. Following 6 days of
exposure to the two compartments, the jays were unexpectedly
given a bowl of cacheable pine nuts after the evening meal;
they were also given free access to the two different compart-
ments in which the caching trays had been placed. The jays
could either eat the pine nuts or store them in one or the other
of the two compartments. The birds opted to store the food,
but not randomly; they presumably anticipated their future
needs and cached more food in the compartment in which
food was not going to be available in the morning. Thus, it
seems that scrub jays can plan for a future motivational state.
Animal Cognition 149
Conceptualization
It is generally admitted that the higher animals possess memory,
attention, association, and even some imagination and reason. If
these powers are capable of improvement, there seems no great
improbability in more complex faculties, such as the higher forms
of abstraction, . . . having been evolved dirough the development
and combination of the simpler ones.
Darwin's speculation that higher-order cognitive capabil-
ities are the evolutionary outgrowth of more primitive cogni-
tive capacities, has bolstered the search for evidence of abstract
concept learning in animals. Over the last quarter century, the
extent to which animals, especially those phylogenetically dis-
tant from humans, can learn abstract concepts and, particu-
larly, understand the relations of sameness and differentness,
has become a focal concern of this work.
Animals must possess at least some primitive sense of
sameness and differentness. Animals are comfortable with
members of their own species, but they flee from members of
alien species. Plus, animals behave consistently in familiar
situations, but they change their actions when something
unusual transpires. For example, when uninformative stimuli
repeatedly occur, a wide range of animals cease responding to
them. This phenomenon of habituation requires the animal to
perceive a particular stimulus as the same as one presented
before. By contrast, when a second novel stimulus is intro-
duced along with the habituated stimulus, the organism starts
responding again - dishabituation. Now, the organism detects
something different in the situation and its behavior changes
accordingly.
We also know that behaviors which are associated with
some stimuli in some contexts generalize to other stimuli in
other contexts. Indeed, generalization and its counterpart, dis-
crimination, are the foundations of categorization, another
familiar instance of conceptual behavior. When an organism
makes the same response 'car' to discriminably different cars, it
is generalizing among all cars; when an organism makes the
same response 'flower' to discriminably different flowers, it is
generalizing among all flowers; at the same time, it is discrimi-
nating flowers from cars. Categorization thus involves general-
ization within a class, but discrimination between classes. If
sameness within a category and differentness between cate-
gories are not perceived, then each event would have to be
treated as entirely unique; animals would need to learn the
appropriate response each time that each event is encountered,
an excruciatingly demanding and inefficient chore.
Pigeons can learn to respond to photographs that contain a
human being and not to respond to similar photographs in
which human beings are absent. Even when people in the
pictures differ in age, size, race, sex, clothing, and posture,
pigeons learn to respond to the photographs when they por-
tray people with a high level of accuracy.
Pigeons can also learn categorization tasks which are even
more similar to the real-world situations that confront
humans. Wasserman and his colleagues wanted to see if
pigeons could report which stimuli belonged to which of
four different categories of objects: cats, flowers, cars, and
chairs. One image at a time from each category was presented
on a viewing screen. Four report keys were also available, each
Figure 4 Illustration of one trial of the categorization task designed
by Wasserman and his colleagues. In this case, a stimulus from the
flower category is presented on the screen. The four report keys
correspond to each of the four experimental categories; cats, flowers,
cars, and chairs. For each of the categories, the pigeons had to choose
the appropriate key. Text labels are included for explanatory purposes;
they did not appear on the screen.
corresponding to a different category; these four different keys
effectively served as four different 'words' for the pigeons (see
Figure 4). Pigeons learned to select the appropriate key for the
different categories. After learning, the birds were given novel
items from the training categories to see if they would exhibit
reliable transfer from the training stimuli, the critical test to
document concept learning. The pigeons passed the test; they
successfully used the four report keys to classify the novel items
from the four trained categories.
It should be noted that items belonging to such categories
as cats, flowers, cars, or chairs share a number of physical
properties; so, it is generally accepted that perceptual similarity
guides classification learning in animals, as it does in humans.
A higher level of conceptualization is represented by relational
concepts, which do not depend on perceptual similarity, but
require learning about the relations between or among two or
more stimuli. Here, the absolute properties of the stimuli must
be transcended and knowledge of universal applicability must
be extracted.
In order to speak of a true same-different concept, rela-
tional learning and generalization to novel situations must be
demonstrated. Habituation and categorization suggest that
abstract conceptualization may be within animals' capabilities,
but more explicit evidence is required. Initial efforts to teach
pigeons to report whether 2 items - the smallest number that is
possible to make a same-different discrimination - are the
same as or different from one another were not successful.
But, perhaps it would be easier for pigeons to learn same-
different discriminations with displays of items containing
more than two stimuli.
Therefore, Wasserman and his colleagues trained pigeons to
peck one button when a stimulus array comprised 16 identical
icons and to peck a second button when a stimulus array
comprised 16 nonidentical icons (Figure 5, top). Now,
pigeons readily learned the discrimination to high levels of
accuracy and, critically, they later transferred the discrimina-
tion with little decrement to both identical and nonidentical
arrays constructed from novel visual items.
150 Animal Cognition
Same Different
Figure 5 Examples of 16-item same and different arrays used in
Wasserman and colleagues’ research on same-different discrimination.
The top panel shows orderly arrays, whereas the bottom panel shows
disorderly arrays. Pigeons are highly adept at discriminating both orderly
and disorderly arrays.
One characteristic of the same and different arrays depicted
in Figure 5 is that they differed in their spatial orderliness; the
same arrays entailed clear horizontal and vertical regularities
that the different arrays lacked. Pigeons could have been
performing a perceptual rather than a conceptual discrimina-
tion. Remember Morgan's canon of parsimony?
Nonetheless, when pigeons were trained to discriminate
16-icon Same from 16-icon Different arrays in which the
items were randomly placed on the screen, the birds readily
acquired the discrimination and showed excellent transfer to
new arrays (Figure 5, bottom). Moreover, randomly rotating
the items in an array did not adversely affect pigeons' discrimi-
nation behavior. Finally, when pigeons were shown successive
lists of same and different items on a one-at-a-time basis -
thereby making spatial orderliness an unusable cue for solving
the discrimination - the pigeons still exhibited robust discrimi-
nation learning and transfer performance. This pattern of results
discredits simple perceptual accounts and strongly suggests that
pigeons in these studies had acquired a same-different concept.
Even Morgan would have been convinced.
To discriminate collections of same and different items
involves understanding what are called first-order relations.
Can animals go to the next level and leam higher-order rela-
tional concepts? Can they understand, not only that several
identical apples are the same and that several identical balls are
the same, but also that the relation between the apples and the
balls is sameness? Judging relations between relations is basic to
analogical reasoning; many authors have proposed that analogi-
cal competence is the very essence of human intelligence.
In an analogical or relational matching-to-sample task, the
animal is given a sample stimulus set (either two or more
identical items on some trials and two or more nonidentical
items on other trials), and two choice stimulus sets (one con-
taining two or more identical items and the other containing
two or more nonidentical items). Critically, none of the items
in the sample is presented in the choice sets; so, if the sample is
AA, then the choices can be BB or CD. To be successful, the
animal must select the set of choice alternatives that instanti-
ates the same relation as the sample set. Given that there is no
overlap between the sample and choice items, only attention
to the matching relations (same sample to same choice and
different sample to different choice) can yield successful
performance.
When baboons were given a relational matching-to-sample
task in which the sample and choice arrays contained 1 6 items,
they successfully learned to choose the 16-item choice array
instantiating the same relation as sample array. Accuracy was
high when sample arrays containing novel items were pre-
sented, thereby attesting to the generality of the relational
matching concept; but, when the number of items in the sam-
ple arrays was reduced from 16 to 12 to 8 to 4 to 2, baboons'
accuracy systematically fell to chance level. For the baboons,
the task proved too taxing when too little pictorial information
was available. Similar results have been reported with pigeons.
In contrast to baboons and pigeons, chimpanzees solve
relational matching-to-sample problems even when only
2 items are presented in the sample and the choice arrays.
The first chimpanzee to exhibit a variety of analogical beha-
viors was Sarah, who could evaluate, complete, and even create
analogies.
Sarah initially learned to use one plastic token for the
concept 'same' and another plastic token for the concept 'dif-
ferent.' In a set of later experiments, Sarah was given four
geometric forms on a display board in a 2 x 2 format. The
2 items on the left represented one relation and the 2 items
on the right represented another relation. Sarah had to choose
the correct plastic token (one for 'same' and one for 'different')
and place it in the middle of the board to indicate whether the
relations between the set on the left and the set on the right
were same (thus representing an analogy) or different. Sarah
chose correctly about 80% of the time.
In other experiments, Sarah was given 2 items on the left
side and only 1 item on the right side. The token for 'same'
was placed in the middle of the board and Sarah now had
to choose, from two alternatives, the item that completed
the analogy. Sarah chose the correct option most of the time.
She even did so when items representing functional relations
were presented. For example, when shown a lock and a key on
the left side, and a can on the right side, Sarah would choose
the can opener to complete the analogy.
Another set of studies explored whether Sarah could also
construct analogies. She was given an empty board and 4 or 5
items that could be used to create a valid analogy. This task was
especially challenging because Sarah had to find unspecified
relations among the items and arrange them on the board so
that they would represent a proper analogical relationship.
When only 4 items were available, Sarah created a valid anal-
ogy 76% of the times. Tier performance dropped when 5 items
were available, but it was still above chance level.
Premack and his colleagues have contended that language
training and/or prior experience with arbitrary symbols for the
Animal Cognition 151
abstract concepts of same and different are needed for animals
to exhibit analogical reasoning. Such training may have
allowed Sarah to display analogical abilities that had been
believed to be uniquely human. Language or symbol systems
may facilitate relational and analogical behavior because they
provide a way for animals to represent abstract relations so that
these relations can be encoded and manipulated.
However, the research described above with baboons and
pigeons suggests that language or symbol training may not be
necessary for disclosing this cognitive capacity. Perhaps criti-
cally, both baboons and pigeons had been trained to discrimi-
nate same from different collections of items before training on
the relational matching-to-sample task. That prior learning of
first-order relations may have provided the scaffolding
required to process second-order relations.
We conclude from all of this research that animals either
have a rudimentary capacity for analogical reasoning or they at
least possess the basic mechanisms that evolved into this
capacity. These observations have important evolutionary
implications. Higher-level cognition was once believed to be
the unique province of human beings; but, we now know that
chimpanzees, baboons, and pigeons show similar intellectual
abilities, at least in their basic form. The roots of abstract
thought may thus lie deep in our animal ancestry.
Metacognition
Demonstrations of numerical and basic mathematical abilities,
different types of memory, as well as abstraction and analogical
reasoning clearly document that animals possess a broad range
of cognitive abilities. But, do animals know what they know?
This question is not a tricky word play, but the core matter of
research in the growing field of metacognition. Metacognition
in humans is said to be associated with conscious awareness of
one's own cognitive states. People know whether they can
retrieve a specific memory, whether they can ascertain if they
have enough information to make a decision, and whether
they can assess the amount of knowledge they have about a
certain topic; in short, people can think about their own cog-
nitive states and processes.
In the last decade, several researchers have studied metacog-
nition in animals as well. Metacognition in animals is plausi-
ble. Imagine this common scene: you are walking through
a park and you encounter a woman walking her dog; the
dog sees you and then it starts looking back and forth to his
owner, as if deciding 'should I stay or should I go?' Or we see
a hesitant squirrel at the base of a wall apparently deciding
if the wall is low enough for her to jump up and land on it
safely. When animals do not know what to do, they might
defer their actions and seek help or information. These beha-
viors may be the result of a metacognitive process; but, do they
really require metacognition? Do animals have access to their
own cognitive states and can they use those states to control
their behavior?
First attempts to address animal metacognition used what
has been dubbed the uncertainty paradigm, in which animals
must learn to discriminate between categories of stimuli, for
example, between high-pitch and low-pitch sounds or between
pixel-dense and pixel-sparse visual images by choosing one
of two different responses for each of the categories. Animals
receive food reward for a correct response and a time-out period
for an incorrect response. When the stimuli are near the extreme
values, the task is easy; but, the task becomes increasingly diffi-
cult the closer values are to the middle of the continuum.
In addition to the two category responses, animals are also
given a third option - the uncertainty response - that avoids
the target discrimination altogether and takes the animal
to another easier task and a smaller amount of food than if
they had chosen the correct response for the training cate-
gories. If animals can monitor their knowledge, then they
might choose the uncertainty response when the values to be
discriminated are highly similar and failure is likely. And, so
they do. Dolphins and monkeys choose the two category
responses when the task is easy, but they choose the third
uncertainty response when the task is more difficult.
Nevertheless, the uncertainty response paradigm has raised
several concerns, because alternative explanations based on
simple associative learning can explain animals' apparently
metacognitive behavior. For example, animals may have
learned to select the uncertainty response for a particular
range of stimuli (the difficult ones near the middle of the
continuum) because of the reinforcement history with those
stimuli (animals are consistently rewarded if they choose the
uncertainty response, but they are inconsistently rewarded if
they choose the category responses), not because of a subjec-
tive feeling of uncertainty.
In order to avoid this problem, other paradigms have been
devised. As we saw earlier, animals have excellent memories for
rich and varied information; but, as in the case of humans,
these memories may fade or become difficult to retrieve over
time. One interesting possibility is to see if animals can report
their having good or poor memory for an event that happened
some time ago.
Hampton trained rhesus monkeys on a matching-to-
sample task in which a delay was introduced between offset
of the sample image and the testing stimuli, with the sample
presented along with three distractors. On some trials, an
intermediate choice was introduced at the end of the delay
interval which allowed the monkeys to either accept the mem-
ory test and receive a preferred reward if they were successful or
to decline the memory test and receive a guaranteed, but less
desirable reward (see method in Figure 6). On other trials, at
the end of the delay interval, only the option to take the test
was given, so that the monkeys had to take the memory test.
If monkeys have metamemory, then when given the option
to accept or to decline the test, they should accept the test
if their memory is strong, but they should decline the test if
their memory is weak.
As a consequence, the monkeys should be more accurate on
those trials in which they are given the choice. They should
accept the test on choice-test trials when they know that their
memory is good. But, the forced-test trials will also include
cases in which the monkeys' memory is poor, thereby, lower-
ing their overall accuracy. Monkeys' performance accorded
with this prediction; they were more accurate on trials in
which they accepted the test than on trials in which taking
test was the only option, suggesting that the monkeys could
distinguish between their different memory states. Similar
results have been found with rats.
152 Animal Cognition
Study
phase
Delay
interval
I
Delay
I
1/3 of trials 2/3 of trials
Choice
phase
Test phase
or small
reward
associations between the presented stimuli and the alleged
metacognitive responses.
Animals can certainly exhibit complex behaviors in these
metacognitive tasks; nevertheless, some researchers believe that
it is premature to conclude that those behaviors are the result
of access to and evaluation of internal cognitive states. Meta-
memory tasks, the gambling paradigm, and other studies do
suggest that animals can adaptively regulate their behavior
under conditions of uncertainty and respond in accord with
the knowledge they posses. But, that behavioral regulation may
be achieved by means other than metacognitive processes. For
example, animals may use their latency to respond as a cue for
subsequent behavior (humans also take into account their
speed of coming up with an answer to decide whether or not
they really know). Although metacognition is a plausible
mechanism for the animals' behavioral regulation, it is not
yet clear whether it is the only possible mechanism. Again,
Morgan's Canon comes into play.
Final Comments
Preferred
peanut
if correct
Primate
pellet
Figure 6 Method for assessing monkeys’ metamemory designed by
Hampton. Colored panels represent what the monkeys saw on the
computer monitor. At the start of each trial, monkeys studied a randomly
selected image. After a delay period, on two-thirds of trials, monkeys
could choose between taking a memory test (right, left-hand stimulus) or
declining the test (right, right-hand stimulus). On the remaining third
of the trials, monkeys were forced to take the test (left). From
Hampton RR (2001) Rhesus monkeys know when they remember.
Proceedings of the National Academy of Sciences 98: 5359-5362.
Work in the US public domain.
Monkeys' confidence in their memory has also been eval-
uated by allowing them to gamble. Monkeys viewed a series of
six pictures, one by one; after the last picture, nine pictures
simultaneously appeared on the screen, only one of which had
been presented in the prior series. The monkeys' task was to
select this picture. But now, before feedback was provided, the
monkeys were given a choice of 2 icons, representing a high-
risk option and a low-risk option. A high-risk choice resulted in
the gain of three tokens (that could be later exchanged for
food) if the monkeys' response in the picture memory test
had been correct, but a loss of three tokens if the monkeys'
response had been incorrect. A low-risk choice resulted in a
sure gain of one token.
The rationale was that a monkey showing metacognitive
capabilities should make a high-risk bet when confident about
its prior response, but it should make a low-risk bet when
unsure about its prior response. In fact, monkeys chose the
high-risk button more often on correct trials than on incorrect
trials, suggesting that they knew whether they had responded
correctly before the presentation of any feedback. Moreover,
monkeys generalized the use of the high- and low-risk options
to a variety of different perceptual discrimination and memory
tasks. This flexibility further helps to discount any specific
Considerable recent research documents many animals' ability
to remember the past, to respond effectively in the present, and
to plan for the future. Research also suggests that animals may
be able to take into account their current state of knowledge to
control their own behavior in an adaptive way. Finally, animals
can master numerical and abstract concepts, perform basic
arithmetic operations, and even exhibit behaviors which sug-
gest that they possess the roots of analogical reasoning.
Dumb beasts? Hardly! Animals of many different species
are sensitive to the rich mosaic of events and relationships that
are woven into the causal fabric of the environment. How
could it be otherwise? Animals evolved under most of the
same constraints and contingencies as the human species. To
study animal cognition is to study the mechanisms and func-
tions of cognition without the complexities of language or the
biases of anthropomorphism. Doing so not only enriches our
understanding of cognition in animals, but it also places
human cognition into a more complete evolutionary
perspective.
See also: Analogical Reasoning; Associative Learning; Memory;
Primate Cognition.
Further Reading
Bekoff M, Allen C, and Burghardt GM (eds.) (2002) The Cognitive Animal: Empirical
and Theoretical Perspectives on Animal Cognition. Cambridge, MA: MIT Press.
Beran MJ (2007) Rhesus monkeys (Macaca mulatta) enumerate sequentially presented
sets of items using analog numerical representations. Journal of Experimental
Psychology: Animat Behavior Processes 33: 42-54.
Beran MJ (2008) The evolutionary and developmental foundations of mathematics.
PLoS Biology 6: el 9.
Boakes R (1984) From Darwin to Behaviorism. Cambridge: Cambridge University Press.
Cantlon JF and Brannon EM (2006) Shared system for ordering small and large
numbers in monkeys and humans. Psychological Science 17: 401-406.
Cantlon JF and Brannon EM (2007) Basic math in monkeys and college students. PLoS
Biology 5(12): e328.
Animal Cognition 153
Clayton NS and Dickinson A (1998) Episodic-like memory during cache recovery by
scrub jays. Nature 395: 272-278.
Clayton NS and Russell J (2009) Looking for episodic cognition in animals and young
children: Prospects for a new minimalism. Neuropsychologia 47: 2330-2340.
Darwin C (1897) The Descent of Man, and Selection in Relation to Sex, 2nd edn.
New York: Appleton (Original work published 1871).
Descartes R (WO) Descartes's Philosophical Letters. Oxford, UK: Clarendon (Original
work published 1646).
Fagot J, Wasserman EA, and Young ME (2001) Discriminating the relation between
relations: The role of entropy in abstract conceptualization by baboons and humans.
Journal of Experimental Psychology: Animal Behavior Processes 27: 316-328.
Gillan DD, Premack D, and Woodruff G (1981) Reasoning in the chimpanzee: I.
Analogical reasoning. Journal of Experimental Psychology: Animal Behavior
Processes 7: 1-17.
Hampton RR (2001) Rhesus monkeys know when they remember. Proceedings of the
National Academy of Sciences 98: 5359-5362.
Hampton RR (2009) Multiple demonstrations of metacognition in nonhumans:
Converging evidence or multiple mechanisms? Special Issue on Metacognition.
Comparative Cognition & Behavior Reviews 4: 17-28.
Hauser M, MacNeilage P, and Ware M (1996) Numerical representations in primates.
Proceedings of the National Academy of Sciences 93: 1514-1517.
Huxley TH (1874) On the hypothesis that animals are automata, and its history.
Reprinted in Huxley TH (1898) Method and Results, Essays. New York, Appleton.
Inoue S and Matsuzawa T (2007) Working memory of numerals in chimpanzees.
Current Biology 17: R1 004-R1 005.
Kornell N, Son LK, and Terrace HS (2007) Transfer of metacognitive skills and hint
seeking in monkeys. Psychological Science 18: 64-71.
Locke J (1849) An Essay Concerning Human Understanding. Philadelphia: Kay &
Troutman (Original work published 1690).
Matsuzawa T (1985) Use of numbers by a chimpanzee. Nature 315: 57-59.
Matsuzawa T (2009) Symbolic representation of number in chimpanzees. Current
Opinion in Neurobiology if): 92-98.
Matsuzawa T, Tomonaga M, and Tanaka M (eds.) (2006) Cognitive Development in
Chimpanzees. Tokyo: Springer-Verlag.
Morgan CL (1896) An Introduction to Comparative Psychology. London: Walter Scott.
Oden DL, Thompson RKR, and Premack D (2001) Can an ape reason analogically?
Comprehension and production of analogical problems by Sarah, a Chimpanzee
(Pan troglodytes). In: Gentner D, Holyoak KJ, and Kokinov BN (eds.) Analogy:
Theory and Phenomena, pp. 472-497. Cambridge, MA: MIT.
Pearce JM (2008) Animal Learning and Cognition. New York: Psychology Press.
Premack D (1 983) The codes of man and beast. Behavioral and Brain Sciences
6: 125-137.
Raby CR, Alexis DM, Dickinson A, and Clayton NS (2007) Planning for the future by
western scrub-jays. Nature 445: 919-921.
Romanes GJ (1892) Animal Intelligence. New York: Appleton (Original work published
1882).
Shettleworth SJ (2010) Cognition, Evolution, and Behavior, 2nd edn. New York:
Oxford University Press.
Smith JD, Shields WE, Schull J, and Washburn DA (1997) The uncertain response in
humans and animals. Cognition 62: 75-97.
Spencer H (1855) The Principles of Psychology. London: Longman, Brown, Green and
Longmans.
Thomas RK (1980) Evolution of intelligence: An approach to its assessment. Brain,
Behavior, and Evolution 17: 454-472.
Wasserman EA and Young ME (2010) Same-different discrimination: The keel and
backbone of thought and reasoning. Journal of Experimental Psychology: Animal
Behavior Processes 36: 3-22.
Wasserman EA and Zentall TR (2006) Comparative Cognition: Experimental
Explorations of Animal Intelligence. New York: Oxford University Press.
Wright AA, Santiago HC, Sands SF, Kendrick DF, and Cook RG (1985) Memory
processing of serial lists by pigeons, monkeys, and people. Science 229: 287-289.
Zentall TR, Wasserman EA, Lazareva OF, Thompson RKR, and Rattermann MJ (2008)
Concept learning in animals. Comparative Cognition & Behavior Reviews
3: 13-45.
Antisocial and Narcissistic Personality Disorder
L Bobadilla, Western Carolina University, Cullowhee, NC, USA
J Taylor, Florida State University, Tallahassee, FL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Construct A working concept.
Nosology Branch of medicine dealing with the systematic
classification of diseases.
Operationalization The process of defining a variable or
variables into measurable factors.
Psychotic Group of psychiatric symptoms that include
hallucinations, delusions, and odd behavior that reflect an
impaired contact with reality.
Reliability The extent to which a test or criterion repeatedly
yields consistent results when assessed by independent raters
or in retesting.
Validity The extent to which a test or criterion measures or
predicts what it is supposed to.
Introduction
The fourth edition of the Diagnostic and Statistical Manual of
Mental Disorders - Text Revision (DSM-IV-TR) defines antisocial
personality disorder (ASPD) and narcissistic personality dis-
order (NPD) as persistent maladaptive personality styles that
cause distress to a person or interfere with his/her social and/or
occupational functioning. ASPD occurs in ~ 5% of men and
2% of women in the general population and is characterized
by a disregard for the rights of others often exemplified by
criminal behavior and a lack of remorse. NPD occurs in ~8%
of men and 5% of women in the general population and is
characterized by a grandiose sense of self and pathological self-
centeredness. Antisocial PD and NPD are included in Cluster B
in the DSM-IV-TR, which groups PDs according to shared
descriptive features. Cluster B is marked by erratic and dramatic
behavior and also includes borderline and histrionic PDs.
Antisocial and NPD often cooccur with one another reaching
20% and 50% comorbidity in clinical samples and impact not
just the individual with these conditions but also the people in
their social/occupational circle and the greater society through
associated aggressive and criminal behavior.
Conceptualizations of ASPD and NPD
The DSM-IV-TR characterizes ASPD as a 'pervasive disregard
and violation of the rights of others.' While this current descrip-
tion emphasizes a behavioral disposition toward breaking
society's rules, a discussion of ASPD cannot be complete with-
out addressing psychopathy, a clinical construct that emphasizes
not just antisocial behaviors but also interpersonal features
such as manipulativeness, glibness, and pathological lying as
well as affective features including lack of empathy and callous-
ness, lack of remorse, shallow affect, and failure to accept respon-
sibility. Although not equivalent to DSM-IV's ASPD definition,
the concept of psychopathy has influenced to varying degrees
the operationalization of ASPD throughout DSM history. At
times, the term psychopathy has been used interchangeably
with ASPD, but it has nonetheless been de-emphasized from
the official nosology. As a construct, psychopathy is reemerging
in psychopathology research and clinical applications and it is
likely to influence future DSM definitions of ASPD.
Descriptions of clinical phenomena that encompass mod-
ern conceptualizations of ASPD and psychopathy date from
the early nineteenth century when psychiatry pioneers in
Europe like Phillipe Pinel and J. C. Pritchard described patients
with a conspicuous absence of overt psychotic symptoms,
neuroses, or other 'mental defectiveness' who habitually vio-
lated societal rules displaying moral insanity (Pritchard) or
maniesans delire (insanity without delirium, Pinel). In addition
to the consistent antisocial behavior and apparent lack of
other psychiatric symptoms, these early descriptions remarked
on personality characteristics such as deceitfulness and lack
of guilt, which are part of the ASPD construct today. The
term psychopathic inferiority was first introduced by the German
psychiatrist Julius L. A. Koch (1891) to encompass 'character
disorders' (what we currently would consider PDs) and
emphasized their organic etiology. Unfortunately, this broad
application of the term 'psychopathic' expanded, and it often
applied to various conditions ranging from what were consid-
ered sexual deviations of various kinds (including homo-
sexuality) to substance abuse. Starting in 1904, Emil Kraeplin
refined the concept to include conditions characterized by poor
impulse control and severe antisocial behavior (sexual crim-
inals, fire-setters, swindlers). By 1915, Kraeplin had described
various 'psychopathic personalities,' which included the callous,
aggressive, and alienated features of modem conceptualizations
of ASPD. Nonetheless, despite refining by Kraeplin, the disorder
remained undefined and the term 'psychopathic' was used to
describe a multitude of unrelated conditions. The issue was
further complicated with the introduction of the term 'socio-
pathic' by Karl Birnbaum (1909) to emphasize the view that
criminality was mainly the product of social factors.
This confused state of affairs remained when in 1941
Hervey Cleckley published his influential monograph 'The
Mask of Sanity.' Based on his extensive clinical observa-
tions, Cleckley provided a series of illustrative case examples
along with 16 diagnostic criteria to operationalize the term
'psychopathy.' Cleckley's criteria described psychopathic
154
Antisocial and Narcissistic Personality Disorder 1 55
persons in terms of maladaptive behavioral attributes such as
unreliability, inadequately motivated antisocial behavior (e.g.,
robbery committed by someone with significant financial
support), as well as affective features including pathological
egocentricity, lack of remorse or shame, and untruthfulness.
Notably, Cleckley's criteria also included healthy mental and
interpersonal characteristics such as superficial charm, average
to high intelligence, lack of delusions or signs of irrational
thinking, and importantly, 'absence of nervousness or psy-
choneurotic manifestations.' Cleckley's book became (and
remains) a cornerstone for research in the field. Based on it,
David Lykken (1957) conducted a seminal psychophysiologi-
cal investigation supporting the hypothesis that psychopathic
behavior may result from impaired affective responses to fear
and anxiety. Lykken's findings have been widely replicated and
form the basis of his influential 'fearlessness' theory on the
etiology of psychopathy.
In 1952, the first edition of the DSM used the term 'socio-
pathic personality disturbance: antisocial reaction' to refer to
persons who habitually violated societal rules, were irresponsi-
ble, did not benefit from experience or punishment, were
frequently 'callous and hedonistic,' and rationalized their
misbehavior. However, under the same diagnostic rubric of
'sociopathic personality disturbance,' the diagnosis of 'dysso-
cial reaction' was also included to account for people who
engaged in antisocial behavior but were able to maintain
meaningful relationships with others. As could be expected,
the inclusion of two versions of the antisocial personality
diagnosis did not help clarify the classification of the syn-
drome. In response, DSM-II, in 1968, attempted to reduce
confusion by discarding the 'dyssocial reaction' category and,
in a marked departure from DSM-I, stressed that the diagnosis
should not be based solely on a history of offending. Also,
other personality characteristics including low frustration to
tolerance and selfishness (common to Cleckley's psychopathy
criteria) were added in the DSM-II and the name of the
disorder was changed to antisocial personality.
As the use of the DSM-II became widespread, it became
evident that a narrative description of disorders resulted in
poor reliability for some of them, including ASPD. To remedy
this, the development of DSM-III included clinical field trials
of the proposed diagnostic criteria for each disorder to deter-
mine their validity and reliability. The final criteria for ASPD
were heavily influenced by the work of Lee Robins (1966),
who conducted research on the development of 'sociopathy.'
Robins obtained archival clinical data of children and adoles-
cents who had been referred to a clinic for juvenile delinquents
from the early 1920s to 1930s and contacted 524 of them
30 years after their original referral. Although Robins initially
included some of Cleckley's symptoms (e.g., lack of guilt about
crimes), her criteria for the diagnosis of sociopathy heavily
favored a behavioral approach to classification (e.g., repeated
arrests, large numbers of sexual partners, substance abuse).
At around the same time that the DSM-III field trials were
being conducted, Robert Hare was developing a structured
interview assessment of psychopathy in prison samples based
on Cleckley's criteria. The result was the Psychopathy Checklist
(1980), which used a clinician rated interview and a review of
file information to rate interpersonal, affective, and behavioral
indices to arrive at a diagnosis. Since its emergence, the Hare
PCL and its revised version (PCL-R, 1991) have become the
gold standard of psychopathy assessment in forensic settings.
Despite its eventual importance, the first version of the PCL
was too late to influence the diagnostic criteria for ASPD in the
DSM-III, also published in 1980. In the DSM-III, the diagnosis
gained its current name (antisocial personality disorder), and it
again departed from its predecessor by exclusively relying on
behavioral criteria and specifically including disregard for social
norms and laws as a diagnostic criterion. To meet diagnosis, a
person would have to persistently display at least four out of
nine symptoms during adulthood and have a history of severe
conduct problems before the age of 15. These changes resulted
in increased interrater reliability, but many researchers pointed
out that it came at the cost of lowered validity. A person could
display an inability to sustain consistent work behavior and/or
function as a parent, have divorces, and be impulsive, but not
have any of the personality/affective traits of psychopathy
(callousness, lack of guilt), and still receive a diagnosis of
ASPD. To address these criticisms, the item 'lacks remorse'
was added in the DSM-III-R in 1987. However, criticism
continued because the added symptom was not necessary for
diagnosis and various researchers suggested that some ASPD
symptoms specified on the DSM-III-R (e.g., not maintaining a
monogamous relationship for more than a year) were not
good indicators of the traits underlying the syndrome.
The current diagnostic criteria for ASPD in the DSM-IV-TR
still have not resolved the identity of this disorder as one marked
more by behavioral antisocial acts or by personality features
reflected in the psychopathy construct. This controversy has
continued as the current field trials for the next edition of the
DSM are underway and is reportedly one of the reasons that
contributed to the recent decision by the American Psychiatric
Association to delay the publication of DSM-V until 2013.
In comparison to the development of the ASPD concept,
the evolution of the NPD construct has been less controversial.
Diagnoses related to pathological narcissism have a relatively
long history in psychiatry. In 1898, Havelock Ellis used the
term 'narcissism' to describe a sexual perversion in which
the self becomes the object of sexual desire. Throughout the
twentieth century, the concept evolved with pathological
egocentricity at its core and heavy influence from Sigmund
Freud's psychoanalytic theory. Various writers including
Sigmund Freud (1914), Ernest Jones (1955), and particularly
Heinz Kohut and Otto Kernberg through the 1960s and 1970s
contributed to the literature that speculated about the disor-
der's etiology, described its symptomatology, and proscribed
its treatment. Although there were some differences between
theorists (notably between Kohut and Kernberg) regarding
the etiology of the disorder, all conceptions of NPD were
remarkably similar in describing it as being characterized
by grandiosity that masked underlying feelings of inferiority
and fueled by an excessive need for admiration.
Although descriptions of the disorder date back to the late
nineteenth and early twentieth centuries, NPD was introduced
into the official psychiatric nosology in the DSM-III in 1980.
The original description of the disorder was very similar to the
current one and included grandiosity, feelings of uniqueness,
entitlement, exploitativeness, lack of empathy, and idealiza-
tion/devaluation of self and others. Since then, the criteria
for the diagnosis have undergone only a few changes.
1 56 Antisocial and Narcissistic Personality Disorder
Specifically, a symptom of preoccupation with envy was added
in DSM-III-R while the symptom of idealization/devaluation
of self and others was removed, and then, in DSM-IV, a symp-
tom tapping arrogance and haughtiness was added while the
symptom of feeling rage, shame, or humiliation in response to
criticism was removed. Although the symptom of anger in
response to criticism was ultimately removed from the diag-
nostic criteria, the text of the DSM-IV-TR explains that rage
and defiant counterattack in response to criticism are often an
associated feature of NPD. Importantly, aggression in response
to criticism has been well documented in clinical and laboratory
settings among NPD patients and nonclinical populations with
high narcissistic traits. Therefore, proneness to anger and aggres-
sion in response to perceived insult remains an important part
of the narcissism construct and it may be a contributor to
irritability and aggression, which characterize ASPD.
Although an initial comparison of the DSM-IV-TR's descrip-
tion of the essential features for NPD (grandiosity, need for
admiration, and lack of empathy) and ASPD (pervasive disre-
gard and violation of the rights of others) may not readily
reveal an obvious overlap, it can be argued that, someone
who engages in repeated deceit, assaults, and violations of
norms while showing no remorse (characteristic of ASPD) is
likely to be entitled, exploitative, unempathic, and arrogant
(characteristic of NPD). Support for this view comes from
various studies that have found a moderate to strong correla-
tion between ASPD and NPD. In fact, the diagnostic overlap
between the disorders was evident during the field trials to
develop the current DSM diagnostic criteria for ASPD. Affective
symptoms such as 'lacks empathy,' and 'inflated, arrogant self-
appraisal' (from a 10-item version of the PCL) provided incre-
mental validity for ASPD, but were not included in the official
list of diagnostic criteria due to concerns that it would make the
differentiation between ASPD and NPD difficult. These find-
ings suggest that personality aspects that are central to NPD
(and psychopathy) are an important aspect of ASPD and will
likely influence future diagnostic criteria for the disorder.
Comorbidity
ASPD, Psychopathy, and Overt Criminality
Both ASPD and psychopathy are associated with aspects of
criminality, including interpersonal violence, substance abuse
(both intake and illegal trade), nonviolent crimes, and recidi-
vism. However, the type of criminal behavior displayed by peo-
ple with ASPD versus those with psychopathy may be an
important factor that distinguishes these two syndromes. Stud-
ies with prison populations have found that up to 50-70% of
people in prison meet ASPD diagnostic criteria, while only 25-
30% meet psychopathy diagnostic criteria. However, within
3 years of release from prison, 80% of psychopathic offenders
violate the terms of parole compared to only 25% of their
nonpsychopathic counterparts. More worrisome are findings
that psychopathic offenders are four times more likely than
nonpsychopathic offenders to recidivate violently, and accord-
ing to an FBI report, almost half of persons who kill law enforce-
ment officers are psychopathic. These findings suggest that
psychopathic offenders are at higher risk for recidivism, in par-
ticular, for violent offenses.
The literature categorizes criminal acts as 'proactive' or
'instrumental' to indicate a level of planning and/or emotional
coldness (e.g., planned homicide to steal drugs from dealer)
versus 'reactive' to indicate impulsivity and higher affective
reactivity (e.g., assaulting a stranger after a perceived insult or
argument). Starting in adolescence, a trend emerges for psy-
chopathic individuals to engage in violent, planned criminal
behavior whereas persons with ASPD are more likely to engage
in unplanned, impulsive criminal behavior. This trend con-
tinues throughout adulthood and extends to a variety of crimes
ranging from petty crimes to sexual homicide. This is not to
imply that persons with ASPD do not engage in proactive
criminal behaviors or that psychopaths are not reactively hos-
tile. Rather, psychopathic individuals appear to have a proclivity
for violence and are more likely to have planned their actions
and/or not be emotionally aroused (e.g., angry) before commit-
ting a crime than people with ASPD. Notably, investigators
examining the relationship between facets of psychopathy and
criminality have found that interpersonal/affective traits of the
disorder appear to be uniquely associated with the instrumen-
tality and planning of crime, while the impulsive/social devi-
ance traits are not. On the other hand, the social deviance traits
of psychopathy are more strongly related to ASPD and play an
important role in the maintenance of antisocial behavior: they
predict violent and nonviolent recidivism and, among adults,
they predict violent sexual offense recidivism.
Other factors such as intelligence may also play a role in the
type of crimes in which people with ASPD or psychopaths
engage. Low scores on measures of verbal ability have been
associated with ASPD and the social deviance aspect of psy-
chopathy but not with the interpersonal affective traits of
psychopathy. Therefore, in the case of people with ASPD,
their lower cognitive ability and aggressive and impulsive
nature makes them likely to repeatedly engage in opportunistic,
nonplanned crimes, often 'in the heat of the moment' or under
the influence of substances. On the other hand, higher verbal
ability, superficial charm, and conning manipulativeness of
psychopathic individuals allow them to better plan crimes
and put them in close contact with victims they may not
know. It should be noted that crimes cannot always be clearly
divided into planned/instrumental or impulsive/reactive and
the majority of crimes have both elements. Research indicates
that while psychopaths may be more likely to commit proactive/
instrumental crimes, the majority of the crimes they commit
have been categorized as 'impulsively instrumental,' leading
some researchers to propose that psychopaths may be able to
restrain their violent impulses but are unwilling to do so.
NPD and Antisocial Behavior
As mentioned earlier, ASPD and NPD have features in com-
mon and it is therefore not surprising that narcissistic traits
and NPD have been associated with antisocial and aggressive
behavior. A large part of the evidence supporting a relationship
between narcissistic traits and aggression comes from social
and personality psychology laboratory studies showing that
highly narcissistic individuals (though not necessarily diag-
nosed with NPD) react aggressively toward people who they
think have criticized them. The same relationship between
narcissism and aggression has been observed in clinical settings
Antisocial and Narcissistic Personality Disorder 1 57
with NPD patients and persons with 'subclinical' NPD. For
example, studies examining domestic violence have reported
a strong association between NPD, high narcissistic traits, and
sexual coercion (among both genders), child abuse, and vio-
lence against women. Importantly, as the levels of narcissism
increase, so do the levels of violence. Therefore, the data appear
to show a pattern in which narcissistic traits and NPD are
clearly related to violence and as narcissistic traits are more
elevated (e.g., among psychopaths), the relationship to aggres-
sion is more robust and crosses the threshold from reactive to
include proactive or predatory aggression.
Comorbidity with Other PDs
Epidemiological data show that nearly two-thirds of people
with NPD also meet criteria for another PD at some point in
their lifetime. The highest comorbidity is found for other Clus-
ter B PDs, particularly ASPD and borderline PD. Men with
NPD are significantly more likely than women with NPD
to have ASPD and the converse is true for borderline PD.
Histrionic PD occurs in 10% of people with NPD regardless
of gender. The next most common lifetime comorbidity for
NPD is with Cluster A PDs, which are marked by odd or
eccentric behavior. Schizotypal and paranoid are the Cluster
A PDs that occur most frequently with NPD and neither
of them shows a gender difference in the comorbidity rate.
Finally, Cluster C PDs are characterized by anxiety and are
found at some point in the lifetime of almost one-quarter of
people with NPD regardless of gender. The comorbidity rate
with Cluster C PDs is driven primarily by NPD's association
with obsessive-compulsive PD. Antisocial PD is fairly similar
to NPD in its pattern of comorbidity with other PDs. Like NPD,
ASPD has its highest comorbidity with other Cluster B PDs, and
ASPD also has substantial comorbidity with paranoid PD from
Cluster A and obsessive-compulsive PD from Cluster C.
Comorbidity with Other Disorders
The rate of comorbid PDs among individuals with NPD is
rivaled by a similarly high rate of substance use disorders
(SUDs). One-half of those with NPD have an alcohol use
disorder and one-quarter have a drug use disorder within
their lifetime. Men with NPD show double the rate of alcohol
and drug use disorders as women with NPD mirroring the
gender difference found in the comorbidity of ASPD with
NPD. Mood disorders such as major depression and bipolar I
and anxiety disorders such as specific phobia and posttrau-
matic stress disorder are found in 20-25% of those with NPD
at some point in their lifetime. Gender differences in the
comorbidity rates for these mood and anxiety disorders are
similar to those for borderline PD, with women with NPD
showing greater rates of all but bipolar I as compared to men
with NPD.
An extensive literature documents the high comorbidity
between ASPD and SUDs, with some research showing that
those with ASPD have an 1 1-fold increase in risk for an alcohol
use disorder and a 14-fold increase in risk for a drug use disor-
der during their lifetime. There is a large literature on subtypes
of alcoholism that consistently identifies two subtypes, one
characterized by comorbid ASPD, more severe substance use
history, and higher likelihood of illicit drug use problems.
According to research examining the structure of mental dis-
orders, ASPD is part of an externalizing spectrum along with
SUDs and behavioral disorders such as attention-deficit hyper-
activity disorder. Antisocial PD's significant comorbidity with
SUDs and other behavioral disorders stems from their associa-
tion to a common factor labeled 'externalizing.' This externa-
lizing factor is proving useful in the search for etiological
influences (e.g., specific genes) that relate to multiple disorders
in the spectrum, including ASPD.
Relationship to Normal Personality Traits
Personality Trait Dimensions Associated with NPD and ASPD
Currently, there is a movement to conceptualize PDs as
extreme, aberrant configurations of otherwise normal person-
ality traits. Most personality trait models coalesce around a
hierarchical structure of three to five 'broad-band' personality
domains comprised of a number of lower-order personality
traits. One of the most widely used models is the five-factor
model (FFM) of personality. Briefly, the FFM is composed of
five broad-band factors: openness (intellectual curiosity vs.
conventionality); conscientiousness (orderliness/planning vs.
disinhibition); extraversion (outgoing/sociable vs. introverted/
quiet); agreeableness (trusting/empathic vs. arrogant/aggres-
sive); and neuroticism (emotionally stable/easy going vs. stress
reactive/prone to negative emotions). Various researchers have
mapped PDs along these dimensions in an effort to understand
the personality traits that contribute to the disorders.
In the case of NPD, studies using dimensional measures of
narcissism and the personality profiles of NPD patients have
found that the disorder appears to be characterized by low self-
reported agreeableness and neuroticism coupled with high
extraversion. More detailed examinations of components of
the narcissism construct such as leadership, exploitativeness,
arrogance, self-absorption, and entitlement are associated with
high levels of FFM facets of assertiveness/dominance of others
but low gregariousness/affiliation. Therefore, the FFM profile
of narcissistic individuals is consistent with the clinical picture
of a person who is aggressive and demands attention and/or
submission from others but without a desire to establish per-
sonal relationships with them. With regard to the relationship
between NPD, narcissism, and the broad-band dimension of
conscientiousness, findings have been mixed. However, recent
studies have found a positive relationship between the motiva-
tional dimension of sensitivity to reward and NPD, as well as
narcissistic traits. In addition, clinical cases show that persons
with NPD endorse high levels of reward driven traits such as
achievement striving, but they also score low in deliberation
(both facets are subsumed under conscientiousness). This
combination of motivational and personality traits contextua-
lizes the clinical picture of persons who might view themselves
as capable and diligent but at the same time are careless or
impulsive in their approach and indeed aggressive or impatient
when frustrated.
Interestingly, studies have shown that the aforemen-
tioned FFM features associated with NPD have some parallel
relationships to ASPD and psychopathy and could explain
their high rates of comorbidity. Specifically, low agreeableness
1 58 Antisocial and Narcissistic Personality Disorder
is common to the three disorders and may be the most
relevant factor influencing their characteristic aggressive-
ness, mistrust of others, arrogance, and lack of empathy. Low
agreeableness has also been found to account for much of
the covariance between the interpersonal/affective and the
social deviance facets of psychopathy. Antisocial PD and
the social deviance traits of psychopathy are positively asso-
ciated with neuroticism from the FFM, particularly the angry
hostility, impulsiveness, and vulnerability (to stress) facets.
On the other hand, the interpersonal/affective factors of psy-
chopathy have been found to be unrelated or negatively related
to neuroticism though positively related to extraversion partic-
ularly to the excitement-seeking and assertiveness facets.
Finally, various studies have found a strong and negative asso-
ciation between all facets of conscientiousness, ASPD, and
the social deviance aspects of psychopathy, completing the
clinical picture of people who are unscrupulous, aimless, dis-
organized, and careless. The interpersonal/affective traits of
psychopathy have also been negatively related to conscien-
tiousness; however, the association tends to be lower than the
one for psychopathy's social deviance traits and ASPD. This
attenuated relationship between conscientiousness and the
interpersonal/affective traits of psychopathy may be the result
of poor insight conferred by inflated self-esteem. Among the
facets subsumed under conscientiousness are competence (self-
competence vs. ineptness) and dutifulness (ethical/scrupulous
vs. lax/casual). Therefore, psychopaths' arrogance may lead
them to report being competent while at the same time endor-
sing a nonchalant attitude toward responsibilities and ethics.
NPD Subtypes, ASPD, and Psychopathy
The above description of NPD presents a paradoxical picture of
persons who may demand attention and recognition yet dis-
like being with others they deem inferior. Also, although some
aspects of narcissism (e.g., vanity, arrogance) are negatively
related to omnibus measures of emotional reactivity (neuroti-
cism), the entitled and exploitative aspects of the disorder are
positively related to it. In addition, as mentioned above, persons
with NPD may have high conscientious traits such as achieve-
ment striving and competence but low deliberation and there-
fore poor control of impulses including their well-documented
aggressive reactions. This 'narcissistic paradox' has been widely
noted in the clinical literature and has led some writers to
propose that there may be subtypes of narcissism: grandiose
(or overt) and vulnerable (or covert). Both subtypes of narcis-
sism have been related to undercontrol of aggressive and erotic
impulses, insistence on self-expression even at the expense
of others, and unrealistic and entitled expectations. However,
grandiose narcissism is positively related to personality traits
measuring social poise, self-assurance, and well-being whereas,
vulnerable narcissism is negatively related to these traits. Vul-
nerable narcissism has also been related to hypersensitivity
and fear of social scrutiny, whereas grandiose narcissism is
related to exhibitionism and attention seeking.
The duality of the narcissistic paradox is similar to the
one observed with psychopathy and ASPD. Psychopathy, par-
ticularly its interpersonal/affective traits, is associated with
high levels of assertiveness and excitement-seeking but low
levels of stress reactivity (similar to grandiose narcissists),
whereas ASPD and the social deviance factor of psychopathy
are associated with high levels of both excitement-seeking and
stress reactivity (similar to vulnerable narcissists). Given the
similarities in personality factors associated with NPD, ASPD,
and psychopathy, recent theoretical proposals have posited
that grandiose narcissism may be related to the interpersonal
aspects of psychopathy while vulnerable narcissism may be
related to the social deviance/ ASPD aspects of the disorder.
These proposals would be in line with earlier theories (e.g.,
Kernberg) suggesting that there is a pathological egocentric
core to psychopathy and ASPD. Though provocative, these
proposals require further exploration.
Etiological Influences: Genes and Environment
Twin studies are a primary methodology for estimating the
extent to which genetic and environmental factors are asso-
ciated with individual differences in a behavior or trait. The
comparison of monozygotic (MZ) twins who share all of their
genes to dizygotic (DZ) or fraternal twins who share, on aver-
age, half of their segregating genes allows for estimates of the
amount of variability in a behavior or trait that is associated
with genetic factors, shared environmental factors (things that
are common to individuals who live together and serve to
make them more alike), and nonshared environmental factors
(experiences that are not shared among relatives which serve
to make them different from one another). Research on the
etiology of NPD is lacking relative to the voluminous literature
on the etiology of ASPD. This is apparent when examining
research on genetic and environmental influence.
Only a handful of twin studies have examined NPD or
its dimensional representations. These studies based on a
Norwegian twin sample suggest that 60-80% of the individual
differences in NPD diagnoses or symptom counts are associated
with genetic factors. The remaining variance is accounted for by
nonshared environmental factors. Studies examining a dimen-
sional representation of narcissism using the Dimensional
Assessment of Personality Pathology in a large Canadian twin
sample have found a similar pattern as found for NPD but
with a more equal distribution of variance across genetic and
nonshared environmental factors. Like NPD, ASPD is influ-
enced by genetic and nonshared environmental factors, but
many twin studies also indicate a small influence of shared
environment on ASPD. Juvenile forms of antisocial behavior
such as conduct disorder and delinquency show larger estimates
of shared environmental influence that decreases in magnitude
as people age.
A natural question given the comorbidity between ASPD
and NPD is whether comorbidity arises from common genetic
and/or environmental factors. This question has been addressed
in two reports on a large Norwegian twin sample and the
results suggest that a common set of genetic factors contributes
to the association between NPD and ASPD (and histrionic and
borderline PD). The data are inconsistent on whether a com-
mon set of nonshared environmental factors contribute to
the disorders. Other research has shown ASPD to be part
of an externalizing spectrum of disorders that also includes
SUDs, attention-deficit hyperactivity disorder, and constraint
(a personality dimension reflecting risk taking, adherence to
Antisocial and Narcissistic Personality Disorder 159
traditional values, and behavioral control). The externalizing
spectrum has a common genetic etiology contributing to the
various disorders as well as some disorder-specific genetic fac-
tors. Future research will need to assess the extent to which
NPD fits within the externalizing spectrum.
Twin studies are useful in providing information on how
much of the variability in a behavior or trait is associated with
genetic or environmental factors, but they have not typically
provided information on the specific genes and environmental
factors involved. Recent developments in data analytic techni-
ques are beginning to spur research on the interplay between
genes and environment, but NPD has not been the focus of
that research. Nor has NPD received attention in molecular
genetic studies aimed at identifying specific genetic variants
associated with the disorder. A growing body of work is begin-
ning to show that there are gene x environment interactions at
work for antisocial behavior and ASPD. Adoption data, which
provide a direct test of shared environment, show that conduct
disorder symptoms are more frequent among adoptees whose
biological parent had ASPD and whose adoptive environment
was disadvantaged (e.g., presence of mental illness in adoptive
parent). Research has also identified peer group as an impor-
tant environmental factor in the development of persistent
antisocial behavior. Finally, numerous studies have examined
associations of genetic variants with ASPD and have found
some support for genes related to the dopamine neurotrans-
mitter system and to monoamine oxidase.
Treatment
The treatment of NPD has received much attention in the
psychoanalytic literature with case studies of difficult to treat
patients and outlines of psychoanalytic treatments for the dis-
order. Studies of clinical samples have shown that NPD is
associated with impairments in interpersonal relationships
and epidemiological studies have shown that impairments
appear to be worse for men than for women. In addition, the
presence of NPD can hinder treatment of other disorders such
as SUDs and depression. Despite the evidence that NPD is a
real phenomenon that impairs relationships and can nega-
tively impact clinical practices, there are very few published
studies of effective cognitive/behavioral treatments for NPD.
Given this lack of attention, it is not surprising that there are no
published reports on the empirical effectiveness of treatments
for NPD using well-controlled clinical trials.
Psychopathy and ASPD have long been considered difficult
to treat conditions. In fact, many PDs are difficult to treat
because the conditions are ego syntonic, which means the
person lacks insight into the fact that their personality is the
root of their distress or functional impairment. The research on
ASPD suggests that it might be more amenable to treatment
than was first thought, but that the type of treatment makes a
difference. Randomized control trials suggest that behavior-
based treatments can be effective for reducing related problems
such as substance use, whereas treatments that depend on the
formation of a strong therapeutic alliance are less effective
for people with ASPD. In the case of psychopathy, some
researchers have presented evidence that treatment may actu-
ally make the disorder worse. Other researchers are less
pessimistic and claim that, like ASPD, some of their behaviors
may be moderated while other researchers suggest that persons
with psychopathy may only be managed or 'contained' (i.e.,
imprisoned and/or closely supervised). There is agreement,
however, that the main reason for inconclusive or even contra-
dictory findings in the area is the dearth of rigorous studies.
In short, it is not clear that the symptoms of ASPD or psychop-
athy themselves are highly amenable to treatment of any kind
once the disorder is present, and this has led some to advocate
for early identification of severely antisocial youth as targets
for intervention.
Summary
Narcissistic and antisocial PDs are persistent, maladaptive per-
sonality styles that interfere with a person's functioning. The
DSM-IV-TR defines NPD as characterized by a grandiose sense
of self and pathological self-centeredness. Antisocial PD is char-
acterized by a disregard for the rights of others often exemplified
by criminal behavior and a lack of remorse. Narcissistic PD
occurs in ~8% of men and 5% of women while ASPD occurs
in ~5% of men and 2% of women in the general population.
Narcissistic PD and ASPD are included in Cluster B in the DSM-
IV-TR which groups PDs marked by erratic and dramatic behav-
ior. Narcissistic PD and ASPD are related to psychopathy, a
disorder that embodies a particularly virulent combination of
traits from both NPD and ASPD. Research has shown that NPD
and ASPD have similar relationships to many normal range
personality dimensions, which might help explain the high
comorbidity among the disorders. Both NPD and ASPD are
also comorbid with other disorders such as depression and
SUDs. The effects of NPD and ASPD extend beyond the indivi-
duals with the diagnoses to people in their social/occupational
circle and the greater society through associated aggressive and
criminal behavior. The causes of NPD and ASPD include both
genetic and environmental influences, and some evidence sug-
gests that the disorders may be underlied by common sets of
genes and environment. Like many PDs, both ASPD and NPD
are often difficult to treat because individuals with the disorders
often lack insight into their personality dysfunction.
See also: Behavior Genetics of Personality; Big Five Model and
Personality Disorders; Borderline Personality Disorder; Histrionic
Personality Disorder; Personality Disorders.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental
Disorders, 4th edn., text revision. Washington, DC: Author.
Cleckley HM (1 941 ) The Mask of Sanity: An Attempt to Reinterpret the So-Called
Psychopathic Personality. St. Louis: C. V. Mosby.
Cook DJ, Forth AE, and Hare RD (eds.) (1998) Psychopathy: Theory, Research and
Implications for Society. Dordrecth, The Netherlands: Kluwer.
Costa PT and Widiger TA (eds.) (2002) Personality Disorders and the Five-Factor
Model of Personality. Washington, DC: American Psychological Association.
Kendler KS, Aggen SH, Czajkowski N, et al. (2008) The structure of genetic and
environmental risk factors for DSM-IV personality disorders. Archives of General
Psychiatry 65: 1438-1446.
1 60 Antisocial and Narcissistic Personality Disorder
Kernberg 0T (1989) The narcissistic personality disorder and the differential
diagnosis of antisocial behavior. Psychiatric Clinics of North America
12:553-570.
Krueger RF, Hicks BM, Patrick CJ, Carlson SR, lacono WG, and McGue M (2002)
Etiologic connections among substance dependence, antisocial behavior, and
personality: Modeling the externalizing spectrum. Journal of Abnormal Psychology
111:411-424.
Livesley WJ (ed.) (2001) Handbook of Personality Disorders: Theory, Research, and
Treatment. New York: Guilford.
Lykken DT (1995) The Antisocial Personalities. Hillsdale, NJ: Erlbaum.
Patrick CJ (ed.) (2006) Handbook of Psychopathy. New York: Guilford.
Stinson FS, Dawson DA, Goldstein RB, et al. (2008) Prevalence, correlates, disability,
and comorbidity of DSM-IV narcissistic personality disorder: Results from the
Wave 2 National Epidemiologic Survey on Alcohol and Related Conditions.
Journal ot Clinical Psychiatry 69: 1 033-1 045.
Relevant Websites
www.hare.org - Robert Hare's website devoted to the study of Psychopathy.
http://www.nlm.nih.gov/medlineplus/personalitydisorders.html - National Institutes of
Health website on Personality Disorders.
Anxiety and Fear
D W McNeil and A M Vargovich, West Virginia University, Morgantown, WV, USA
B J Ries, Minnesota State University, Mankato, MN, USA
C LTurk, Washburn University, Topeka, KS, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Anxiety An emotional state characterized by verbal reports
of distress (e.g., apprehension, worry), physiological arousal
(e.g., increased heart rate), behavioral activation (e.g.,
avoidance), and/or disruption of cognitive processing (e.g.,
hyperawareness). Anxiety is associated with more cognitive
symptoms and less visceral activation, and cues for its
manifestation are more diffuse and changeable, relative to
fear, as the focus of anxiety is on objects/events that are more
physically removed in space and/or time.
Fear An emotional state involving verbal reports,
physiological arousal, overt behavior, and/or cognitive
disruption similar to that of anxiety. Fear, however, involves
greater mobilization for physical action, as the threat
typically is related to specific objects or situations that are
immediately present.
Panic Sudden onset of a discrete period of intense fear,
typically involving feelings of terror and often accompanied
by a sense of imminent doom. Physiological (e.g.,
hyperventilation) and/or cognitive (e.g., feelings of
unreality) symptoms are involved. Panic attacks can either
be associated with specific situations or are manifested with
no known situational trigger; they can be either expected or
unexpected, or situationally predisposed.
Phobia Extreme and persistent fear and/or anxiety of
specific object(s) or a situation(s) that is out of proportion
to any actual danger involved, typically provokes an
immediate response, is recognized by the person affected as
excessive, involves avoidance or considerable discomfort,
and results in interference with life activities or great
suffering.
What Is Anxiety and What Is Fear?
Use and Origins of the Words Fear and Anxiety
The importance of anxiety and fear in the human experience is
reflected by the rich vocabulary that describes these emotions.
A few of the many words used in everyday language to describe
fear and anxiety include agitation, apprehension, concern,
dread, edginess, fright, horror, jitters, nervousness, restless-
ness, scare, tension, terror, uneasiness, wariness, and worry.
Historically, the word fear evolved from the Old English word
faer, which meant peril or calamity but later came to mean a
feeling of uneasiness caused by possible danger. The word
anxiety originated from the Latin word anxius, which means
troubled in mind, solicitous, or uneasy.
While the terms anxiety and fear are used interchangeably in
much of the literature, recent theorizing suggests important
differences between them. Fear, being an evolved basic emo-
tion, typically is in response to a more immediate threat, and
so is more highly associated with psychophysiological activita-
tion. Anxiety, on the other hand, primarily is a fear reaction to
distal or non-present threat, and so is more cognitive in nature.
Fear and Anxiety in Discrete and Dimensional Models of
Emotion
Discrete models of emotion propose the existence of a finite
number of basic, fundamentally different emotions. Izard, for
example, proposes ten fundamental emotions including anger,
fear, sadness/distress, contempt, disgust, guilt, shame/shyness,
joy, interest/excitement, and surprise. These basic emotions are
considered innate and are thought to motivate adaptive action.
For example, in fear, the corresponding 'flight' or 'fight' action
tendencies help motivate the organism to avoid or escape or
fight to survive a dangerous situation. Individual emotions
may combine with one another to produce a large variety of
affective states. Fear typically is regarded as a basic emotion
while anxiety is viewed as a blend of emotions.
Dimensional models generally assert that all emotions can
be classified in terms of their relationship to two or three
major orthogonal, bipolar dimensions of affect. The two
dimensions proposed by Watson and Tellegen are positive af-
fect (feeling active, elated, vs. dull, sleepy) and negative affect
(feeling distressed, nervous vs. calm, relaxed). These dimensions
are usually depicted visually as two axes intersecting at a 90°
angle within a circumplex. In Watson and Tellegen's model,
anxiety is associated with heightened negative affect.
Lang's bioinformational model proposes three rather than
two orthogonal, bipolar dimensions: (1) a valence dimen-
sion, pleasantness-unpleasantness (feeling happy, content
vs. afraid, sad), (2) an activity dimension, arousal-calm (feel-
ing excited, tense vs. relaxed, sleepy), and (3) a control
dimension, dominance-submissiveness (feeling in control,
dominant vs. submissive, guided). These dimensions are
visualized as three axes intersecting at 90° angles in three-
dimensional space. According to this model, anxiety involves
the experience of unpleasantness, arousal, and lack of control.
Components of Anxiety and Fear
Some psychological perspectives have regarded fear and anxiety
as amorphous entities inside an individual. Critics have labeled
these as lump theories. The lump perspectives have been criti-
cized as being unscientific and nonconducive to empirical veri-
fication. More recently, fear and anxiety have been regarded as
161
1 62 Anxiety and Fear
comprised of verbal reports of distress (with underlying cogni-
tive activity), physiological arousal, and overt behaviors. These
three systems are fear and anxiety themselves and comprise
their structure.
In many cases, verbal report is the most salient of anxiety/
fear responses. The verbal report component is quite compli-
cated in that people can not only describe private, internal
states, such as feelings and thoughts, but can also report on
their overt behaviors and physiological responses. Accuracy
and reliability of reporting are an issue, due to the effects of
social desirability and acquiescence, reactive effects of mea-
surement, bias, and faking.
Behaviors associated with fear and anxiety are often motoric
and observable. In fact, actual avoidance of an object or situa-
tion has long been a hallmark of phobia. Escape is a related
behavior, but instead of avoiding the object or situation alto-
gether, the individual confronts it, but leaves prematurely.
While avoidance and escape behaviors are the most classic
overt behaviors relevant to fear and anxiety, Marks has identi-
fied a total of four types of overt behaviors associated with fear
and anxiety:
1. withdrawal (avoidance, escape),
2. immobility (freezing, unresponsiveness),
3. submission (vulnerable posturing, appeasement),
4. aggression (verbal attack, threats).
Immobility often is seen in the animal kingdom as an
adaptive response to threatening situations, such as when pre-
dators are nearby and neither escape nor aggression would be
prudent. Immobility can either be attentive, in which an ani-
mal freezes, or tonic, in which an organism is unresponsive,
perhaps appearing dead. Immobility may be less common in
human beings, but can be seen in extreme reactions to major
stressors. For example, victims of brutal attacks have reported
reacting by being paralyzed and unable to respond during the
event. Individuals with height anxiety typically report feeling
'frozen' in place, which is adaptive to their fear as they are less
likely to fall.
In submission, organisms attempt to deflect attacks on
themselves (or their young) or try to appease a threatening
organism. In human beings, appeasement can be observed in
a variety of social encounters, such as in situations in which
individuals with less status attempt to placate higher status
persons by avoiding eye contact or making self-deprecating
remarks.
Aggression is less commonly associated with fear or anxiety.
Nevertheless, fight is one option in the fight or flight reaction.
Sometimes, colorful aggressive displays, like the threatening
rustling of plumage in some peacocks are involved in such
reactions. Responding to threat with fear- or anxiety-associated
aggression can be adaptive, discouraging attack by others.
Anxiety and fear are associated with changes in various
bodily organs and systems. Of greatest interest in this area
has been the nervous system, particularly the autonomic ner-
vous system. The parasympathetic and sympathetic nervous
systems (divisions of the autonomic nervous system) are asso-
ciated with changes in physiological function related to anxiety
and fear.
In fear/ anxiety responses, cardiovascular and other systems
are activated. The heart rate is increased. Blood flow to the
extremities is decreased because of the constriction of periph-
eral blood vessels. In order to aid in defense, blood pressure
increases to allow blood to be directed to the large skeletal
muscles. Blood is stored in the torso so that it is readily avail-
able for vital organs. Skeletal muscles prepare to contract and
so generate electrical activity. Muscles of the face change, which
may be evident in facial expression. Sweat gland activity
increases to prepare the organism to cool itself after exertion.
The skin surface becomes cooler because of decreased blood
flow to the skin. Breathing becomes more rapid to supply the
blood with increased oxygen. An increase in the release of
glucose into the bloodstream prepares the muscles and organs
for response. Pupils dilate and hearing becomes more acute.
This mobilization of the organism for action is not the only
aspect of fear. A unique diphasic response, specific to blood,
injury, and/or illness stimuli, ultimately decreases heart rate
and blood pressure, sometimes causing the organism to faint.
This response is usually triggered by the sight of injections,
blood, or injury. Blood flow is reduced, presumably to lessen
the danger of shock if an injury were actually to occur, thereby
rendering an organism more likely to survive.
Relationship of Anxiety and Fear to Courage
Courage is persistence in the face of fear and anxiety. To carry
on with behavior despite both responding physiologically and
acknowledging fear or anxiety verbally is to act courageously.
For example, rescuing a child from a burning building would
constitute courageous behavior for a passerby. It is also coura-
geous behavior, however, for someone with an intense fear of
public speaking to give a speech to an audience. Repeated and
successful practice of courageous behavior leads to a decrease
in verbal reports of fear and psychophysiological responsivity,
which can lead to a state of fearlessness.
Individuals 'deficient' in fear and anxiety are not necessarily
seeking treatment, but can suffer negative consequences. Marks
and Nesse describe this phenomenon as 'hypophobia,' and
explain that it is more beneficial to be appropriately cautious
than to take risks. When a potentially harmful situation is
encountered, anxiety can be beneficial and used in a way to
warn and prepare for the possible threat. Hypophobic indivi-
duals may not experience as much fear and anxiety as the
average person, which means they are at a higher risk of
engaging in risky and possibly injurious behavior.
Measuring Fear and Anxiety
Assessing fear and anxiety can follow the three systems model,
in which the focus of measurement is what a person says (e.g.,
expressing oneself in speech or writing by saying 'I'm afraid'),
what a person does (e.g., avoiding a feared situation), and how
their body responds (e.g., muscle tension). Interestingly, these
systems appear to operate somewhat independently of one
another. This phenomenon is known as desynchrony or discor-
dance; anxiety or fear is present in one or more systems, but to a
lesser or greater degree in others. For example, some people
appear outwardly calm while making a speech and do not
avoid or escape, but their hearts pound and they later report
having been terrified. People who are highly fearful, however,
typically show consistency in fear across the three systems.
Anxiety and Fear 1 63
Anxiety can be measured in response to a specific situation
at a discrete time, or can be assessed as a more general response
tendency. Spielberger has characterized anxiety as being either
state or trait in nature. State anxiety is a transitory emotional
condition characterized by feelings of tension and apprehen-
sion which fluctuate over time as a function of environmental
stressors and the individual's coping response to them. Trait
anxiety is a relatively stable personality attribute that is an
individual difference in proneness to experience stressors as
distressing.
In assessment, the question arises as to which of the three
response systems is the 'gold standard.' No one system has
precedence over the others, nor is any one considered the
most important, although self-reporting is the most commonly
used approach, as a matter of convenience. Nevertheless, ide-
ally the use of methods to assess each of the categories of self-
report, psychophysiological response, and overt behavior is
applied for the most comprehensive approach.
Self-reports are what people indicate through spoken, writ-
ten, or other symbolic communication. Methodologies such as
questionnaires and interviews are commonly used self-report
assessment strategies. Questionnaires that directly prompt the
individual to indicate the presence and intensity of anxiety/fear
typically use numerical ratings. Some nonverbal assessment
methods involve presentation of symbolic stimuli and require
an individual to select stimuli that match internal feeling
states. For example, the Self-Assessment Mannikin prompts
individuals to rate dimensions using an abstract stick figure,
changing characteristics of the mouth (for smiling or frown-
ing), bodily activity (to indicate arousal), and body size (to
report on feelings of dominance).
Another self-report methodology, interviews, can be used to
collect information about anxiety and fear and can have the
advantage of allowing a trained interviewer to use judgment in
assigning ratings to compensate for interviewee biases such as
over-reporting or minimizing. The interview format can be
unstructured and general, such as open-ended interviews con-
ducted by most mental health professionals in purely clinical
settings. Structured interview protocols can be utilized to assess
broadly (e.g., Anxiety Disorders Interview Schedule) or in
highly specific areas (e.g., Dental Fear Interview).
Overt motoric behaviors are typically evaluated by observa-
tion. Observation of behavior can take place in the clinic,
laboratory, or natural environment. Self-monitoring is often
used with clinical patients, asking them to record event fre-
quency, context, and associated thoughts and feelings (e.g.,
recording when panic attacks occur and in what situations).
Monitoring can also be done by parents, partners, and others.
In clinical settings, role plays can be used to emulate actual
situations so that behavioral responses of patients can be
observed.
It is possible to utilize trained personnel for standardized
observations, such as monitoring of facial expressions or as-
sessment of social skills. A standardized strategy for observing
behavior in clinical and research settings is the behavioral
assessment test (BAT). In this method, the individual being
assessed is asked to confront a fear/ anxiety-eliciting stimulus in
a standardized format (e.g., how long a person can carry on a
speech). It is also possible to observe more subtle behaviors
(e.g., stammering during a speech), which can be quantified.
Typically, measures involve approach (e.g., how close a person
can get to a snake in a BAT for snake phobia) or endurance
(e.g., how long an individual can stay seated in a dental chair
in a BAT for dental phobia).
Psychophysiological data most often are collected in the
laboratory or clinic, but can be acquired in the natural envi-
ronment using ambulatory monitoring equipment. Typically,
psychophysiological levels are measured at resting baselines
and then compared to precisely timed events that provoke
fear and/or anxiety. The most frequently utilized measures of
physiological arousal in fear and anxiety research are cardio-
vascular. Of these, heart rate has been the most often selected
index. Another commonly used physiological measure is elec-
tromyography, the assessment of skeletal muscle activity. By
measuring electrical activity from a muscle, it is possible to
infer muscular tension. Assessment of electrodermal activity
is yet another type of psychophysiological measurement; it
focuses on sweat gland activity in the skin, using measures of
skin conductance and resistance. An increase in perspiration
will be reflected in a decrease in resistance and an increase in
conductance. Other ways of measuring psychophysiological
fear and anxiety reactions include endocrinological measures.
Research has examined changes in catecholamine levels before,
during, and after stress in both normal and anxious subjects.
Endocrinological studies have also investigated changes in
epinephrine, norepinephrine, and adrenocortical functioning.
Across the three systems, specialized instruments allow eval-
uation of aspects of anxiety and fear that would ordinarily be
impossible to measure accurately or at all. Instruments are com-
monly used to measure psychophysiological responses as well as
to record speed and accuracy of cognitive processing. Moreover,
brain imaging techniques that allow measurement of very basic
physiological processes are available now. Novel assessment
methods have been introduced to accurately record cognitive
processes. These techniques can be used to assess responses to
stimuli that are within the consciousness (supraliminal) or out-
side the awareness (subliminal). Assessing cognitive processes
in fear/anxiety research has focused on attention and memory,
assessing reaction time, processing time for complex tasks, accu-
racy of recall and recognition, comprehension of ambiguous
stimuli, and effects of interfering probe stimuli, among others.
What Makes Human Beings Anxious and/or Fearful?
Anxiety and Fear as Normal, Adaptive Emotional States
Fear and anxiety are normal responses in everyday life that are
an important part of being human. Fear and anxiety can be
adaptive reactions to realistic threats or to the possibility of
some negative event(s). In dangerous situations, fear can
mobilize people to respond rapidly and efficiently. Without
this capacity, human beings would not have been able to
survive when faced by the dangers present in the natural envi-
ronment. Cannon's emergency action tendencies of 'fight'
or 'flight' represent the organism's alarm reaction to life-
threatening emergencies. With regard to anxiety, the ability to
anticipate possible negative futures and to prepare to deal with
them is adaptive as well.
Fear and anxiety continue to have adaptive value in every-
day life. Without some minimal amount of fear and anxiety,
1 64 Anxiety and Fear
people might become careless and lack the motivation to
accomplish things that need to be done. People might drive
recklessly on the highway, approach dangerous animals, neglect
work, or not plan for the future. According to the Yerkes-
Dodson law, human and animal performance generally im-
proves with increases in arousal until some optimal level is
reached, at which point further increases in arousal hinder per-
formance. With too little arousal, a pilot may become inattentive
and make a critical mistake; with too much activation, the pilot
may become paralyzed with fear and unable to respond effec-
tively to danger.
People do not always avoid situations which provoke fear
and anxiety and at times may actively seek out dangerous
activities. Many sky divers, mountain climbers, and race car
drivers enjoy the challenge of mastering danger. Most people
enjoy some sort of danger vicariously such as by viewing horror
movies, or reading suspense novels. Some individuals also
enjoy the fear involved in riding roller coasters and going to
Halloween 'haunted houses.'
Pathological Fears and Anxieties
While a bit of fear and anxiety can keep people alert and ready
for action, excessive amounts are disruptive and even paralyz-
ing. It is important to clarify the relationship of normal fears
and anxieties to pathological states, diagnostically known as
anxiety disorders. The immediate context is particularly impor-
tant with fear, as when there is danger present, even high levels
of fear may be appropriately activating, to remove the individ-
ual from threat. Without such danger, however, higher degrees
of fear likely interfere with functioning. Similarly, some lower
levels of transient anxiety may be a cue for individuals to plan
ahead and overcome procrastination, while excess worrying
probably interferes with performance.
As illustrated in Figure 1, fears and anxieties exist along a
continuum of severity, ranging from fearlessness at one end to
typical levels of anxiety and fear to their pathological manifes-
tations in anxiety disorders at the other end. Inherent in this
model is the idea that normal anxieties and fears are continu-
ous with psychopathological states. The anxiety disorders are
not disease states completely unlike that found in 'normal'
individuals. They are, however, quite extreme and persistent
versions of emotional states experienced by almost all people.
Anxiety and fear are considered pathological when they are
out of proportion to the actual threat present, interfere with
one's daily activities, involve maladaptive behavior such as
avoidance or escape, and are recognized by the individual as
excessive or unreasonable. The prevalence of anxiety disorders
surpasses that of any other category of mental health problems
in the general population. Additionally, substance abuse, espe-
cially alcohol abuse, is common among individuals suffering
from anxiety disorders and sometimes reflects an attempt at
self-medication.
Situations Eliciting Anxiety and Fear
A considerable amount of research has been directed at identi-
fying the objects and situations that evoke fear in human
beings. New fears and anxieties emerge or are lessened period-
ically. Only in the last several decades have there been concerns
about terror attacks and AIDS; concerns about contracting
certain diseases which were common prior to and early in the
20th century are considerably less now due to availability of
certain medications. The specific focus of fear and anxiety are,
in part, culturally and otherwise environmentally determined.
It may be more common to fear violence, for example, in high-
crime areas in certain cities and in war-torn lands, given the
greater vulnerability and probability associated with attack.
Nevertheless, work by Arrindell and others has made it possi-
ble to categorize anxieties and fears of adults into the following
five basic dimensions:
1 . social situations,
2. environmental and/or agoraphobic situations,
3. blood, bodily injury, illness, and death,
4. sexual and aggressive scenes,
5. small animals.
Fear and Anxiety Across the Lifespan
While fears and anxieties can develop at any time of life, some
have greater incidence in particular developmental stages. Fears
identified as being common in infancy typically include fear
of heights, loss of physical support, loud noises, and other
sudden, intense, unexpected stimuli from the environment.
A fear of strangers typically develops between 6 and 12 months.
Stranger fear appears to be different from distress associated
with separation from parents. Fear of strangers can emerge
as early as 4 months and peaks between 13 and 18 months.
A program of research by Kagan and colleagues on tempera-
ment evident in infancy suggests that some human beings are
constitutionally predisposed to being anxious and/or fearful,
inhibited, and physiologically reactive, which may have long-
lasting implications.
Common fears of children 1-3 years old include fear of toilet
training activities, injury, loud noises, animals, dark rooms,
masks, and novel stimuli. During preschool, kindergarten, and
Little or no Extreme
anxiety/fear anxiety/fear
Fearlessness
Figure 1 Continuum of severity for anxiety/fear.
‘Normal’
anxiety/fear
Phobia or other
anxiety disorder
Anxiety and Fear 1 65
first grade, prominent fears include fear of separation from
parents, animals, darkness, bodily harm, thunder, lightning,
sleeping or staying alone, and supernatural beings. Older ele-
mentary school children frequently experience fears and anxi-
eties related to test taking grades, physical appearance, injury,
illness, death, and natural phenomena such as thunder and
tornadoes. Most fears and anxieties of adolescents are focused
on personal adequacy, school events, physical health, sexual
matters, and political and economic concerns. In general, over
the course of childhood, fears and anxieties change from having
a formless and imaginary quality to being specific and reality-
based.
Many fears and anxieties are common in the general popu-
lation. The five most common fears among adults, in devel-
oped Western societies, in order of prevalence, are:
1 . snakes (most common),
2. heights,
3. storms,
4. flying in an airplane,
5. dental treatment.
What human beings fear and find anxiety-provoking
changes across the lifespan. During the course of development,
virtually all children display a large number of fears (e.g., of
injections, doctors, darkness, strangers), most of which are
transitory or short-lived. Fears of snakes, animals, heights,
storms, enclosures, and social situations are frequently ac-
quired during childhood and adolescence, but appear to be
more long lasting, as their prevalence remains relatively high
throughout adulthood and into old age. Fears which reach
their peak of prevalence at the upper end of the lifespan
include fear of crowds, death, injury, illness, and separation.
Fears and anxieties in older adults are complicated by frequent
coexistence with other disorders such as depression and
dementia. Symptoms of anxiety, including worry, are similar
in younger and older adults, although there may be some
unique considerations in the elderly, such as the greater like-
lihood of medical illnesses. While some anxiety disorders
decrease in prevalence in older adults, phobic disorders and
generalized anxiety disorder are relatively frequently
manifested.
Sex and Gender Differences
Studies consistently have shown that, in most domains (with
social fears a notable exception), men generally report less fear
and anxiety, while women report more such distress. This sex
difference appears during childhood, and is seen throughout
life. What males and females fear and find anxiety-provoking
also seem to differ. Women express more environmental and
agoraphobic fears, distress about small animals, and concern
about sexual and aggressive scenes than men. Fear of social
situations, as well as fear of bodily injury, death, and illness
appears to be more equivalent. During childhood, girls report
more fears and anxieties related to animals and physical injury
and illness, while boys report more fears and anxieties related
to economic and academic failure.
Variables other than biological sex may account for many of
the differences observed between females and males regarding
the prevalence of fears, anxieties, and anxiety disorders. While
there is more generalized anxious response in women relative
to men, with females reporting higher and males reporting
lower levels of anxiety, the response of females may be health-
ier for them, consistent with the literature on the benefits of
emotional expression. Craske suggests sex differences in anxi-
ety may be related to four main issues: gender role expectations
(females allowed greater social latitude to report fear), life
events (females have more negative events than males), physi-
ology (males more physically reactive to acute stress, females
more reactive to contextual cues), and cognition (females more
likely to worry). Wilson and Daly contest that males are socio-
biologically developed to be competitive risk takers, especially
when there is a good chance of success. This predisposition to
risk taking may result in either less fear and anxiety, or perhaps
less reporting of fear and anxiety.
Cultural Considerations
Fear and anxiety are emotional states that are experienced by
people across the world, although their environmental con-
tent, context, and manifestation vary widely. Traditional Abori-
ginals in Australia, for example, prefer open areas that are not
closed in, and some may react quite negatively to being shut in
a room. This reaction may be related to thousands of years of
nomadic existence, in which closed in shelters were uncom-
mon. As another example, Chinese males in Southeast Asia
may suffer from koro, the fear of the penis retracting into the
abdomen and death occurring as a result. This phenomenon
results in acute panic and various devices may be employed to
prevent retraction of the penis. Occasionally this disorder is
seen among women as well; they fear retraction of the nipples,
the breasts, or the labia. In this culture, male genitals are
considered to be vital for life and excessive sexual activity is
believed to be unhealthy. Therefore, it is not surprising to find
that patients who experience koro may relate the cause to past
sexual transgression.
Fright-related disorders are interesting phenomena that
have been observed in many different cultures. In Iranian
society, a variety of problems ranging from physical illness to
psychotic episodes may be explained as being caused by a prior
startle or fright experienced by the individual. Similarly, in
some Hispanic cultures, susto is a condition that is caused by
a sudden fright, the evil eye, or the casting of a spell and is
characterized by a state of anxiety, fearfulness, or illness. The
result of these beliefs may lead to what is termed Voodoo Death.
The victim of the curse or evil eye can be in a state of autonomic
hyperarousal for an extended period of time, which presum-
ably may lead to damage of the internal organs and eventually
death.
While these culture-specific conditions differ across the
globe, the prevalence rates of many classic anxiety disorders
are relatively similar. Nevertheless, there are differences be-
tween ethnic and racial groups within nations and geographic
areas. For example, there is a higher rate of phobic disorders
among African Americans than Caucasians in the United
States. This distinction may be due to the stress of being in a
minority group that has experienced prejudice and which has
higher proportions of low socioeconomic status relative to the
general population, placing them at higher risk for stress and
negative life events.
1 66 Anxiety and Fear
Comparisons Across Species
Fear and anxiety responses have been instrumental in the sur-
vival of both human beings and other species throughout the
course of evolution. Humans and animals have often experi-
enced similar hazards to their existence and, through natural
selection, have come to share many similar, adaptive fears. For
example, human infants, like the young of other land-dwelling
species such as cats, goats, and monkeys, display an adaptive
fear of heights and withdraw from a visual cliff. Aquatic species
such as ducks show no such fear. Similarly, novel objects, foods,
and situations elicit adaptive approach-avoidance behaviors in
humans and other species such as birds and primates; such
behaviors are beneficial to organisms in that the unfamiliar
carries with it the possibility of both danger and opportunity.
Other fears common to both humans and many other species
include fear of too much or too little light or space, fear of
sudden touch, seizure, proximity, or movement, and fear eli-
cited by visual and auditory alarm cues from members of
the same species.
Theories of the Development and Maintenance of
Anxiety and Fear
Origins of Angst
For centuries, philosophers and clinicians have pondered the
cause(s) of diffuse and objectless anxiety. Kierkegaard used
the German word angst to describe this experience. For Kierke-
gaard, angst results from the distinctively human, self-conscious
capacity to discern the possibility of freedom and the ultimate
threat of nonbeing. Kierkegaard thought that the achievement
of selfhood was possible only if one was able to recognize and
confront anxiety and to move ahead and actualize one's possi-
bilities despite it.
Psychodynamic Models
Early theories of anxiety and fear arose from psychodynamic
traditions. Freud used the word angst (anxiety) to refer to
vague, objectless anxiety and the word fur chi (fear) to label
such distress when it was focused on an object. Freud viewed
anxiety as the reaction to and signal of unconscious memories
of real or imagined threats to the child. Anxiety is related to
fantasized situations that were perceived as real and dangerous
to the child, who felt helpless in the face of them. The genera-
tion of anxiety acts as a signal to elicit the psychological
defense mechanisms, which, if adequate, reduce anxiety and
allow for a high level of functioning. If the defenses are not
adequate, phobic or compulsive symptoms symbolically re-
lated to the unconscious source of anxiety may arise.
Learning Theories
Although the capacity to experience fear is innate, most fears
and anxieties are learned through association. Fear and anxiety
may develop through classical conditioning. In this type of
learning, an unconditioned stimulus (UCS), such as pain, is
the original source of fear (unconditioned response). After
pairing the UCS with a new, previously neutral stimulus, this
second stimulus becomes the conditioned stimulus (CS). The
CS then can elicit a fear response (conditioned response) with-
out the UCS being present. Watson and Raynor demonstrated
this process in their case of little Albert. Albert, a young child,
initially showed no fear of rats. He was, however, frightened by
loud noises (UCS). Watson and Raynor repeatedly exposed
little Albert to a loud noise (UCS) while viewing a rat (CS).
Little Albert soon developed a conditioned fear to rats. This fear
generalized to other objects that resembled rats, such as rabbits
and other soft, furry animals.
An early adaptation to classical fear conditioning theory
was the avoidance learning model, which hypothesized that
fears fail to decrease or remit completely if the individual learns
to avoid the stimulus. Mowrer stated that the learned fear of
the CS was acquired according to classical conditioning prin-
ciples, and that the motor response of avoidance was learned
by instrumental conditioning. As it involves both classical
conditioning and instrumental conditioning, it is known as
the 'two-factor theory.' It purported to explain why phobias
do not resolve even with repeated exposure. Mowrer reasoned
that the motor response was first made to escape a negative
situation, but on subsequent trials, the behavior was an avoid-
ance response motivated by fear, and maintained by fear reduc-
tion. Despite the fact that the two-factor theory has not fared
well when subjected to rigorous empirical investigation, it has
made significant contributions toward the understanding of
the maintenance of fear and anxiety.
Other mechanisms have been introduced to resolve the
limitations of conditioning theories to account for the devel-
opment and maintenance of anxieties and fears.
Vicarious Learning Through Observation and Direct Instruction
Vicarious learning has been extensively studied, and undoubt-
edly can play a major role in the acquisition of anxiety and fear.
Such learning through observation, hearing accounts of fearful
situations, or someone telling another person that an event
(e.g., dental care) is scary can be linked to learned sensitivity
of bodily sensations (e.g., suffocation cues related to panic
attacks) and social phobias (e.g., socially anxious parents fear-
ing and avoiding social situations, thereby serving as a model
for their children's social behaviors). Although vicarious learn-
ing certainly does not always lead to phobic responses, or even
anxiety or fear, it may serve as a specific type of psychological
vulnerability for the development of anxiety disorders.
Cognitive-Behavioral and Cognitive Theories
Cognitive-behavioral and cognitive formulations of fear and
anxiety have grown in influence within the scientific literature.
Barlow regards panic as a critical component to anxiety disor-
ders, with biological diathesis and psychological vulnerability
potentially predisposing individuals to develop problems with
anxiety. Barlow's work suggests anxiety, fear, and panic may
be described as cued or uncued, expected or unexpected. Panic
is considered cued when the individual is able to identify an
external or internal event that precipitates it; panic is consid-
ered uncued when the individual is not able to identify a
trigger. If a panic attack is anticipated, then it is expected;
otherwise, it is unexpected. Thus, panic attacks fall into one
Anxiety and Fear 1 67
of four categories: cued and expected, cued and unexpected,
uncued and expected, uncued and unexpected. For example, a
panic attack is both cued and expected when a person with
a snake phobia encounters a snake. Spontaneous panic that
comes 'out of the blue' is uncued and unexpected. In this
conceptualization, fear is largely cued, while anxiety is uncued.
Research suggests there are three particular categories
of anxiety/fear-linked cognitive bias. Specifically, individuals
experiencing anxiety ( 1 ) attend selectively to threat-related infor-
mation, (2) show facilitated memory for threat-related informa-
tion, and (3) demonstrate an interpretive bias favoring more
threatening meanings of ambiguous information. In terms of
the content of thinking, anxiety has been found to include
negative self-talk, low self-efficacy, and self-abnegation.
Beck's cognitive theory depicts a three stage approach in
responding to anxiety and fear: (1) registration of the threat,
(2) activation of an automatic response to the threat, and (3)
the subsequent evocation of secondary, elaborative checking.
Further, this theory emphasizes that beliefs and cognitive pro-
cessing of errors play an important role in producing and
sustaining negative emotional states. For example, an anxious
individual may believe that 'others are critical' and subse-
quently feel anxious when meeting new people. This individ-
ual will be quick to attend to threatening information in a
social situation (e.g., frowns), will likely remember and rumi-
nate about the most negative aspects of the situation (e.g.,
mispronouncing someone's name), and interpret ambiguous
information negatively (e.g., someone ending a conversion
is an indication of dislike rather than some more neutral
explanation).
Metacognitive theory is a branch of cognitive psychology
which emphasizes thinking about the manner in which indi-
viduals' process cognition, or 'thinking about thinking.'
Instead of focusing on maladaptive beliefs, this theory places
importance on awareness and control over cognitive (and
emotional) processes, as well as overarching beliefs and apprai-
sals of specific situations. Therefore, as described by Wells, an
individual with generalized anxiety disorder may utilize vari-
ous types of worrying as a means of appraising a situation and
coping with it.
Network Models
As discussed earlier, Lang has forwarded a bioinformational
model of emotion, differentiating the dimensions of valence,
arousal, and control. In this theory, anxiety and other emotions
are represented in memory via stimulus characteristics (e.g.,
hearing a dental drill sound), response propositions (e.g., grip-
ping the arms of a dental chair), and meaning elements (e.g.,
being in a dental office for root canal therapy versus attending
for routine care). Lang's theory may best be considered in the
context of a defensive motivational system, in which the behav-
ioral reactions of anxiety and fear are reflected in defensive
immobility (freezing, hypervigilance response) and defensive
action (flight-or-fight response) . In Lang's recent research, there
has been a focus on these responses, specifically the phenome-
non of a startle potential. A noticeably stronger startle response
is the result of the activation of the defensive motivational
system, resulting from high arousal and a high level of
unpleasantness.
Evolutionary Significance and Biologically Prepared Stimuli
Encounters with environmental stressors that cause fear and
anxiety have been present throughout the history of human-
kind. Early encounters included contact with feared beasts that
have shaped humans' reactionary responses. It has been pro-
posed that animal fear originates in a predatory defense system
that allows animals, including humans, to avoid or escape
potentially dangerous animals such as poisonous snakes.
The evolutionary significance of fear and anxiety is directly
linked to the biological idea of survival of the species. Fear and
anxiety serve a protective function for all members of the
animal kingdom who are capable of experiencing this bio-
logical warning. Some theorists believe that the ability to
plan for the future is also dependent on anxiety and fear.
Seligman, Ohman, and others posit that humans, and some
lower animals, are prepared or predisposed to fear certain ani-
mals and other stimuli. Human beings more frequently fear
stimuli that can potentially cause harm (e.g., rodents). This
theory of evolutionary preparedness implies that some aspects
of what humans learn to fear, along with the fear response,
may be innate. For example, investigators have suggested that
humans may have an innate fear of small stimuli that move
rapidly or abruptly (e.g., snakes, spiders). It is somewhat less
frequent for people to fear things that have little or no apparent
danger, although quite unusual fears are observed clinically
(e.g., distress about opening large dictionaries).
This fear of evolutionarily relevant creatures (e.g., poison-
ous snakes and spiders) is observed even in individuals living
in geographic regions in which no such threat exists. Therefore,
even without the risk of encountering a dangerous animal, the
fear persists. In contrast, the likely harmful aspects of modern
life (e.g., overconsumption of alcohol, tobacco use, excess
dietary fat, obesity, casual unprotected sex), are typically not
feared, even though they are encountered more often and are
more of an actual threat to individual well being.
Genetic and Familial Transmission
A variety of studies have asked, 'Are fear and anxiety inherited?'
and 'Are fear and anxiety acquired through learning?' These
questions are an example of the 'nature versus nurture' debate
that permeates much of psychology. Researchers have looked
at family histories, parenting characteristics, and prevalence
rates of anxiety disorders within families to answer these ques-
tions. Comparative psychologists have demonstrated that it
is possible to breed anxiety/fear in rats and dogs and have
observed severe social anxiety to be naturally occurring in a
subset of certain nonhuman primates. When exploring the
heritability of the global category of anxiety neurosis, one
study noted that as many as 50% of the children with parents
with an anxiety neurosis also carried that diagnosis. Studies
with twins suggest a genetic component in transmission of
anxiety disorders. Studying the connection between phobias
and the stress-diathesis model, Kendler, Myers, and Prescott
suggested the importance of nonassociative models of fear
acquisition, proposing vulnerability that primarily is biologi-
cally determined and innate. Most studies with humans sup-
port the idea that in some individuals, tendencies to develop
fear and anxiety are inherited, but learning through familial
1 68 Anxiety and Fear
transmission has been demonstrated to be extremely influen-
tial as well. These interacting influences are represented in
the diathesis-stress model, which integrates genetic or other
biological vulnerability (diathesis) with environmental influ-
ences, particularly early learning. Further development of this
model in the stress-vulnerability-protective factors approach
also introduces the idea that there can be genetic and other
biological hardiness as well, and certain earlier learning can
'inoculate' an individual from otherwise negative environmen-
tal events.
The contribution of heritability has ignited another ques-
tion: What is it that is inherited 7 The most widely accepted
answer is that one inherits a vulnerability to fear and anxiety
generally, which may or may not develop in pathological
proportions based on an individual's environment. Chronic
stress is an example of an environmental influence that may
induce problem fear and/or anxiety in a vulnerable person.
Without that stress, the individual might never have developed
an anxiety disorder.
Conclusions
The knowledge base about anxiety and fear continues to
evolve. There are scientific journals devoted to anxiety and
anxiety disorders as their exclusive subject matter. In the last
several decades, national organizations (e.g., Anxiety Disorders
Association of America) have come into being, with member-
ships including lay people afflicted with problem fear and/or
anxiety, and professionals who provide treatment and/or re-
search in the area.
In the anxiety and fear literature, theories of emotion and
emotion regulation have gained prominence in recent decades.
Learning, cognitive, and particularly cognitive-behavioral
models continue to be important, but, theorists increasingly
try to understand fear and anxiety in the context of the theoret-
ical work and basic research emerging from the field of
emotion.
Anxiety and fear are important emotional states, but are
in truth like the proverbial double-edged sword. On the one
hand, they are motivating emotions, prompting people to
behave in healthy and safe ways. Conversely, in extreme
forms, they are associated with misery and distress, causing
chronic and lifelong problems that can disrupt an individual's
life entirely and indefinitely if left untreated. While anxiety
and fear are scientifically fascinating, and reveal much about
the human condition, the associated distress can be
overwhelming. Anxiety disorders are also often associated
with other emotional problems, particularly depression, as
well as problems with alcohol and other drug abuse. While
there is an extensive knowledge base on anxiety and fear,
further conceptual work and more research is needed to better
understand and more fully understand these emotions and the
treatment of their pathological manifestations in anxiety
disorders.
See also: Aggression; Anxiety Disorders; Behavioral Genetics;
Classical Conditioning; Motivation; Phobias; Sex Differences; Sex
Roles; Social Anxiety Disorder; Stress and Illness; Subjective Culture;
Work Efficiency and Motivation.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental
Disorders, 4th edn., text rev. Washington, DC: American Psychiatric Association.
Arrindell WA, Kolk AM, Pickersgill MJ, and Hageman WJJM (1993) Biological sex,
sex-role orientation, masculine sex-role stress, dissimulation and self-reported
fears in a non-clinical sample from Britain. Advances in Behaviour Research and
Therapy AS 103-146.
Barlow DG (2002) Anxiety and its Disorders: The Nature and Treatment of Anxiety and
Panic, 2nd edn. New York: Guilford.
Beck AT and Emery G (1985) Anxiety Disorders and Phobias: A Cognitive Perspective.
New York: Basic Books.
Craske MG (1999) Anxiety Disorders: Psychological Approaches to Theory and
Treatment. Colorado: Westview Press.
Craske MG (2003) Origins of Phobias and Anxiety Disorders: Why more Women than
Men? Oxford, UK: Elsevier.
Craske MG, Hermans D, and Vansteenwegen D (2006) Fear and Learning: From Basic
Processes to Clinical Implications. Washington, DC: American Psychological
Association.
Kendler KS, Myers J, and Prescott CA (2002) The etiology of phobias: An evaluation of
the stress-diathesis model. Archives of General Psychiatry 59: 242-248.
Marks IM (1 987) Fears, Phobias, and Rituals: Panic, Anxiety, and Their Disorders.
New York: Oxford University Press.
Marks IM and Nesse RM (1994) Fear and fitness: An evolutionary analysis of anxiety
disorders. Ethology and Sociobiology 15: 247-261 .
McLean PD and Woody SR (2001 ) Anxiety Disorders in Adults: An Evidence-Based
Approach to Psychological Treatment. New York: Oxford University Press.
McNeil DW, Vrana SR, Melamed BF, Cuthbert BN, and Lang PJ (1993) Emotional
imagery in simple and social phobia: Fear versus anxiety. Journal of Abnormal
Psychology 102: 212-225.
Morris TL and March JS (2004) Anxiety Disorders in Children and Adolescents,
2nd edn. New York: Guilford.
Rachman SJ (1990) Fear and Courage, 2nd edn. New York: Freeman.
Rachman SJ (1998 ) Anxiety. United Kingdom: Psychology Press.
Tuma AH and Maser JD (eds.) (1985) Anxiety and the Anxiety Disorders. Hillsdale, NJ:
Erlbaum.
Wells A (2000) Emotional Disorders and Metacognition: Innovative Cognitive Therapy.
New York: Wiley.
Anxiety Disorders
J L Rodriguez and E A Meadows, Central Michigan University, Mount Pleasant, Ml, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Agoraphobia Fear and avoidance of situations associated
with panic, or from which escape might be difficult, due to
the fear of having a panic attack or not being able to
obtain help.
Diagnostic and statistical manual of mental
disorders The reference manual used by mental
health professionals to diagnose psychological disorders.
The DSM is published by the American Psychiatric
Association and contains information on the diagnostic
criteria, associated features, prevalence, and course for
each disorder.
Exposure A treatment technique that involves facing a
feared stimulus (including images, thoughts, objects, events,
or situations), and remaining with that stimulus until
anxiety has decreased.
Panic attack A rush of intense fear or discomfort
accompanied by physical sensations.
Phobia Excessive and/or unreasonable fear of an object or
situation.
What Is Anxiety?
In general, anxiety is good and is an adaptive survival mecha-
nism. It helps in keeping us from danger and in remaining
motivated. Without anxiety, we would not jump out of the
way from a car about to hit us, or study for a test, or be alert to
signs of potential danger. Anxiety can be described in terms
of behavior, thoughts, and physiological sensations. For exam-
ple, behavior related to anxiety involves moving out of the
way of the car that is about to hit us or paying attention
to our surroundings in potentially dangerous situations. An
anxiety-related thought might be, 'If I don't study, I may fail
my test.' Physiological sensations of anxiety might include
palpitations, sweating, or trembling. Thus, anxiety can be
very adaptive, such as when it leads us to take precautions to
protect ourselves from real dangers or to take action to run
from or fight off real dangers (e.g., the fight-or-flight response).
Even in safe situations, mild anxiety typically leads to better
performance than does none at all.
When anxiety is excessive for a situation, however, it may
become harmful rather than helpful. Maladaptive anxiety may
include the same behaviors as adaptive anxiety, but in different
circumstances. For example, the level of hypervigilance for
potential danger that makes sense while one is walking alone
in the street at night would likely be excessive while one is
walking in a hallway in an office building during the workday.
Another example of maladaptive anxiety is avoidance, such as
refusing to participate in class due to fear of evaluation. The
Diagnostic and Statistical Manual of Mental Disorders - Fourth
Edition, Text Revision (DSM-IV-TR) describes 11 disorders
related to anxiety, and as a whole, anxiety disorders are
among the most prevalent of all psychological problems.
It is important to recognize that anxiety falls along a con-
tinuum. So while everyone experiences anxiety and many
people experience common symptoms of an anxiety disorder,
this does not in itself mean they have an anxiety disorder;
diagnosable anxiety disorders require that the anxiety leads to
distress and/or functional impairment. For example, it is very
common for individuals to fear public speaking due to fear of
evaluation by others, a symptom of social phobia. Yet, many
people with this common fear will not meet the criteria for
social phobia because it does not bother them to have this fear,
nor does the fear or fear-related avoidance affect their lives. On
the other hand, if the same individuals, with the same fear level,
were to receive a job promotion that required them to give
presentations, the same fear of public speaking might now be
problematic for them, and might then be considered a diagnos-
able social phobia. Alternatively, the same individuals might find
themselves developing stronger fears of the same situation, and
becoming more distressed by their fear, which would also lead to
their reaching the threshold of a diagnosable social phobia.
Thus, distress or interference with functioning may be prompted
by symptoms becoming more intense or more frequent, or by
the same symptoms becoming problematic in a way that they
had not been before (e.g., having a new job requirement to give
presentations). Without distress or interference, however, even
excessive or irrational anxiety would constitute nonclinical
features of a disorder rather than a clinical disorder itself.
The Anxiety Disorders
Panic Disorder and Agoraphobia
Panic disorder is a common mental health condition that
affects between 1% and 3.5% of the population in their life-
time. This disorder requires unexpected panic attacks, at least
initially (often people with panic disorder then begin expecting
to panic in specific situations, and thus may experience cued
attacks as well as uncued ones). Panic attacks are periods
of intense fear or discomfort that peak within 10 min and are
accompanied by at least four of the following symptoms: palpi-
tations; pounding or racing heart; trembling or shaking; feelings
of choking; chest pain or discomfort; dizziness, unsteadiness,
lightheadedness, or faintness; fear of losing control or going
crazy; fear of dying; chills or hot flushes; numbness or tingling
sensations; sweating; nausea or abdominal distress; sensations
of shortness of breath or smothering; and derealization. Panic
attacks can occur in all anxiety disorders, but in panic disorder,
the attack is unexpected, rather than caused by a specific stim-
ulus. Further, panic disorder requires at least 1 month of one or
169
170 Anxiety Disorders
more of the following symptoms: persistent concern about
additional attacks, worry about the implications of an attack,
or a significant change in behavior linked to the attacks.
Finally, the attacks must not be due to a medical condition or
substance use.
The change in behavior noted above can take many forms.
It might include always having one's cell phone charged to be
able to call for help if necessary. Using distraction, such as
playing the radio loudly while driving, is another form of
behavior change. One of the most common forms of behavior
change is agoraphobia, or fear/ avoidance of being in places
where escape might be difficult or where help might not be
available if the individual were to have a panic attack. One-
third to one-half of all individuals with panic disorder also have
agoraphobia, and are thus diagnosed with panic disorder with
agoraphobia. (Those who do not develop agoraphobia are
diagnosed with panic disorder without agoraphobia.) Com-
mon situations avoided in agoraphobia include bridges and
tunnels, highways, and crowds. Such individuals avoid these
types of situations entirely or in part (e.g., they will enter such
situations with another person, but not alone) or endure them
with marked distress.
More infrequently, agoraphobia may exist in the absence
of panic disorder. This is most common in individuals who
experience limited symptom attacks, which include panic
symptoms fewer in number than are required for a panic attack.
It may also occur in the presence of other nonpanic symptoms,
such as someone with irritable bowel syndrome avoiding situa-
tions for fear of not having immediate access to a bathroom, in
those who have never experienced the symptoms they fear but
who avoid situations because 'it could still happen,' or for a
small number of people, without the individuals being able to
specify exactly what the fear is.
Specific Phobia
Specific phobia is characterized by excessive or unreasonable
fear that is induced by the anticipation or presence of a specific
object or situation, such as flying, animals, the sight of blood,
or heights. The five specific phobia subtypes classified within
the DSM-IV-TR are animal, natural environment (e.g., storms),
blood-injection-injury (BII), situational (e.g., flying), and other
(e.g., vomiting); however, there has been some discussion about
these subtypes in the preparation of DSM-V. The person must
recognize that the fear is unreasonable and the phobic stimulus
must be avoided or endured with significant distress.
As noted earlier, it is important to keep in mind that many
people have very strong fears that would not be considered
phobias, due to a lack of distress or functional impairment.
For example, a person may be very fearful of air travel;
however, if the person does not need or want to fly anywhere,
and does not mind having that fear, it would not rise to the
level of a diagnosable phobia. If that person is offered a job
that regularly requires air travel, however, the same fear may
now become a phobia (e.g., if having to fly now leads to
significant distress, refusing to fly would lead to poor job
performance reviews, or the person must turn down the
job despite otherwise wanting to accept it). Approximately
7-11% of people meet criteria for a specific phobia within
their lifetimes.
Social Phobia (Social Anxiety Disorder)
Social phobia, also referred to as social anxiety disorder (with
this term being considered as the official diagnostic term for
DSM-V) is characterized by fear of negative evaluation or of
embarrassing or humiliating oneself. Social phobia can be very
specific, limited to just one or two situations such as public
speaking, or it can extend to almost all social interactive situa-
tions. As with specific phobias, some social evaluative fears
(such as public speaking) are quite common, and do not neces-
sarily rise to the level of a diagnosable social phobia due to a
lack of distress regarding the fear or impairment (such as avoid-
ance of social evaluative situations) caused by the fear. In addi-
tion to the distress/impairment requirement, a diagnosis of
social phobia requires that the individual should recognize
that the fear is unreasonable, that anticipation of and/or expo-
sure to such social situations induces anxiety, and that these
situations are either avoided or endured with significant distress.
Social phobia is one of the most commonly diagnosed
anxiety disorders, affecting between 3% and 13% of the popu-
lation in their lifetime. Most commonly, people fear public
speaking and performance. Typically, people fear more than
one social situation; for those who fear most social situations
due to fear of evaluation, the specifier 'generalized' is given.
In some severe forms of social phobia, even situations not
usually thought of as socially evaluative might be avoided,
such as driving during the daytime when other drivers might
be encountered at stop lights, for example.
Obsessive-Compulsive Disorder
Obsessive-compulsive disorder (OCD) is characterized by
obsessions and compulsions. Obsessions are defined by the
DSM-IV-TR as "persistent ideas, thoughts, impulses, or images
that are experienced as intrusive and inappropriate and that
cause marked anxiety or distress." Typically, individuals with
OCD feel that the obsessions are not under their control and
often feel alien to them, in that they are not the kind of
thoughts the individuals wish to have, or that make sense to
them. Similarly, compulsions are defined as "repetitive beha-
viors or mental acts the goal of which is to prevent or reduce
anxiety or distress, not to provide pleasure or gratification,"
and thus are engaged in to neutralize the obsessions or other-
wise prevent feared thoughts or impulses from happening.
Thus, many behaviors often referred to as 'compulsive,' such
as gambling or shopping, would not be considered symptoms
of OCD, as these behaviors are both enjoyed and engaged in
for pleasure, albeit sometimes regretted later.
Common obsessions include contamination (e.g., being
contaminated by touching a door handle), a need to have
things in a certain order (e.g., feelings of distress when objects
are asymmetrical), aggressive or horrific impulses (e.g., to yell
out profanity in public or assault someone), repeated doubts
(e.g., wondering whether one has turned off the stove), and
sexual imagery (e.g., obscene thoughts or images); common
compulsions include washing and cleaning, checking, repeat-
ing actions, ordering, requesting or demanding assurances, and
counting. Often the compulsions are readily linked to the
obsession, although far in excess of what would realistically
make sense, as in the case of washing compulsions to neutralize
Anxiety Disorders 171
contamination fears. In other cases, the compulsions are not
realistically connected to the obsession the individual is
attempting to neutralize or prevent; for example, someone
with an obsession of hurting another person intentionally
may attempt to neutralize the thought by counting to ten back-
wards and forwards 10 times. Compulsions may include both
overt (e.g., hand washing) or covert (e.g., bringing up positive
images mentally to counteract horrifying images) behaviors.
For a diagnosis of OCD, individuals may experience obses-
sions only, compulsions only, or both, and these symptoms
must cause significant distress or functional impairment, and/
or take up at least one hour per day. Thus, as with the other
disorders reviewed, it is important to note that many indivi-
duals engage in some forms of compulsive behaviors at levels
that would not be considered clinically significant, such as
occasionally feeling as if one needs to check the locks again,
or even doing so again despite having done so already; without
causing distress or interference or being time consuming, such
symptoms would not be considered OCD.
There is also increasing attention being paid to problems
sometimes referred to as 'obsessive-compulsive spectrum
disorders,' such as skin picking, trichotillomania, and body
dysmorphic disorder, and there is some discussion about con-
solidating these disorders, currently classified elsewhere in the
DSM, into an obsessive-compulsive spectrum disorders group
in the DSM-V or a later edition. Further, compulsive hoarding,
currently classified as a form of OCD, has been the subject of
increased attention in both the popular and research literature,
and there is some discussion of classifying hoarding as a sepa-
rate disorder rather than as one form of OCD. At the time of
this writing, however, no such decisions have yet been made
for the DSM-V.
OCD is somewhat less common than other anxiety disor-
ders, affecting between 0.5% and 2.1% of the population in
their lifetime. It typically begins in adolescence and is as com-
mon among women as it is in men (as opposed to other anxiety
disorders, which are more common in women); however,
childhood-onset OCD is more common in boys than in girls.
Posttraumatic Stress Disorder
Posttraumatic stress disorder (PTSD) is one of only two anxiety
disorders that specify a cause for the onset of the disorder:
extreme trauma. The traumatic event must involve serious
injury or actual or threatened death, a threat to one's physical
integrity, or the witnessing or learning of such events. The
individual's response at the time of the trauma must be one
of fear, helplessness, or horror. Additional required symptoms
include reexperiencing of the trauma, avoidance of stimuli
associated with the event and general numbing of responsive-
ness, and increased arousal. Reexperience can range from flash-
backs to intense distress when exposed to reminders of the
trauma. In children, reexperiencing symptoms may take the
form of distressing dreams or repetitive play related to
the trauma. Avoidance can be effortful (avoiding thoughts, feel-
ings, or situations related to the trauma) and could also include
symptoms of emotional numbing, such as diminished interest
in significant activities, and feelings of detachment. Increased
arousal can take the form of difficulty in sleeping, outbursts
of anger, trouble in concentrating, and hypervigilance. These
symptoms must be present for over a month, must represent a
change since the trauma, and must cause distress or impairment
in the individual's functioning.
While PTSD does have a specific cause identified, it should
be noted that not all people with exposure to an extreme
trauma go on to develop PTSD. The severity of, proximity to,
and duration of the trauma affect the likelihood of an indivi-
dual's developing PTSD. Estimates of lifetime prevalence of
exposure to trauma range from 40 to 60% in the general
population, yet only about 8% of the population actually
develops PTSD. Clearly, exposure to an extreme trauma is
necessary but not sufficient for the development of PTSD.
Acute Stress Disorder
Acute stress disorder (ASD) is similar to PTSD in the require-
ment of a precipitating traumatic event; however, the time frame
needed for a diagnosis differs as do some of the prominent
symptoms. For ASD, the individual must exhibit symptoms
within 4 weeks of exposure to the traumatic event and these
symptoms must last a minimum of 2 days and a maximum of
4 weeks. If symptoms last longer than 1 month, a diagnosis
of ASD is no longer applicable, and a diagnosis of PTSD or
another disorder should be considered.
Like PTSD, ASD requires exposure to an extreme trauma;
response of fear, helplessness, or horror; symptoms of reexper-
iencing of the trauma, avoidance, and increased arousal and
significant distress or interference from these symptoms.
In addition, the individual must experience at least three
dissociative symptoms such as a subjective sense of numbing,
reduction in awareness of surroundings, derealization, deper-
sonalization, or an inability to recall important events of
the trauma; and at least one symptom of reexperiencing the
trauma; these symptoms must also interfere in the individual's
daily functioning.
Less information is available on the prevalence of ASD than
other anxiety disorders, but the lifetime prevalence rate is
estimated to be between 14% and 33% for individuals exposed
to severe trauma. Individuals with ASD also have an increased
risk of developing PTSD.
Generalized Anxiety Disorder
Generalized anxiety disorder (GAD) is characterized by
excessive and persistent worries that the individual feels are
uncontrollable. The worry must be experienced more days than
not for at least 6 months, and focus on multiple topics. The
diagnosis of GAD also requires at least three of the following:
restlessness, fatigue, difficulty in concentrating, irritability,
muscle tension, or sleep disturbance. Finally, the individual
must experience marked distress or impairment in functioning
due to these symptoms.
Those with GAD worry about common concerns, such as
work, health, and finances, but the degree of worry is excessive,
given the actual likelihood or impact of the event. Further,
those with GAD feel as if they cannot stop worrying even if
they want to. Often, the worries interfere with other tasks
because the person is unable to focus on other tasks due to
worrying, or engages in reassurance-seeking behaviors such as
calling home to make sure everyone is okay.
1 72 Anxiety Disorders
The lifetime prevalence rate for GAD is ~5%, and the
disorder is more common in women than in men. Often,
people with GAD report worries as far back as they can remem-
ber, and describe their worry as a personality trait.
Anxiety Disorder due to a General Medical Condition
Anxiety disorder due to a general medical condition is charac-
terized by symptoms of anxiety that are deemed the direct
consequence of the medical condition, with the anxiety symp-
toms causing distress or functional impairment. The relation-
ship of anxiety to the general medical condition is usually
determined by examining the history, laboratory findings, or
physical examination of the individual. Some medical condi-
tions that have been known to cause anxiety symptoms include
hyperthyroidism, cardiac arrhythmia, pneumonia, vitamin
B12 deficiency, and encephalitis.
Substance-Induced Anxiety Disorder
Anxiety symptoms that cause marked distress or an impairment
in functioning and are the direct effect of substance use are
categorized as substance-induced anxiety disorder. Here again,
the relationship of the anxiety to the substance must be deter-
mined by examining the history, laboratory findings, or physi-
cal examination of the individual. The anxiety symptoms must
occur during intoxication or within 1 month of use and should
be in excess of typical anxiety symptoms incurred during use or
withdrawal from the substance.
Anxiety Disorder Not Otherwise Specified
Anxiety disorder not otherwise specified (NOS) is diagnosed
when someone experiences a clinically significant level of anx-
iety but does not meet criteria for a specific anxiety or adjust-
ment disorder. For example, a person who reports excessive
worry that interferes with functioning but without enough
accompanying physical symptoms (e.g., fatigue, muscle ten-
sion) to warrant a diagnosis of GAD could be diagnosed with
anxiety disorder NOS.
Etiology
As with the majority of mental disorders, it is difficult to
identify a single cause for the development of anxiety disor-
ders. Anxiety disorders are thought to develop due to a combi-
nation of factors. While no single gene has been found that
passes on an anxiety disorder to an individual, we know that
children who have parents with anxiety disorders are more
likely to have anxiety disorders. Some researchers believe that
rather than a deficient gene that is passed from one generation
to another, it is a general vulnerability or sensitivity to anxiety
that is transmitted. This hypothesis may be supported by the
fact that it is not uncommon for children to have a different
anxiety disorder than that of their parents.
Environmental factors also play a role in the development
of anxiety disorders. Parents may model anxious behavior for
their children, or may transmit messages suggesting that vari-
ous objects, situations, or feelings are dangerous and should be
avoided. Excessive levels of stress may trigger biological and
psychological vulnerabilities that can manifest in an anxiety
disorder.
Many clinicians focus less on causal than on maintaining
factors in anxiety disorders, with the idea that one cannot
change the past, but can change the present. Specifically,
behavioral, cognitive, and physiological factors are thought to
maintain anxiety. Generally speaking, individuals with anxiety
disorders tend to avoid anxiety-provoking stimuli or to distract
themselves so that they do not feel as anxious. While these
behaviors can be helpful in temporarily reducing anxiety, they
are harmful in the long term because the individual does not
learn that the anxiety will decrease eventually on its own, even
without such avoidance. Cognitive factors also play a role in
the maintenance of anxiety disorders, with negative and unre-
alistic thoughts often perpetuating anxiety. For example, some-
one with social phobia may believe that everyone around her
thinks she is stupid for dropping her books when in reality
most people did not notice, and those who did notice did
not think about the event or pay any further attention to it.
Finally, physiological factors, such as higher autonomic labil-
ity, leading to frequent fluctuations in bodily sensations such
as heart rate, can also maintain anxiety. Often, these maintain-
ing factors interact to perpetuate a cycle of anxiety.
Treatment
Cognitive Behavioral Treatments
Currently, the psychosocial treatment with the most empirical
evidence to support its use in the treatment of anxiety dis-
orders is cognitive-behavioral therapy (CBT). CBT may include
a variety of treatment components, such as psychoeducation,
relaxation, exposure, and cognitive challenging; exposure is
by far the method most implicated in the successful treat-
ment of anxiety disorders. Psychoeducation refers to providing
information about the nature of anxiety, such as the interac-
tion of different components described in the previous section.
Relaxation training, often in the form of breathing retraining,
is taught as a coping strategy. Breathing retraining involves
breathing more slowly, and diaphragmatically rather than from
the chest, to dampen physiological sensations. When using
such physical relaxation methods, however, especially in con-
junction with exposure treatment, therapists generally empha-
size that the idea is not that one must use these methods to
decrease anxiety, or that experiencing anxiety is a dangerous
thing that must be avoided by using such methods. Rather,
such methods are presented as ways for individuals to decrease
anxious responding at times when it is inconvenient (e.g.,
during an exam), and the anxiety should then be revisited at
a later time.
The cognitive component of CBT typically involves recogniz-
ing and challenging maladaptive and/or unrealistic thoughts.
Once the thoughts have been recognized, clients engage in an
examination of the evidence for and against such thoughts and
of the consequences of these thoughts, leading to the develop-
ment of a rational response that best fits the evidence and the
individual's desired outcome. Common unrealistic thoughts
include overestimating the likelihood of a particular outcome
(such as thinking that a panic attack will lead to a heart attack)
Anxiety Disorders 1 73
and catastrophizing, or thinking things are much worse than
they are, or that they cannot be coped with (such as fearing
embarrassment as something one cannot live with). Generally,
consequences that are likely are also the least serious, while
those that are the most serious also tend to be the least likely,
and so countering the two thinking errors described above is
generally sufficient for most anxiety-related thoughts. Another
component in cognitive challenging is to continue drawing out
one's fears to the fullest; this generally allows the individual to
recognize that most of these fears are in fact quite irrational,
something that had not been noticed due to the refusal to
consider the next steps in a chain of anxious thoughts. For
example, those with social phobia often fear being laughed at
or being embarrassed in front of others, but may not continue
drawing out those scenarios. When the scenarios are drawn out
further, they may realize that their fears include having to run
out of the room, that everyone will continue talking about
their performance, that it will be printed in the paper for others
to laugh at, etc. Using this method, it becomes much clear that
these further consequences are truly unlikely, and that without
those additional consequences, temporary embarrassment is
not that big a deal.
Exposure, a key component in most CBT programs and
the component with by far the most research to support its
use, requires individuals to confront anxiety-provoking stimuli
and to remain with those stimuli until their anxiety decreases.
Exposure can be conducted in a gradual manner, beginning
with easier situations and moving on to harder ones, or all at
once, referred to as flooding. With flooding, the individual
confronts the most anxiety-provoking stimuli from the start.
While flooding has been found to be as effective as exposure
conducted in a gradual manner, most clinicians use gradual
exposure over flooding due to the decreased perceived distress
related to working up to the most feared anxiety-provoking
stimulus. Flooding may be preferred by some individuals,
however, such as those who wish to overcome their problems
in the quickest manner possible.
There are many different forms of exposure used to target
different kinds of stimuli. In in vivo exposure, or real-life
exposure, the individual physically confronts the feared sti-
muli. For example, a person with a specific phobia of snakes
would remain in a room with a snake, touch a snake, or hold
a snake. A person with a fear of public speaking would give a
speech in front of others. Imaginal exposure is often used when
in vivo exposure is not feasible or the individual is reluctant to
begin with in vivo exposure, or when the anxiety-provoking
stimulus is an image or memory. In the previous example,
if the snake-phobic individual refused to get anywhere near a
real snake, imaginal exposure might be used first, having the
person imagine touching a snake, with the goal of then moving
on to the in vivo exposure. Imaginal exposure is often used in
the treatment of PTSD, wherein the memory of the trauma
itself is anxiety-provoking. Interoceptive exposure, generally
used for panic disorder, involves exposure to bodily sensations
such as fast breathing or increased heart rate. Exercises such as
running in place or spinning in a chair are used to induce
feared sensations and continued until anxiety diminishes. Pho-
bias are particularly helped by exposure treatment, as the focus
of fear in these disorders is so specific. As noted above, these
phobias are typically treated with in vivo exposure, with imaginal
exposure sometimes being used to start off, and sometimes with
the use of modeling as well, wherein clients first observe the
therapist confront the feared object before confronting it them-
selves. In general, exposure treatments are conducted similarly
across the anxiety disorders, as only the focus of the exposure
needs to vary. One exception to this is in the treatment of BII
phobia, in which the phobic response is often a bit different
from that in other phobias. Specifically, while most anxiety
responses include increased physiological arousal, BII phobics
often respond with decreased blood pressure, lightheadedness,
and even fainting, an opposite physiological response. Thus, for
BII phobia, applied muscle tension is generally added (gener-
ally by clenching one's fists), to prevent the fainting response by
boosting blood pressure during the exposure.
Another addition to exposure treatment is response preven-
tion, which is used for individuals with OCD. Individuals
are first exposed to the thought, image, or impulse that triggers
the obsession, but then they must also refrain from engaging
in any behaviors or mental rituals aimed at neutralizing that
obsession. For example, an individual with contamination
obsessions would touch objects felt to be contaminated, and
then refrain from washing. In complete response prevention,
sometimes used in the treatment of more severe cases of OCD,
the neutralizing behaviors are completely avoided; for exam-
ple, contamination-fearful individuals would not touch any
water at all for a period of several days, and would immediately
'recontaminate' upon the completion of the allowed shower at
that time. Of course, in any exposure treatment it is important
that clients do not engage in safety behaviors, whether they are
ritualizing, distraction, or other behaviors, that detract from
the idea of fully confronting one's fears.
Finally, exposure treatment may be augmented by the addi-
tion of D-cycloserine, an antibiotic that has been shown in
some recent studies to enhance the learning process that occurs
during exposure. This medication is not used as a treatment in
itself for anxiety, but rather, is taken prior to the exposure
session, to facilitate the exposure treatment.
Medication
Two different classes of drugs have been extensively studied for
the treatment of anxiety disorders: anxiolytics and antidepres-
sants. The anxiolytic drugs, benzodiazepines, are often used to
treat anxiety disorders, but it is typically recommended that
they be used as a short-term rather than long-term treatment
due to the increased likelihood of dependency with long-term
use. Further, both low-potency benzodiazepines (such as diaz-
epam) and high-potency benzodiazepines (such as alprazo-
lam) tend to decrease anxiety only temporarily. In addition
to the dependency issue and the related withdrawal symptoms,
benzodiazepines tend to have fairly high relapse rates once the
medication is discontinued, and alprazolam for panic disorder
may also lead to rebound panic attacks, where the panic attacks
are more frequent and/or severe, following discontinuation,
than they had been before treatment.
Antidepressants, such as selective serotonin reuptake inhibi-
tors (SSRIs), tricyclic antidepressants (TCAs), and monoamine
oxidase inhibitors (MAOIs), have been shown to be helpful in
treating a number of anxiety disorders. SSRIs (e.g., fluvoxamine,
fluoxetine, sertraline) are often the first-line pharmacological
1 74 Anxiety Disorders
treatment for anxiety disorders due to their tolerability and their
effectiveness in many anxiety disorders. The TCA most com-
monly used in the treatment of anxiety is imipramine, which
is most indicated in the anxiety realm as a treatment for panic
disorder; another TCA, clomipramine, is indicated for the treat-
ment of OCD. MAOIs have also been shown to be effective in
some anxiety disorders, especially social phobia and panic dis-
order; however, they carry a greater risk. Specifically, those on
MAOIs must adhere to a strict diet, avoiding foods (such as red
wine and aged cheese) with tyramines and related substances in
order to avoid a steep rise in blood pressure, which can be fatal.
It should be noted that while pharmacological treatments
have been shown to have a positive effect in terms of symptom
remission for many with anxiety, they are often found to be
no more effective, and sometimes less effective, than cognitive-
behavioral treatments. Further, cognitive-behavioral treatments
tend to be more effective in the long term, have a lower likeli-
hood of relapse, involve lower overall cost, and have fewer
adverse effects.
Other Psychosocial Treatments
While cognitive-behavior therapies and medication treatments
enjoy considerable empirical support for their use, they are
not the only treatments used for anxiety disorders. Another
common intervention for anxiety disorders is psychodynamic
therapy. In this form of therapy, the anxiety is not seen as the
problem in itself, but rather as the symptom of an underlying
problem that needs to be recognized and resolved. Thus, the
focus of psychodynamic treatment is not on reducing anxiety
symptoms directly. While this treatment does not have the
extensive research to support its use in decreasing anxiety that
CBT or medication treatments do, some individuals prefer
to focus on the possible causes of their problems rather than
on the alleviation of their anxiety symptoms, and thus may
prefer this more insight-oriented approach to treatment.
Similarly, there are numerous self-help approaches for the
treatment of anxiety, as well as other therapist-led treatments
such as group therapy, that individuals sometimes find more
helpful and/or prefer, compared to the CBT or medication
approaches.
Issues in Diagnosis
Among the anxiety disorders, there is significant overlap in
symptoms, which is reasonable given that this group of dis-
orders requires the presence of anxiety that causes the person
marked distress or impairment in functioning. While the
reason for symptom overlap is expected, sometimes differen-
tiating between different anxiety disorders can be difficult.
For example, people with social phobia as well as those
with agoraphobia may fear going to a store. They may both
experience significant physiological sensations while in this
situation, such as increased heart rate, sweating, and trembling.
They may both attempt to leave the situation or distract them-
selves to feel less anxious, and both may exhibit impairment in
their social functioning due to their anxiety. They may even
fear evaluation by others in this situation. However, what
differentiates these two disorders is the specific focus of their
fear. For individuals with social anxiety, their fear centers
on others negatively evaluating them, such as the belief that
others will think they are stupid. Individuals with agoraphobia
fear having symptoms of a panic attack or not being able to
escape a situation if they experience these panic symptoms.
While they may also fear the evaluation by others, this fear
would be focused on others noticing their panic symptoms,
not the social phobia focus of tripping, saying something
stupid, or other such foci separate from the anxiety symptoms
themselves; further, the person with social phobia is likely to
feel anxious only in the presence of others, whereas the person
with panic disorder will likely be anxious even when others are
not around.
As treatment often stems from the specific diagnosis (e.g.,
applied tension for blood phobia) and certainly on the specific
focus of the anxiety, it is important to make an accurate diag-
nosis. For example, treating all panic attacks with interoceptive
exposure ignores the reality that only in panic disorder is the
panic attack itself the focus of anxiety, and thus an appropriate
target for interoceptive exposure; an individual who experi-
ences panic attacks when near a dog would not be expected
to benefit from interoceptive exposure, and should instead
receive in vivo exposure to dogs, if properly diagnosed with a
specific phobia of dogs.
Special Populations
Children
It is not uncommon for children to experience excessive
anxiety or worry that warrants an anxiety disorder diagnosis;
in fact, some anxiety disorders, such as GAD, often have an
onset in childhood. Separation anxiety disorder, excessive anx-
iety related to separation from parents or other caregivers,
is relatively common in childhood, yet rarely diagnosed in
adulthood. Children with separation anxiety disorder often
experience significant distress at the idea of separation from
caregivers, and may be reluctant, or outright refuse, to attend
school, go on sleepovers or to camp, or even be in a different
room from their parents while at home. This fear is generally
focused on the possibility of caregivers being harmed while
away from the child, their not coming back for the child,
or harm befalling the child while away from the caregivers;
the fear is often expressed in nightmares and/or physical symp-
toms, such as stomach aches instead of, or in addition to,
words. Symptoms must persist for at least 4 weeks for such a
diagnosis to be warranted; ~4% of children meet criteria for
separation anxiety disorder in their lifetime.
In general, children are more likely to express their
anxiety through somatic symptoms such as headaches, vomit-
ing, and stomachaches, than are adults. This may be due to
their developmental stage and their lack of insight into their
symptoms. For example, children with a phobia may demon-
strate freezing, tantrums, or crying when presented with their
particular feared stimulus. Often, children with an anxiety
disorder do not report distress related to the fear, especially if
a parent accommodates the child and his or her fears, such as
by allowing a child with separation anxiety to stay with the
parent throughout the day.
Anxiety Disorders 175
Elderly Adults
As is true of children, elderly adults may also experience the
same anxiety disorders as other adults. Differential diagnosis
may become even more important in this population, however,
when one recognizes that medical conditions and neurological
difficulties, more common among geriatric patients, can cloud
the diagnostic picture. For example, it is not uncommon for
those with dementia to exhibit symptoms of anxiety, especially
in the early stages of the disorder when the dementia has not
been clearly recognized. While both dementia and anxiety
disorders can coexist, for a specific anxiety disorder diagnosis
to be warranted, the anxiety symptoms must be in excess of
those typically found in dementia.
It is also fairly common for older adults to experience
worries and anxieties related, in full or in part, to aging. For
example, while some worries tend to decrease as people age,
other specific worries, such as those about one's health, per-
sonal safety, finances, and independence may increase with
age, due to changing life and personal circumstances, which
make these issues more salient.
Dually Diagnosed/Severely Mentally III Patients
It is not uncommon for individuals with an anxiety disorder to
also have another diagnosis. Anxiety disorders tend to co-occur
with one another. For example, as many as half of the indivi-
duals with GAD may also have another anxiety disorder, espe-
cially social phobia. Anxiety disorders also commonly co-occur
with depressive disorders and substance use disorders. Some
individuals with anxiety disorders self-medicate with alcohol
or recreational drugs, using these substances to cope with
anxiety-provoking situations or to mask their symptoms of
anxiety.
Due to this high comorbidity, careful assessment is needed
not only of anxiety, but also of related problems. Additional
treatment may be needed to target these other problems,
although treating the primary anxiety problem may also gener-
alize to other anxiety disorders, or alleviate the co-occurring
depression or substance use.
Individuals who suffer from severe mental illness, such as
psychotic disorders, along with an anxiety disorder may also
benefit from anxiety treatment. For example, a person with
schizophrenia who also suffers from panic disorder with ago-
raphobia may avoid leaving home for fear of having a panic
attack. This could interfere with the person's obtaining treat-
ment for schizophrenia. With treatment for agoraphobia, the
individual may be more likely to properly manage the schizo-
phrenia. Further, even anxiety that is not separate, but instead
is better considered part of the psychotic (or other) disorder,
may be alleviated by targeted intervention; this symptom
reduction, while not 'fixing' the primary problem, may cer-
tainly improve the individual's life by removing at least some
source of distress.
Non-Western Cultures
Anxiety disorders can be seen across the world; they are not
simply a Western phenomenon. A number of culture-bound
syndromes, or disorders specific to a particular region, resemble
DSM-IV-TR diagnoses, but often with a focus that more reflects
culturally specific fears . Ataque de nervios, seen in Latin American
and Latin Mediterranean cultures, is associated with a sense of
being out of control, uncontrollable shouting, trembling, crying
heat in the chest rising to the head, and fainting or seizure-like
episodes, and somewhat resembles panic disorder. Koro, found
in south and east Asia, refers to extreme anxiety and panic regard-
ing the penis (in men) or vulva and nipples (in women) retract-
ing into the body and causing death. In Korea, shin-byung is
characterized by excessive anxiety, dissociation, and somatic
complaints (e.g., weakness, dizziness, insomnia, gastrointestinal
problems) that are attributed to possession by ancestral spirits.
Numerous other culture-bound syndromes exist as well.
Medically III Patients
Anxiety disorder symptoms may also overlap with a range of
medical conditions; for this reason, physical exams are gener-
ally recommended before anxiety treatment begins, to rule
out a medical explanation for the anxiety and/or to ensure
that anxiety treatment is safe medically. For example, hyper-
thyroidism can result in palpitations, nervousness, insomnia,
fatigue, trembling, and sweating, and can resemble a panic
attack. While the symptoms may be similar for both issues,
the treatments differ greatly.
Medical conditions can also lead to diagnosable anxiety
disorders, when the initial symptoms caused by the medical
problem become exacerbated by the attention to those symp-
toms or what they may lead to. For example, a person with
hyperthyroidism may become quite frightened of the increased
heart rate, and fear that it is leading to a heart attack and death.
This fear may prompt increased monitoring of heart rate and
heart rate changes, leading to more frequent fears of heart attack,
well beyond what the hyperthyroidism itself would involve.
In such cases, where the anxiety is beyond that expected from
the medical condition alone, an anxiety disorder may also be
diagnosed.
Treatment for anxiety may also need to take medical con-
ditions into account. For example, while interoceptive expo-
sure may be among the best treatments for panic disorder,
it may be contraindicated for an individual who also has a
heart condition and must avoid the increased heart rate that
such exposure often entails. In such cases, exposure may still be
used, perhaps with modifications such as monitoring heart rate
to ensure it does not exceed the medically advised level, or an
alternative treatment may be used instead.
Conclusion
Anxiety disorders are common across gender, age, and culture.
Accurate diagnosis of an anxiety disorder rests on determining
the specific feared stimuli and the presence of functional
impairment or distress related to this fear. Generally speaking,
CBT is the psychosocial treatment of choice for all anxiety
disorders, and it has also been shown to be effective when
treating clients with severe mental illness, dual diagnoses,
and medical illnesses. Medication is frequently used to treat
anxiety disorders and is often preferred by many for its initial
lower cost in terms of money, time, and effort; however,
176 Anxiety Disorders
medication tends to be less cost effective in the long term
and involves higher relapse rates when discontinued than
does CBT.
See also: Agoraphobia and Panic Disorder; Anxiety and Fear;
Childhood Mental Disorders; Clinical Assessment; Cognitive Behavior
Therapy; Evidence-Based Practice; Generalized Anxiety Disorder;
Hypochondriasis or Health Anxiety; Obsessive-Compulsive Disorder;
Obsessive-Compulsive Personality Disorder; Phobias; Posttraumatic
Stress Disorder; Psychotherapy; Separation Anxiety; Social Anxiety
Disorder.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental
Disorders, 4th edn.. Text Revision. Washington, DC: Author.
Antony MM, Orsillo SM, and Roemer L (2001) Practitioner's Guide to Empirically Based
Measures of Anxiety (AABT Clinical Assessment Series). New York: Kluwer
Academic/Plenum.
Barlow DH (2002) Anxiety and Its Disorders: The Nature and Treatment of Anxiety
Disorders, 2nd edn. New York: Guilford.
Barlow DH (ed.) (2007) Clinical Handbook of Psychological Disorders: A Step-by-Step
Treatment Manual, 4th edn. New York: Guilford.
Clark DA and Beck AT (2009) Cognitive Therapy of Anxiety Disorders: Science and
Practice. New York: Guilford.
Craske MG (1999) Anxiety Disorders: Psychological Approaches to Theory and
Treatment. Denver, CO: Westview Press/Basic Books.
Craske MG (2003) The Origins of Phobias and Anxiety Disorders: Why More Women
than Men. Oxford, England: Elsevier Science.
Foa EB, Keane TM, and Friedman MJ, et al. (eds.) (2008) Effective Treatments for PTSD:
Practice Guidelines from the International Society for Traumatic Stress Studies,
2nd edn. New York: Guilford.
Morris TL and March JS (2004) Anxiety Disorders in Children and Adolescents,
2nd edn. New York: Guilford.
Orsillo SM and Roemer L (eds.) (2005) Acceptance and Mindfulness-Based
Approaches to Anxiety: Conceptualization and Treatment. New York: Springer.
Oxford University Press Treatments that Work series (a set of client and clinician
manuals for numerous anxiety disorders as well as other disorders); details of all
manuals available at http://www.oup.com/us/catalog/general/series/
TreatmentsThatWork/?view=usa.
Rothbaum B (2005) Pathological Anxiety: Emotional Processing in Etiology and
Treatment. New York: Guilford.
Tolin D, Frost R, and Steketee G (2007) Buried in Treasures: Help for Compulsive
Acquiring, Saving, and Hoarding. New York: Oxford University Press.
Relevant Websites
http://www.adaa.org - Anxiety Disorders Association of America.
http://www.ocfoundation.org - International 0CD Foundation.
http://istss.org - International Society for Traumatic Stress Studies.
http://www.nimh.nih.gov/health/topics/anxiety-disorders/index.shtml - National
Institute of Mental Health Anxiety Disorders Page.
http://www.abct.org - The Association for Behavioral and Cognitive Therapies.
http://www.therapyadvisor.com - Therapy Advisor Website.
www.trich.org - Trichotillomania Learning Center.
Aphasia
H S Kirshner, Vanderbilt University School of Medicine, Nashville, TN, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Agraphia An acquired inability to write.
Alexia An acquired disorder of reading.
Aphasia An acquired disorder of language caused by
brain disease.
Frontotemporal dementia or frontotemporal lobar
degeneration A neurodegenerative condition in which
frontal lobe behavioral syndromes or aphasia are the
presenting features.
Language A system of symbols used to convey thoughts, or
internal, nonverbal images and representations, from one
individual to another.
Lexical Of or pertaining to a word.
Literal (phonemic) paraphasic error An incorrect utterance
based on substitution of an erroneous sound.
Morpheme A word ending which may vary depending on
the use of a word, for example, in singular or plural,
masculine or feminine, present or past tense.
Phoneme The smallest unit of speech which can alter
the meaning of a word.
Semantics The meaning of words.
Syntax A system of grammatical rules used in language.
Verbal (semantic) paraphasic error An incorrect utterance
based on substitution of an erroneous word.
Aphasias are disorders of language, or symbolic communica-
tion, acquired as a result of brain disease. Aphasia strikes at that
most human of capabilities, symbolic language, and deprives
the subject of communication with others.
By this definition, aphasia can be distinguished from several
related conditions. First, aphasia is a disorder of language, and
not a motor speech disorder. Motor speech disorders are
abnormalities of speech articulation brought about by weak-
ness or incoordination of the muscles of the speech apparatus.
Neurological motor speech disorders include dysarthrias,
abnormalities of speech articulation; dysphonias, disorders
of voice; and speech apraxias, impairments of the ability to
sequence a series of phonemes. In motor speech disorders,
language comprehension and production are normal; in a
patient with abnormal sounding speech output, language pro-
duction can be ascertained by examining the patient's writing or
by transcribing speech output into type. Second, aphasia is an
acquired language disorder, not a congenital or developmental
disorder. In North America, developmental language disorders
are often called 'dysphasias,' just as developmental reading
disorders are 'dyslexias,' and developmental writing disorders
are 'dysgraphias.' It should be noted that in the UK, and in
Europe, these terms are sometimes used in the sense of partial
and not total language disorders. Third, the language disorder
of aphasia must relate to brain disease, and not to psychiatric
disorders. Psychotic disorders such as schizophrenia and bipo-
lar affective disorder typically derange thought, or the content
of language, rather than language itself. This last distinction is
the most problematic, for two reasons. First, language is diffi-
cult to separate from thought. Even severely aphasic patients,
however, seem capable of nonverbal thought, and their behav-
ior, outside of language function, usually seems normal. On the
other hand, patients with psychoses such as schizophrenia
usually express their disturbed thoughts in normal articulation
and syntax. Second, these psychotic disorders, though they
remain in the domain of psychiatry rather than neurology, are
likely brain diseases as well.
Introduction
History of Aphasia
Historically, language was the first of the higher cortical func-
tions to be correlated with focal lesions of the brain. The
French physician Paul Broca, in 1861, provided a clear descrip-
tion of nonfluent speech associated with lesions of the domi-
nant hemisphere, especially the left frontal lobe. The principal
syndrome of nonfluent aphasia is still called 'Broca's aphasia.'
Similarly other nineteenth century physicians correlated areas
of brain damage, usually resulting from stroke, with specific
syndromes of abnormal language. Carl Wernicke, in Germany,
described the fluent aphasia syndrome which now bears his
name, as well as conduction aphasia; Lichtheim outlined the
'transcortical aphasias'; and Dejerine designated the syndromes
of pure alexia with and without agraphia. The late nineteenth
and early twentieth centuries saw slow progress in the under-
standing of aphasia, largely because physicians had to outlive
their patients to correlate an aphasic syndrome with a brain
lesion at autopsy. In addition, the early twentieth century wit-
nessed a decline in interest in behavioral syndromes, in favor
of the 'hard' neurological deficits of paralysis, sensory loss,
and incoordination. The higher cognitive functions were either
accorded little interest or assumed to have a diffuse distribution
throughout the brain.
Modern Techniques in Aphasia Research
In the late twentieth and early twenty-first centuries, aphasia
has undergone a revolutionary expansion of knowledge. First,
brain imaging techniques such as computed tomography (CT)
and magnetic resonance imaging (MRI) have permitted the
study of brain lesions in living patients, simultaneously with
behavioral studies. Functional brain imaging modalities such
as positron emission tomography (PET), single photon emis-
sion computed tomography (SPECT), and functional MRI
(fMRI) allow the mapping of brain regions which are
177
178 Aphasia
metabolically active or show increased blood flow during spe-
cific behavioral tasks in normal subjects or in patients with
aphasia. In addition, the language tasks themselves have been
refined. Speech/language pathologists administer standardized
test batteries to compare performance between subjects, or in
the same subject over time. Neuropsychologists, cognitive neu-
robiologists, and linguists have constructed language tasks to
test specific steps in language processing at which deficits
occur, thereby creating models of the language process. Finally,
knowledge about language and the brain has been acquired
not only from the study of patients with strokes, head injury,
and brain tumors, but also in normal subjects, with the
use of functional brain imaging. In addition, in the evaluation
of patients with epilepsy, to prepare for surgical resection of
epileptogenic foci, techniques of electrical stimulation of brain
loci have permitted mapping of areas of the language cortex.
These areas can then be correlated with the results of surgery to
ablate epileptogenic foci. These additional sources of informa-
tion about language and the brain have modified our maps of
language function in the brain. Stimulation of the cortex in
epileptic patients, for example, has confirmed the general con-
cepts of the localization of language functions derived from
stroke patients, but electrical stimulation has identified new
language areas and has suggested that the traditional areas may
be smaller and more variable than previously thought.
Handedness and Cerebral Dominance
As Broca observed, the vast majority of right-handed patients
with aphasia have left hemisphere lesions, suggesting that the left
hemisphere is dominant for language function. In general, at
least 95% of right-handed patients and a majority of left-handed
patients have relative left hemisphere dominance for language.
Cerebral dominance can be determined definitively by Wada
testing, in which sodium pentobarbital is injected into the inter-
nal carotid artery, thus temporarily 'putting to sleep' the frontal,
temporal, and parietal cortical areas of one hemisphere. Wada
testing is carried out to determine whether an area of the brain
can be surgically removed without producing aphasia. Newer
methods of determining language dominance include mapping
of language activity during language tasks by PET or fMRI,
magnetoencephalography, and other techniques. The Wada
test, however, remains the gold standard for determination of
language dominance. The dominance of language in the left
hemisphere appears genetically programmed, since it is present
even in illiterate people. Geschwind and Levitsky pointed out
that, in most human brains, the superior temporal plane ('pla-
num temporale') is larger in the left hemisphere than in the right.
Subsequent measurements using CT and MRI scans have shown
similar asymmetries in living patients. The most reliable mea-
surement is the occipital length, which is longer in the left than
the right hemisphere in most right-handed people. This asym-
metry is present at birth, again indicating that the enlargement of
the left language cortex is genetically programmed, rather than
acquired through use in language function. Studies have sug-
gested that patients with atypical cortical asymmetries on brain
imaging may have abnormal language dominance and atypical
recovery patterns after injury to the language cortex.
Occasionally, a right-handed patient with an acquired right
hemisphere lesion develops aphasia, a phenomenon termed
'crossed aphasia,' or 'crossed aphasia in dextrals.' These
patients presumably have atypical cerebral dominance, and
some have atypical hemispheral asymmetries on CT scan. The
specific 'crossed aphasia' syndromes are frequently atypical,
milder than would be expected with a lesion of similar size in
the left hemisphere. Patients may also have expressive language
in the left hemisphere, and receptive language in the right, or
vice versa, leading to partial syndromes in which deficits are
more specific to expressive or receptive functions than in the
typical, expected aphasia syndrome.
Aphasia in left-handed patients also differs from the typical
syndromes of right-handed patients with aphasia secondary to
left hemisphere lesions. More left-handed patients than right-
handed patients develop aphasia after a stroke, regardless of
the side of the stroke, suggesting that left-handed patients have
some language representation in both cerebral hemispheres.
Though aphasia is more common in left-handers, recovery
may be better. Taken together, these findings imply that the
language dominance of left-handed patients is less rigid than
that of right-handers, and either hemisphere can subserve recov-
ery of language function. Most left-handed aphasic patients in a
large series had typical aphasia syndromes, but two patients had
large right hemisphere infarctions, with nonfluent aphasia
but preserved comprehension. These patients appeared to have
language expression arising from the right hemisphere, but lan-
guage comprehension from the left. Most left-handers, like
right-handers, have left hemisphere dominance for speech
and language, but there may be separate loci of dominance
for handedness, motor speech, and auditory comprehension.
There is some evidence that women have less rigorously defined
language dominance than men, and recovery may be better
in women.
Epidemiology and Etiologies of Aphasia
Aphasia has classically been described in association with focal
brain lesions. The most common such lesion is stroke, long
the leading etiology of aphasia in research studies. Stroke can
be thought of as an 'experiment of nature,' in which one area of
the brain is damaged, while the rest remains intact. Both major
types of stroke, cerebral infarctions, in which the brain is
damaged by reduced blood flow, and hemorrhages, or bleeds
into the brain, cause aphasia. Approximately 750 000 strokes
occur annually in the United States. About 20% of stroke
patients, or over 100 000 persons each year, develop aphasia
following a stroke. The total prevalence of aphasic stroke vic-
tims in the United States approaches 1 million.
Traumatic brain injury frequently disrupts discourse and
produces other language impairments, which are often part
of a syndrome of general cognitive dysfunction. Penetrating
brain injuries cause focal areas of damage which resemble
strokes, except that the lesion locations are different, and the
'remote effects' of swelling, increased intracranial pressure, and
contrecoup injuries are greater with traumatic brain injuries
than with stroke.
Brain tumors, abscesses, and other mass lesions may also
disrupt language. These syndromes typically develop gradually,
over weeks or months, unlike the abrupt onset of aphasia
secondary to stroke or head injury. As in head injuries, there
are remote effects related to mass effect and edema, in addition
Aphasia 179
to the focal damage. For this reason, these disorders have been
less useful than stroke in aphasia research. Focal brain damage
from encephalitis and other infections also causes aphasia. The
Herpes simplex virus causes encephalitis, with a predilection
for the orbital frontal lobes and temporal lobes. Patients may
present with fluent aphasia, as well as confusion, memory loss,
seizures, and fever. Patients undergoing epilepsy surgery have
also provided new knowledge of the language areas of the
brain, as discussed earlier.
In addition to the language disorders associated with focal
brain lesions, generalized brain diseases produce language dis-
orders. Aphasia accompanies such diffuse syndromes as ence-
phalopathies, acute confusional states, dementias, and chronic
syndromes of memory and cognitive loss. These syndromes
will be discussed after the focal aphasias.
Evaluation of Aphasia
Bedside Evaluation
Language functions are tested informally at the bedside to arrive
at a preliminary syndrome diagnosis of aphasia. Deliberate
testing is carried out for spontaneous and automatic speech,
naming, repetition, comprehension, reading, and writing. Phy-
sicians correlate these language tests with other findings of the
neurological examination such as motor paralysis, visual field
defects, and sensory loss to arrive at a localization of the respon-
sible brain lesion. Brain imaging studies are then performed to
confirm the diagnosis.
Standard Aphasia Batteries
In addition to the bedside language evaluations carried out by
neurologists and other physicians, a number of standardized
aphasia tests are available. Most are used by speech/language
pathologists or neuropsychologists for documentation of the
type and severity of aphasic deficits and of progress during
recovery. Among the widely used batteries are the Boston
Diagnostic Aphasia Examination (BDAE), its Canadian adap-
tation the Western Aphasia Battery (WAB), and the Porch Index
of Communicative Ability (PICA). More specific language tests
have also been developed by cognitive psychologists, linguists,
and speech pathologists for research studies.
Classification of Aphasias
Aphasia experts have used a variety of terminologies and clas-
sification systems. This article utilizes the 'Boston' aphasia
classification, developed from classical neurology, as reflected
both in the bedside examination and in the BDAE and WAB.
Table 1 summarizes the language characteristics of the major
aphasia syndromes.
Broca’s Aphasia
Broca's aphasia was the first aphasia syndrome described, and
the first correlated with a specific brain region. The syndrome
involves reduced speech output, sometimes with complete
mutism or hesitant, struggling efforts to produce single words
or short phrases. Subjects utter the most important, meaning-
ful words of a sentence, leaving out the small grammatical
words. This speech pattern is called 'telegraphic speech' or
'agrammatism.' Patients struggle to recall names, often stutter-
ing on the initial phoneme ('tip-of-the-tongue' phenomenon).
Repetition is hesitant, like spontaneous speech. Auditory com-
prehension is intact for simple conversations, but detailed
testing reveals deficits, especially for complex grammatical
relationships, which also cause difficulty in expressive speech.
Reading is often less preserved than auditory comprehension.
Writing is influenced by the usual association with right-sided
paralysis, or hemiparesis, but patients with Broca's aphasia usu-
ally cannot write well even with the nonparalyzed left hand,
indicating the presence of an expressive language disorder. Inter-
estingly, studies have found that patients write better with the
paralyzed right arm, aided by a mechanical device resembling
a skate. This technique removes the necessity of transmission
of language information to right hemisphere motor centers.
The lesions of Broca's aphasia traditionally involve the
inferior frontal gyrus, just anterior to the face area of the
motor cortex (Figure 1). Small lesions in this area are asso-
ciated with excellent recovery of language function. Larger
lesions, involving the frontoparietal cortices, produce an early
global aphasia, or loss of all language function, followed by
gradual evolution toward Broca's aphasia, or 'mixed nonfluent
aphasia.' Studies by the Boston VA group of CT scans in
patients with nonfluent aphasia have established three lesion
patterns: (1) lesions restricted to the lower motor cortex of the
precentral gyrus produce only dysarthria and mild expressive
disturbance; (2) those involving the more anterior frontal
operculum (Brodmann areas 44 and 45) result in difficulty
initiating speech without true language difficulty (see discus-
sion of 'aphemia,' below); and (3) lesions combining these
two lesion sites, plus the subcortical white matter and periven-
tricular white matter cause deficits of Broca's aphasia. These
additional aphasic disturbances may result from interruption
Table 1 Characteristics of aphasic: syndromes
Syndrome
Fluency
Naming
Repetition
Comprehension
Reading
Writing
Broca
—
—
_
+
:!:
—
Wernicke
+
-
-
-
-
-I-
Global
-
-
-
-
-
-
Conduction
+
±
-
+
+
Anomic
+
-
+
+
+
+
TCM
-
+
+
■+,.
+
±
TCS
+
-
-I-
-
-
-
Isolation
-
-
+
-
-
-
1 80 Aphasia
of pathways from the posterior language areas. Damage to
two subcortical areas, the rostral subcallosal fasciculus deep
to Broca's area and the periventricular white matter adjacent to
the body of the lateral ventricle, appears necessary for the
permanent loss of functional expressive speech (Figure 2).
Aphemia
The term 'aphemia' was Broca's original choice for the name of
his nonfluent aphasia syndrome, but the term 'aphasia,' pro-
moted by Trousseau, was adopted instead. Aphemia was then
Parietal
Frontal lobe
Figure 1 A drawing of the lateral surface of the left hemisphere of the
brain. The approximate locations are given for Broca’s area in the left
inferior frontal convolution, Wernicke’s area in the left superior temporal
convolution, and the angular and supramarginal gyri of the inferior
parietal lobule.
Figure 2 MRI scan from a patient with Broca’s aphasia immediately
after stroke. MRI images are displayed in the standard orientation used in
radiology, with the left side of the brain on the right side of the image. This
is a diffusion-weighted image, showing an acute, left frontal infarction.
applied to a more restricted, often transitory syndrome of
muteness or nonfluent speech with difficulty of initiation of
speech, dysarthria, and phoneme substitutions. Repetition is
impaired in a similar manner, and to a similar degree, as sponta-
neous speech. Both auditory and reading comprehension are
preserved, and writing is usually much superior to speech.
Most patients have right facial weakness, and some have an
associated right hemiparesis. Aphemia is more closely related
to the motor speech disorders, apraxia of speech and dysarthria,
than to aphasia. Some authorities have even equated it to apraxia
of speech, a disorder in which hesitancy of initiation and
inconsistent substitutions of phonemes, are also characteristic.
Aphemia is associated with lesions of the lower motor cortex for
the face in the precentral gyrus, with extension in some cases
into the inferior frontal gyrus and underlying white matter.
Wernicke’s Aphasia
In contrast to the nonfluent speech of the patient with Broca's
aphasia, the Wernicke's aphasic speaks fluently, with abnormal
language content. Spontaneous speech contains empty phrases,
circumlocutions, neologisms (new or nonexistent words), and
errors reflecting sound substitutions (literal or phonemic para-
phasic errors, e.g., 'poon' for 'spoon') and word substitutions
(verbal or semantic paraphasic errors, e.g., 'fork' for 'spoon').
Sometimes no meaning at all is conveyed, amounting to 'jargon
speech.' To a nonnative speaker of English, however, the fluent
speech productions sound normal. Naming often produces
paraphasic utterances that have no obvious relationship to the
target word, and repetition is also disturbed. Comprehension
of spoken language is severely impaired. Reading is typically
affected in the same way as auditory comprehension, but occa-
sional cases show interesting dissociations between compre-
hension of auditory versus printed language. Spared reading
comprehension may be important in providing a means of
communication with the patient. Writing is often well-formed,
since patients with Wernicke's aphasia usually have no hemipar-
esis, but the written productions share the same lack of mean-
ingful words as the speech output. Written language productions
also contain misspellings; patients with mild Wernicke's aphasia
may be detected most easily by analysis of writing.
Patients with Wernicke's aphasia typically have no motor
or sensory deficits; some have a partial or complete loss of
vision in the right visual field. As Wernicke originally demon-
strated in 1 874, the lesions of patients with Wernicke's aphasia
typically involve the posterior portion of the left superior
temporal gyrus, the traditional 'Wernicke's area.' Studies by
Naeser and colleagues have suggested that damage to a major
part of Wernicke's area is necessary for lasting loss of compre-
hension, while more temporary loss of comprehension can be
seen with lesions elsewhere in the temporal lobe or in the
inferior parietal lobule. Some patients with lesions restricted
to the temporal lobe may have spared reading ability in the
presence of severe auditory comprehension deficits, whereas
those with lesions primarily affecting the inferior parietal lob-
ule may have disproportionate reading difficulty (Figure 3).
Pure word deafness
Pure word deafness is a rare syndrome involving the inability
to understand or repeat spoken language, in the absence of
Aphasia 181
Figure 3 MRI scan, diffusion weighted image, showing an acute stroke
in the left temporal lobe, resulting in Wernicke’s aphasia.
other language difficulty. The patient is not deaf to pure tones
or to meaningful nonverbal noises such as animal cries or the
ringing of a bell. Classically, the deficit occurs with bilateral
temporal lesions. By the 'disconnection' theory, these lesions
disconnect the left hemisphere Wernicke's area from both
primary auditory cortices (Heschl's gyrus in the superior tem-
poral gyrus of each hemisphere), such that sounds can be
heard but not analyzed by the receptive language cortex in
Wernicke's area. Destruction of the auditory cortices them-
selves causes cortical deafness, or complex auditory disorders
referred to as 'auditory agnosia.' Patients with cortical lesions
are usually not completely deaf; some cannot recognize either
words or nonverbal sounds. Cases of unilateral, left temporal
stroke have also been reported to develop pure word deafness.
As mentioned earlier, some patients with Wernicke's aphasia
and damage predominantly affecting the temporal lobe may
have disproportionate involvement of auditory comprehen-
sion, as compared to reading comprehension, thus resembling
the deficit of pure word deafness. In fact, pure word deafness is
often not entirely 'pure,' in that mild paraphasic speech dys-
function is usually associated.
Global Aphasia
Global aphasia may be thought of as the sum of Broca's and
Wernicke's aphasia. It is a loss of all six of the commonly tested
language functions: spontaneous speech is mute or severely
nonfluent, naming is reduced, repetition and comprehension
are severely impaired, and the patient cannot read or write.
Most such patients have large areas of damage in the left
hemisphere and profound associated neurological deficits
such as right hemiplegia, right-sided sensory loss, and right
hemianopsia. Occasionally, strokes can damage both Broca's
and Wernicke's areas, without massive involvement of the
motor system, and global aphasia without hemiparesis can
result. When lesser degrees of deficit are seen across all lan-
guage functions, the syndrome is called 'mixed aphasia.'
Conduction Aphasia
Conduction aphasia is an unusual but interesting language
syndrome, in which repetition is affected, out of proportion
to all other aphasic deficits. Patients speak relatively fluently,
though some make frequent literal paraphasic errors and pause
frequently for self-correction, giving the speech output a hesi-
tant, choppy pattern. Naming is often impaired, but auditory
comprehension is intact. Reading aloud and copying may be
affected in a way similar to repetition.
Wernicke originally postulated that conduction aphasia
might be caused by a lesion that did not damage either
Wernicke's or Broca's area, but disrupted the connections be-
tween the two areas, namely the arcuate fasciculus. Geschwind
rediscovered this explanation of conduction aphasia as a 'dis-
connection syndrome,' in which the language impairment is
caused by disconnection between cortical language centers, in
the absence of damage to the centers themselves. In fact, most
cases of conduction aphasia are related to stroke and do have
evidence of cortical damage, either in the left temporal lobe,
without complete destruction of Wernicke's area, or in the left
inferior parietal lobule. Recent research has indicated that the
arcuate fasciculus, originating in Wernicke's area, terminates
not in Broca's area, but in the precentral gyrus, or motor cortex,
shedding doubt on the importance of arcuate fasciculus dam-
age to the syndrome of conduction aphasia. While the discon-
nection theory of conduction aphasia has remained popular,
some investigators have favored a disorder of auditory proces-
sing, with some evidence favoring the supramarginal gyrus as
a site where auditory inputs are processed for transmission
to Broca's area, thus enabling repetition. Others have analyzed
the deficit as a disorder of immediate memory for word
sequences, such that phrases cannot be maintained in the imme-
diate memory store long enough to allow repetition. These
theories sound very different, but they are difficult to separate,
in terms of practical details of language deficits.
Anomic Aphasia
As its name implies, anomic aphasia involves the selective loss
of naming ability, out of proportion to other language deficits.
Patients speak fluently, though with word-finding pauses and
circumlocutions. Repetition, auditory comprehension, reading,
and writing are intact. This syndrome is less well correlated with
focal pathology than the other classical aphasia syndromes,
though anomic aphasia can occur with focal lesions of the left
temporal or inferior parietal region. Anomic aphasia may also
be the last stage in the recovery process from other aphasia
types, including both Broca's and Wernicke's aphasia. Anomic
aphasia is also common in the early stages of dementing con-
ditions such as Alzheimer's disease.
Transcortical Aphasias
The transcortical aphasias were named to describe the sparing
of the perisylvian language circuit, and thereby of repetition.
The word 'transcortical,' as used by Lichtheim, implies that the
damage is not to the language cortex itself, but to other areas of
the brain which project onto the language cortex. Lichtheim
referred to these other cortical areas as the 'area of concepts.'
1 82 Aphasia
Current neuroanatomy texts do not list an area of concepts, but
this would be part of the association cortex of the frontal,
temporal, and parietal lobes. The three transcortical aphasia
syndromes share the sparing of repetition, despite marked
aphasic deficits.
Transcortical motor aphasia is similar to Broca's aphasia, in
that the patients speak nonfluently, often with long pauses
before utterances and single word or whispered replies to
questions. Patients may remain mute when asked questions
requiring sentence length responses, but if a single word will
suffice, the patient may answer accurately. Auditory compre-
hension is usually quite good. These patients, unlike Broca's
aphasics, can repeat long sentences, even with complex gram-
matical constructions. The lesions of transcortical motor apha-
sia spare Broca's area itself but involve adjacent portions of
the left frontal cortex, including medial frontal areas such as
the supplemental motor cortex, or the frontal subcortical white
matter. These lesions all fall within the distribution of the
anterior cerebral artery, a relatively uncommon locus of stroke,
as compared to the middle cerebral artery, parts of whose
territory are affected in Broca's, Wernicke's, global, and con-
duction aphasias.
Transcortical sensory aphasia resembles Wernicke's apha-
sia, in terms of fluent, paraphasic speech and poor auditory
comprehension, but these patients can repeat normally. The
lesions are located posteriorly, in the left temporo-occipital
confluence. This syndrome is seen uncommonly in stroke
patients, but it occurs in patients with Alzheimer's disease, in
later stages of the disease. In one electrical stimulation study
in normal subjects, stimulation of adjacent groups of neurons
in the temporal lobe could produce either Wernicke's or trans-
cortical sensory aphasia.
Mixed transcortical aphasia, also called the 'syndrome of
the isolation of the speech area,' is a syndrome resembling
global aphasia, except for the sparing of repetition, which
may even be excessive or palilalic. These patients cannot
speak spontaneously or comprehend spoken or written lan-
guage, but they can repeat fluently. Some patients have a
tendency to complete partial utterances, as in poetry lines or
common idioms (if the examiner says 'roses are red,' the
patient may repeat 'roses are red, violets are blue'). Geschwind
and colleagues reported the case of a young woman who had
extensive, bihemispheral damage from carbon monoxide poi-
soning, sparing the perisylvian language circuit. This patient
could not speak in any meaningful way, could not compre-
hend conversation, but she could repeat long sentences and
could even memorize lyrics to songs which became popular
after her illness implying that the medial temporal lobe mem-
ory circuit was also spared. Mixed transcortical aphasia has also
been reported in patients with severe dementia.
Subcortical Aphasias
Unlike all of the aphasia syndromes described up to this point,
subcortical aphasias are defined more by the location of the
responsible brain lesion than by the language characteristics of
the aphasia. While aphasia is usually thought of as a cortical
phenomenon, several patterns of aphasia with subcortical
lesions have been described, all within recent decades since
brain imaging became widely available. First, lesions of the left
thalamus have been associated with fluent aphasia, usually
with better comprehension and repetition as compared to
Wernicke's aphasia. A 'dichotomous state' has been described,
in which the patient may speak clearly while awake, but drift
into incomprehensible, paraphasic speech when sleepy. The
thalamus has a role in activating the language cortex; Luria
referred to thalamic aphasia as a 'quasiaphasic disturbance
of vigilance.' 'Thalamic aphasia,' as it is called, was first
described in patients with thalamic hemorrhage, but many
cases have also been reported with ischemic sttokes involving
the thalamus.
Ischemic lesions of the basal ganglia and subcortical white
matter also cause aphasia. Lesions of the head of the caudate,
anterior limb of the internal capsule, and anterior putamen
produce a nonfluent aphasia with dysarthria, termed the 'ante-
rior subcortical aphasia syndrome.' The syndrome bears some
resemblance to Broca's aphasia, but these patients usually have
longer phrase length and better comprehension and repetition
than patients with Broca's aphasia from cortical lesions. More
extensive damage in the subcortical structures can produce
other aphasic syndromes; lesions which extend posteriorly in
the basal ganglia and laterally into the deep temporal lobe
white matter can cause syndromes resembling global and
Wernicke's aphasia. Similar aphasias are seen with hemor-
rhages into the basal ganglia, but deficits tend to be more
severe because of pressure effects on surrounding structures
and on the overlying language cortex.
Alexias
Alexia is an acquired disorder of reading secondary to brain
disease. Since reading is a language function, alexia falls under
the definition of aphasia as an acquired disorder of language.
The alexias can be either largely restricted to reading, or they
may be part of an aphasic syndrome.
Pure alexia with agraphia is an acquired illiteracy, a syn-
drome in which patients cannot read or write, but they can
speak, repeat, name, and comprehend spoken language. In
fact, most patients with alexia with agraphia do not have
such a 'pure' syndrome; they often have a mild fluent aphasia,
and the syndrome overlaps with Wernicke's aphasia. The
lesion involves the inferior parietal region, particularly
the angular gyrus. This type of alexia is also referred to as
'central alexia.'
Pure alexia without agraphia is, in effect, a linguistic blind-
folding. Patients cannot read, but they speak normally and can
understand spoken language, even when words are dictated in
spelled form. These patients can write, but they cannot read
their own written productions, a striking feature of the disor-
der. Naming for objects is usually preserved, but patients may
be unable to name colors. They have not lost the concept of
color names, as shown by their ability to name the usual colors
of common objects such as bananas or apples, but they cannot
name a perceived color. In addition, they usually have short-
term memory loss and a right visual field defect. Pure alexia
is associated with a lesion of the left occipital cortex, within the
territory of the left posterior cerebral artery, usually also involv-
ing the splenium of the corpus callosum, and sometimes the
medial temporal cortex as well. Pure alexia without agraphia is
a classical disconnection syndrome, as originally postulated by
Aphasia 1 83
the French physician Dejerine in 1891, and later rediscovered
by Norman Geschwind. By the disconnection theory, the non-
damaged right occipital cortex can subserve vision in the left
visual field, but visual language and color information cannot
be transmitted properly to the left hemisphere language cortex.
Like the disconnection theory of conduction aphasia, this
model does not account for all of the complexity of the pure
alexia syndrome. Some investigators have pointed to a distur-
bance of visual immediate memory as a factor in pure alexia; if
the patient cannot keep a string of letters in mind, then the
letters cannot be processed into a word. In fact, patients with
pure alexia often can read letters one by one, but they cannot
read a word at a glance. The ability to recognize words
(or 'word forms') at a glance is likely a left occipital lobe
function. Alternate names for the syndrome are 'letter-by-letter'
alexia and posterior or occipital alexia.
Many patients with aphasia have associated reading dis-
turbance, termed 'aphasic alexia.' Benson called the alexia of
Broca's aphasia the 'third alexia' following the syndromes of
alexia with and without agraphia. It is also called 'frontal alexia.'
This type of alexia can vary from a mild alexia related to difficult
syntactic material to a severe alexia, in which patients can read
only by recognition of a few familiar words. Wernicke's apha-
sics often have deficits in reading and writing similar to the
syndrome of alexia with agraphia, though patients have been
described with more partial, selective deficits.
In recent years, neurolinguists and cognitive psychologists
such as Marshall, Newcombe, and Coltheart have divided alex-
ias according to the specific stages in the reading process which
go awry (Figure 4). Four patterns of alexia (or 'dyslexia,' as
these investigators often call it) have been recognized: letter-
by-letter, deep, phonological, and surface dyslexia (Table 2).
Printed
1, 2,3
Spoken output
Figure 4 Three separate routes to reading. (1) The grapheme-
phoneme conversion system used in learning reading. This is the only
route available in surface dyslexia. (2) Reading only from recognizing
the meaning of familiar words, the only route available in deep dyslexia.
(3) A direct conversion of words to sounds, available along with route 2 in
phonological dyslexia.
Letter-by-letter dyslexia is equivalent to pure alexia without
agraphia, while the other three patterns are usually seen with
aphasia.
Deep dyslexia involves several distinctive features: intact
reading of familiar words, inability to read nonsense syllables
or nonwords, semantic and visual errors in reading aloud, and
pronounced effects of word frequency, word imageability, and
word class. Imageable words, those which denote an object or
animal that can be seen by the mind's eye, are read more
accurately than abstract words, while nouns and verbs are read
better than adjectives, adverbs, and prepositions. Word reading
is not much affected by word length or by regularity of spelling;
for example, patients have been described who can read 'be' but
not 'B,' or 'ambulance' but not 'am.' Examples of semantic errors
are 'jail' for 'prison' and 'soccer' for 'football'; examples of visual
errors are 'perform' for 'perfume' and 'banish' for 'blush.' Deep
dyslexia involves the loss of reading by the 'phonological,'
or grapheme-to-phoneme conversion route, with residual
reading involving only recognition of familiar words (route 2
in Figure 4). Most cases have severe aphasia, with extensive left
frontal and parietal damage. The residual reading of patients
with deep dyslexia may involve the right hemisphere.
Phonological dyslexia is similar to deep dyslexia, except
that single content words are read in a nearly normal fashion,
and semantic errors are rare. As in deep dyslexia, reading of
nonwords is poor. Patients sometimes read words without
access to meaning. The syndrome of phonological dyslexia,
like deep dyslexia, involves the loss of grapheme-to-phoneme
conversion. In deep dyslexia, only words whose meaning is
recognized can be read aloud; only a semantics-to-phonology
reading route can operate. In phonological dyslexia, words
can also be read via a lexical-phonological route (route 3 in
Figure 4), in which printed words can be recognized and
transferred into their sounds, without access to semantics, or
meaning. Only the nonlexical-phonological route, by which
nonwords must be read, is inoperative. Most patients with
phonological alexia have severe, nonfluent aphasia.
In surface dyslexia, subjects can read laboriously by
grapheme-phoneme conversion, but they cannot recognize
words at a glance. They can read nonsense syllables, but not
words of irregular spelling, such as 'yacht' or 'colonel.' Their
errors tend to be phonological rather than semantic or visual;
an example would be to pronounce 'pint' as if it rhymed with
Tint.' Reading in these patients is accomplished without access
to semantics until the word is pronounced. Individual gra-
phemes are sounded out by the most common way they are
Table 2 Characteristics of alexic syndromes
Characteristic
Deep
Phonological
Letter-
Surface
dyslexia
dyslexia
by-letter
dyslexia
dyslexia
Nonwords
—
—
—
+
Word class effect
+
+
-
-
Imageability
+
-
-
-
Word length
-
-
+
+
Regular spelling
-
-
-
+
Source: Patterson KE (1981) Neuropsychological approaches to the study of reading.
The British Journal of Psychiatry 72: 151-174.
1 84 Aphasia
pronounced in the language, producing errors for both irregu-
larly spelled words and regular words which have atypical
pronunciations. For example, a surface dyslexic might pro-
nounce 'rough' and 'though' alike. Surface dyslexia represents
a syndrome in which only one route of reading, the grapheme-
phoneme route, is operative (route 1 in Figure 4). Patients
cannot read words without pronouncing their component gra-
phemes. This syndrome is also seen in patients with aphasia,
and also in patients with primary progressive aphasia (PPA), or
the frontotemporal dementia (FTD) syndrome (see below).
Agraphias
Like reading, writing may be affected either in isolation ('pure
agraphia') or in association with aphasia ('aphasic agraphia').
In addition, writing can be deranged by motor disorders, by
visuospatial or constructional impairments, and by apraxia, a
disorder of learned motor activities. Isolated agraphia has been
described with left frontal and occasionally left parietal lesions.
In the past, a cortical writing center called 'Exner's area' was
postulated, just above Broca's area. The existence of a specific
writing area has not been supported, however, either by stroke
cases or by electrical stimulation studies; stimulation of Broca's
area interferes with writing, while excitation of other frontal
regions does not produce agraphia without disturbing speech.
Agraphias have also been analyzed in ways similar to the
alexias. A syndrome of 'phonological agraphia' has been de-
scribed, in which patients cannot convert speech sounds (pho-
nemes) into letters (graphemes). These patients cannot write
pronounceable nonsense words but can write familiar words,
even when irregular in spelling. Reported cases have been rela-
tively few, with limited neuroanatomic data, but some cases have
had lesions of the supramarginal gyrus. Most patients have had
associated aphasia, but of varied types. 'Deep' dysgraphia is
similar to phonological agraphia, but as in deep alexia, mainly
familiar nouns and verbs can be produced, with a marked word
class effect.
A 'lexical agraphia' syndrome, also called 'surface agraphia,'
has also been described. Patients can write regularly spelled
words and pronounceable nonsense words, but not irregularly
spelled words. These patients have intact phoneme-grapheme
conversion but cannot write using a whole word, or 'lexical'
strategy. Lesions have been reported in the posterior angular
gyrus and occipitotemporal gyrus, but other cases have had
more anterior localizations.
Recovery of Aphasia
Patients with aphasia from acute disorders such as stroke gen-
erally show spontaneous improvement over the first several
months. The traditional opinion in rehabilitation is that this
improvement is maximal during the first 3 months, but, as
Sarno has pointed out, patients with severe, global aphasia
may actually improve more in the second 6 months than the
first. The aphasia type often changes during this improve-
ment; global aphasia often evolves into Broca's aphasia, and
Wernicke's aphasia may evolve into conduction or anomic
aphasia. The neuroanatomical correlates of language recovery
appear to involve the taking over of language functions by
adjacent cortical areas of the left hemisphere, and the right
hemisphere may also participate. Studies of activation of lan-
guage areas by PET, SPECT, and fMRI techniques promise to
advance our understanding of the neuroanatomy of language
recovery. To summarize a growing and not always congruent
literature, if left hemisphere language cortex can resume lan-
guage functioning, recovery is excellent, but if centers in the
right hemisphere take over, the recovery is incomplete. Recov-
ery using the contralateral hemisphere to assume lost functions
may be a 'second best' strategy.
Speech therapy, carried out by trained speech/language
pathologists, seeks to facilitate improvement in language func-
tion by a variety of techniques. Repetitive exercises are carried
out to promote better articulation, and matching vocabulary
tasks are designed to improve receptive language. Other tech-
niques in speech therapy include melodic intonation therapy,
which attempts to involve the right hemisphere in speech
production through the use of melody, and computer tech-
niques originally developed for primate communication.
Patients who cannot speak can learn to produce major nouns
and verbs and associate them into simple phrases or sentences.
A variety of augmentative devices are also available to help
make language outputs available to others via printers or
voice simulators. Therapy techniques are now chosen based
on the patient's aphasia test results, and some investigators are
using brain imaging studies to predict which techniques are
likely to succeed. While many speech therapy techniques have
been developed empirically, large, randomized ttials have
clearly indicated that patients who undergo formal speech
therapy recover more communicative function than untreated
patients.
New techniques in aphasia therapy include drugs, such as
amphetamines or beta blockers, both of which have been
shown to improve fluency and naming, and dopaminergic
agents such as bromocriptine, which may improve speech
output in nonfluent aphasias. All the evidence of the benefit
from these drugs has come from small studies and series, and
there is a lack of strong evidence of clinical effectiveness.
Another avenue for aphasia therapy involves brain stimulation
techniques. Transcranial magnetic stimulation is a noninvasive
way of stimulating the language cortex, and some preliminary
studies have indicated benefit. Again, large clinical trials are
lacking. Direct, transcranial electrical stimulation, or surgical
placement of electtodes in the brain or subdural space has also
been attempted, but these procedures are more invasive and
will require much better evidence of benefit to justify the risks.
Transplantation of stem cells has also been attempted in
patients with poststroke aphasia, though this research, too, is
at an early stage of development. These new techniques carry
new promise for patients with aphasia secondary to strokes
and other causes of brain damage.
Miscellaneous Topics in Aphasia
Aphasia in Polyglots
Aphasia in patients who speak more than one language ('poly-
glots') has been the subject of considerable interest. In 1895,
Pitres postulated that the language used most before the onset
of aphasia would be the first to be recovered. In contemporary
Aphasia 1 85
terms, the language in which the patient works every day,
and in which hospital personnel and therapists speak, is likely
to be recovered faster and better than another language the
patient has learned, or even grew up speaking. Ribot, in 1906,
predicted that the first-learned or native language would be
the most resistant to later brain injury. These two rules are
now known as Pitres' and Ribot's laws. Other investigators
have considered both the time of acquisition and the relative
skill level in each language. Thus, Canadians who acquired
French and English simultaneously in growing up ('com-
pound bilinguals') have more congruent language involve-
ment than those who learned one language later than the
other ('coordinate bilinguals'). In recent studies, coordinate
bilinguals have even developed different aphasia syndromes
in two languages; for example an Israeli patient was globally
aphasic in Hebrew but had only a mild anomic aphasia in
Russian. In general, the language used more extensively before
the brain injury, as Pitres originally postulated, appears to be
less disturbed than the less used language, regardless of which
was learned first.
Aphasia in Acute Confusional States and Dementias
Patients who develop acute delirium or confusion frequently
utter meaningless or nonsensical phrases. In most cases this
involves abnormal thought, or language content, rather than
abnormal language, resembling a psychosis. In some cases,
however, delirious patients produce fluent, paraphasic speech
and even neologisms. Naming is frequently disturbed in acute
confusional states. Comprehension is at least partially pre-
served, if the patient's attention and memory can be entrained
to the task. Many delirious patients have abnormal writing. The
diagnosis of delirium rests on a general confusional state, with
impaired attention, reasoning, and often hallucinations and
delusional thinking as well as agitation or drowsiness and
autonomic signs such as tremor, sweating, tachycardia, hyper-
tension, and fever. All of these 'positive' cognitive symptoms
such as agitation, delusional thinking, and hallucinations, and
also the autonomic signs, are much less prominent in dement-
ing illnesses, except in advanced cases.
Patients with dementing diseases such as Alzheimer's
disease frequently demonstrate abnormal language functions.
The dementia of Alzheimer's disease usually begins with
memory loss, and among the earliest symptoms is a loss of
memory for names. Recollection of proper names and rapid
generation of a series of names are very sensitive to aging and
dementia. For example, patients with early dementia have
marked difficulty with the 'animal naming' subtest of the
BDAE, which requires the subject to name as many animals
as possible in one minute. Articulation, repetition, and audi-
tory comprehension tend to remain intact into later stages of
dementia. At an early stage, an Alzheimer's disease patient
often shows the profile of anomic aphasia. As the disease
advances, reading and writing deteriorate, and auditory com-
prehension begins to decline. Repetition and articulatory flu-
ency remain preserved, and the patient may then show the
language profile of transcortical sensory aphasia. The content
of expressive speech is also severely impoverished, devoid of
abstract content. Ultimately, language expression becomes
limited almost to statements of biological need. Some patients
become mute in late stages, resembling global aphasia, while
others repeat fluently.
Alzheimer's disease is a 'diffuse' degeneration of the cere-
bral cortex, but pathological studies indicate early involvement
of the association cortex, particularly of the parietal and later
the frontal lobes. Medial temporal structures such as the hip-
pocampus, which are important to memory function, and
medial forebrain structures such as the nucleus basalis, also
show neuronal loss and senile plaques.
Other dementing diseases also produce language disorders.
After Alzheimer's disease, the most common cause of dementia
is multiinfarct or vascular dementia. The specific features
depend on the size and location of the cerebral infarcts, but
many patients manifest aphasia.
Pick's disease was described in the nineteenth century as a
lobar atrophy involving predominantly the frontal and tempo-
ral cortices, with microscopic stains showing intraneuronal,
silver-staining inclusions called 'Pick bodies.' Many cases with
similar clinical and neuroradiological features lack Pick bodies,
and the newer term for this family of focal neurodegenerative
disorders is 'FTD' or 'frontotemporal lobar degeneration.'
Some cases begin with 'frontal' syndromes of behavioral
abnormality, others have isolated aphasia at onset. What
was described in North America as 'PPA' is now thought to
be a variant of the syndrome of FTD, described in Europe.
Some cases of PPA have been followed for several years, with
steadily worsening language function, usually with nonfluent,
or Broca-like aphasia, but no significant memory loss or
generalized dementia. Brain imaging studies show only a
lobar atrophy. PET studies have consistently shown left frontal
and/or temporal hypometabolism, with more advanced cases
showing more widespread metabolic changes. The pathology is
usually of lobar atrophy, without the typical changes of
Alzheimer's disease, but with neuronal loss and glial scarring.
A variant of the syndrome is 'semantic dementia,' in which
patients speak fluently but lose their ability to name or even to
comprehend the meaning of names given to them. This syn-
drome can be seen in either FTD or Alzheimer's disease. A great
deal of research has been published recently on this family of
conditions. Two different gene mutations have been found in
familial cases of FTD and PPA, both entirely separate from
the genes associated with Alzheimer's disease. Another disease
that can present with progressive aphasia is corticobasal degen-
eration, a variant of Parkinson's disease also associated with
mutations in the tau protein, one of the two gene mutations
associated with FTD.
Creutzfeldt- Jakob disease is a more rapidly progressive
dementia, often with myoclonus and seizures, and often a
fatal course within several months. Numerous variants of the
syndrome have been described. Aphasia has been documented
as a presenting symptom of Creutzfeldt-Jakob disease, but the
development of a rapidly progressive dementia with myoclo-
nus and seizures usually makes the diagnosis clear.
Language and the Right Hemisphere
While the right hemisphere has traditionally been called the
'minor' hemisphere because of its lack of language dominance,
research has shown that patients with right hemisphere disease
have significant problems with verbal communication. While
1 86 Aphasia
they can produce and understand appropriate words and sen-
tences, patients with right hemisphere lesions speak monoto-
nously, lacking the emotional tone or 'prosody' which makes
normal language colorful. These patients lose the ability to
understand emotional tone, connotation, irony, sarcasm, satire,
and humor, responding only to the literal meaning of the words;
they understand what is said, but not how it is said. The right
hemisphere may have a similar organization to the left in terms
of speech prosody, the emotional intonation, and emphasis
present in spoken language; frontal lesions disturb production
of prosodic speech, while temporal lesions disturb comprehen-
sion of emotional tone. Right hemisphere stroke patients have
difficulty even with nonemotional aspects of prosody, such as
placement of emphasis within sentences or inflections, as in the
difference between a statement and a question. Despite their
intact language skills, these patients are at a great disadvantage
in normal human communication.
Aside from these effects of right hemisphere disease, the
right hemisphere appears to have considerable linguistic abil-
ity. Patients with surgical commissurotomy, or section of the
corpus callosum, are able to recognize words flashed in the left
visual field, or presented in tactile fashion to the left hand.
They can perform simple matching tests between words or
short phrases which are semantically related. The mystery
about these abilities is the failure of patients with left hemi-
sphere lesions and global aphasia to perform similar tasks,
despite the presence of an intact right hemisphere. It is possible
that the corpus callosum contains fiber tracts which inhibit the
right hemisphere from expressing its full language capability. If
such systems could be therapeutically disrupted, the recovery
of aphasia might be significantly enhanced.
Conclusion
The subject of aphasia is a rich field. Language was the first of
the higher brain functions to be analyzed in terms of specific
localization of brain lesions, and it remains the best studied
and understood of all of the cognitive functions. Increased
understanding of both normal language and aphasia is likely
to emerge from new studies combining specific language test-
ing measures with precise neuroanatomic techniques such as
MRI and functional activation methodologies such as PET and
fMRI. The result should be not only a better knowledge of
language and the brain, but also improved therapies for
patients afflicted with aphasia.
Further Reading
Albert ML, Bachman DL, Morgan A, and Helm-Estabrooks N (1988) Pharmacotherapy
for aphasia. Neurology 38: 877-879.
Benson DF and Ardila A [WS) Aphasia. A Clinical Perspective, pp. 1-441. New York:
Oxford University Press.
Bernal B and Ardila A (2009) The role of the arcuate fasciculus in conduction aphasia.
Brain 132: 2309-2316; Epub 2009 Aug 18.
Catani M, Jones DK, and ffytche DH (2005) Perisylvian language networks of the human
brain. Annals of Neurology 57: 8-16.
Damasio AR (1992) Aphasia. New England Journal of Medicine 326: 531-539.
Davies RR, Hodges JR, and Krill JJ (2005) The pathological basis of semantic dementia.
Brain 128: 1984-1995.
Devlin JT and Watkins KE (2007) Stimulating language: Insights from TMS. Brain
130: 610-622.
Faber-Langendoen K, Morris JC, Knesevich JW, LaBarge E, Miller JP, and Berg L
(1988) Aphasia in senile dementia of the Alzheimer type. Annals of Neurology
23: 365-370.
Hillis AE (2007) Aphasia: Progress in the last quarter of a century. Neurology 69:
200-213.
Josephs KA (2008) Frontotemporal dementia and related disorders: Deciphering the
enigma. Annals of Neurology 64: 4-14.
Kirshner HS (1995) Handbook of Neurological Speech and Language Disorders,
pp. 1-532. New York: Marcel Dekker.
Kirshner HS (2002) Behavioral Neurology: Practical Science of Mind and Brain,
pp. 1-474. Boston, MA: Butterworth Heinemann.
Kirshner HS (2008) Aphasia. In: Bradley W, et al. (ed.) Neurology in Clinical Practice,
5th edn., Chapter 12A, pp. 141-160. Boston, MA: Butterworth Heinemann.
Kreisler A, Godefroy 0, Delmaire C, et al. (2000) The anatomy of aphasia revisited.
Neurology 54: 1117-1123.
LaPointe LL (2005) Aphasia and Related Neurogenic Language Disorders, 3rd edn.,
pp. 1-274. New York: Thieme.
Mesulam M-M (2003) Primary progressive aphasia - A language-based dementia.
New England Journal of Medicine 349: 1535-1542.
Naeser MA, Palumbo CL, Helm-Estabrooks N, Stiassny-Eder D, and Albert ML (1989)
Severe non fluency in aphasia. Role of the medial subcallosal fasciculus and other
white matter pathways in recovery of spontaneous speech. Brain 112: 1-38.
Neary D, Snowden J, and Mann D (2005) Frontotemporal dementia. Lancet Neurology
4: 771-780.
Posner Ml, Petersen SE, Fox PT, and Raichle ME (1988) Localization of cognitive
operations in the human brain. Science 240: 1627-1631.
Sarno MT (19^) Acquired Aphasia, 2nd edn. San Diego, CA: Academic Press.
Small SL and Llano DA (2009) Biological approaches to aphasia treatment. Current
Neurology and Neuroscience Reports 9: 443-450.
Thompson CK and den Ouden DB (2008) Neuroimaging and recovery of language in
aphasia. Current Neurology and Neuroscience Repods 8: 475-483.
Weintraub S, Rubin NP, and Mesulam M (1990) Primary progressive aphasia:
Longitudinal course, neuropsychological profile, and language features. Archives of
Neurology 47: 1329-1335.
Wertz RT, Weiss DG, Aten LJ, et al. (1986) Comparison of clinic, home and deferred
language treatment for aphasia: A VA cooperative study. Archives of Neurology
43: 653-658.
See also: Agraphia and Alexia; Alzheimer's Disease; Handedness;
Language Development; Neuroimaging of Dementia; Sign Languages;
Syntax.
Appetite
N Geary, Institute of Food, Nutrition and Health, Schwerzenbach, Switzerland
© 2012 Elsevier Inc. All rights reserved.
Glossary
Anorexia Decreased appetite and eating caused by illness,
aversion, or abnormal interference with the normal
physiological controls of eating.
Appetite The subjective experiences and behaviors
associated with the urge to eat including food seeking and
meal initiation and with eating, including flavor hedonics
and satiation.
Body mass index A measure of normal or abnormal
adiposity equal to weight divided by height squared
(BMI, kgm~ 2 ); normal BMI is 18.5-25, overweight is
25-30, and obese is >30.
Flavor hedonics The component of food reward referring
to the pleasure or displeasure accompanying eating a
particular food and the associated increase or decrease
in selection and ingestion of that food; hedonic
reactions are responses of the brain, not intrinsic properties
of foods.
Flomeostasis The maintenance of crucial physiological
variables within relatively narrow ranges by active
physiological mechanisms.
Flunger A synonym for appetite; formerly often limited to
urges to eat elicited by physiological consequences of
nutrient depletion.
Meal An organized bout of eating which is the functional
unit of eating upon which physiological controls act.
Obesity A degree of excess adiposity associated with
markedly increased health risks, often defined as BMI > 30.
Palatability A synonym for flavor hedonics.
Postprandial satiety The subjective experiences and behaviors
associated with the lack of eating during intermeal intervals.
Satiation The subjective experiences and behaviors
produced during a meal by the ingestion of food that cause
the termination of the meal; the physiological mechanisms
of satiation and postprandial satiety are thought to be
partially independent.
Introduction
Scope
The physiology of eating is a crucially important problem
because of the epidemic proportions of obesity and the immense
societal costs of obesity-related health disorders. Thus, why so
many of us eat more than we need to and how this might be
controlled are pressing scientific questions. Psychiatric eating
disorders, although far less prevalent than obesity, are significant
causes of morbidity and mortality, and therefore are also impor-
tant basic research medical challenges. Unfortunately, the physi-
ology of eating is very poorly understood. A reflection of this
situation is that at present no truly effective physiologically
based medications for disordered eating are available. Such
medications are urgently needed to aid obese persons to diet or
to aid persons with eating disorders to overcome their afflictions.
My aim in this article is to introduce the physiology of
eating to students of physiology, neuroscience, and behavioral
science. Fundamental physiological knowledge is emphasized,
with some current areas of intense research interest highlighted.
Social and cognitive aspects are not included. Disordered eating,
as occurs in obesity and psychiatric eating disorders (anorexia
nervosa, bulimia nervosa, etc.) are very briefly introduced.
Some background in neuroscience and physiology, such
as could be easily obtained in basic texts, is assumed. Table 1
lists hormones involved in the control of eating (except those
discussed in connection with sodium appetite) together with
their abbreviations; Table 2 lists neurochemicals involved in the
control of eating, only some of which are discussed, with their
abbreviations. Because space requires that the review be incom-
plete, many mechanisms and mediators are not covered;
this includes several of the molecules listed in Tables 1 and 2.
The role of learning in the physiological control of eating
can hardly be overestimated. Although I cover this topic in a
separate section, several examples of how learning shapes physi-
ological controls are included. Finally, my discussion is organized
around a few concepts that, although in wide use, are not univer-
sal or have varying definitions in the physiological literature.
Therefore, I begin by describing my use of these concepts.
Concepts and Categories
The basic behavioral and subjective concepts used in various
scientific approaches to eating - for example, appetite, hunger,
palatability, satiety, and satiation - have no standard defini-
tions. These days, as physiological and psychological methodol-
ogies are growing ever closer, as exemplified by functional brain
imaging work, the semantic opportunities for miscommunica-
tion and confusion abound. The student of eating must keep in
mind what actually is measured, rather than the names used to
describe the measures.
The physiology of eating comprises: (1) the behaviors and
subjective phenomena of eating; (2) the types of exteroceptive
and interoceptive information that affect eating; (3) the periph-
eral neural and endocrine sensory mechanisms relaying this
information to the brain; and (4) the brain neural networks
that process and integrate this, and other, information to control
acts and subjective phenomena associated with it. An important
aspect of this processing is the plasticity of the brain's eating-
control systems - that is, experience shapes every aspect of the
physiology of eating.
Eating in humans and other mammals is organized into
discrete meals, which, in turn, are thought to result from
four functional processes: (1) hunger processes related to the
187
1 88 Appetite
Table 1 Peripheral within- and across-meal feedback signals in the control of eating
Control
Species
Within-meal
Across-meal
Physiological role
Flavor hedonics
R-M, H
y
♦♦
Gastric mechanoreception
R-M, H
y
♦
Cholecystokinin (CCK)
R-M, H
V
♦♦
Glucagon-like peptide-1 (GLP-1)
R-M
y
y
♦>
GLP-1
H
V
♦
Peptide YY (PYY)
R-M, H
y
Amylin
R-M
y
♦♦
Amylin
H
y
♦>
Amylin
R-M
y
Ghrelin
R-M, H
y
♦
Metabolic signals
R-M
y
y
♦
Insulin in overweight
R-M
y
-
Insulin in underweight
R-M
y
-
Leptin in overweight
R-M
y
-
Leptin in underweight
R-M, H
y
♦>
Summary of within- and across-meal controls discussed in the text. Several other candidate mechanisms are not included. Species refers to laboratory rats or mice (R— M; for
simplicity, the sometimes-important species differences between rats and mice are ignored) or to humans (H). Physiological role means the likelihood that the control produces
a graded, dose-related effect on normal eating, rated as unlikely (-), somewhat likely with very incomplete evidence (❖), probable due to promising but incomplete evidence (♦),
or well established (♦♦).
Table 2 Brain neurotransmitters and neuromodulators involved in
the control of eating
Neurotransmitter
Typical abbreviation
Agouti-related peptide
AgRP
Cholecystokinin
CCK
Cocaine- and amphetamine-related transcript
CART
Corticotropin-releasing hormone
CRH
Dopamine
DA
Dynorphin
Dyn
Enkephalin
Enk
Galanin
GAL
y-aminobutyric acid
GABA
Gastrin-releasing peptide
GRP
Glutamate
Glut
oc-Melanocyte-stimulating hormone
or-MSH
Neuropeptide Y
NPY
Norepinephrine
NE
Orexin-A and orexin-B
Or-A, Or-B
Oxytocin
OT
Melanin-concentrating hormone
MCH
Serotonin (5-hydroxytryptamine)
5-HT
Thyrotropin-releasing hormone
TRH
This list is not exhaustive. Many neurotransmitters are named for their initially
discovered endocrine roles. This is an example of how signaling molecules often have
different roles in different parts of the body. There are variants for several of the
abbreviations listed.
initiation of meals; (2) flavor hedonics processes related to the
evaluation of the food that stimulate or inhibit eating during
the meal; this is one aspect of food reward; (3) satiation pro-
cesses related to inhibitory feedbacks from postingestive food
stimuli that act to terminate eating at the end of the meal;
and (4) postprandial satiety processes inhibiting eating during
the intermeal interval. These four processes appear to have at
least partially independent underlying neural mechanisms.
Although each process includes both behavioral and subjective
phenomena, for simplicity, I use single names for both aspects.
We make the decision to select this food or that food, or
to have a little more of a particular food during a meal, many
times a day, usually with little conscious effort or thought.
Nevertheless, these are important decisions that can be
subject to many influences, some physiological, some not. Fur-
thermore, these various influences may push us in opposite
directions at the same time - for example, feedbacks from the
stomach and intestines may signal us to stop eating at the same
time that sensory feedbacks related to food flavor tell us to keep
on. For example, the influence of flavor hedonics is often in
opposition to the influence of another category of controls,
homeostatic or regulatory controls of eating. Both hedonic (section
'Flavor Hedonics') and homeostatic (section 'Eating and Homeo-
stasis') controls of eating are discussed in more detail below.
Finally, my review of the physiology of eating is organized
around within- and across-meal controls of eating. Within-meal
controls refer to mechanisms in which a signal (e.g., changed
peripheral sensory neural signals or changed plasma hormone
concentrations) and a change in eating occur during the same
meal; across-meal controls are those in which the signal and
the eating response are not linked to the same meal. The major
peripheral within- and between-meal signals are presented in
the sections 'Within-Meal Controls of Eating' and 'Across-Meal
Controls of Eating,' respectively. As it becomes clear, however,
within- and between-meal controls function along a contin-
uum, so that the distinction is mainly heuristic.
Eating and Homeostasis
Because eating is our only source of metabolic fuel and of
essential nutrients, it is an integral part of homeostatic regula-
tion, that is, the maintenance of crucial physiological variables
within relatively narrow ranges. The concentrations of glucose,
0 2 and C0 2 in the blood, blood pressure, body temperature,
and a number of other variables are maintained within ranges
that are much smaller than would occur if there were no
Appetite 1 89
Figure 1 Schematic diagram of the organization of a negative-feedback
control system, such as is thought to regulate micronutrient homeostasis
and energy homeostasis. An afferent feedback signal related to the
regulated variable is compared by the brain to the desired level of the signal
and to other inputs. This integrated signal then causes the brain to initiate
or adjust effector signals that affect control of the selection of food,
amount eaten, energy expenditure, etc., in ways that bring the regulated
variable back to the appropriate level, that is, reestablishes homeostasis.
regulation. Similarly, physiological regulations providing for
appropriate supplies of metabolic energy, of protein, and of
micronutrients can powerfully influence how much is eaten
and what is eaten. As a consequence, homeostasis is a major
conceptual scheme used to understand eating. Figure 1 shows
the basic organization of a negative-feedback regulatory sys-
tem. Interestingly, many regulatory controls of eating seem to
be organized asymmetrically, in that they protect from deficits
but not from surfeits.
Specific Appetites for Micronutrients
Sodium appetite is an excellent example of the contribution of
eating to micronutrient homeostasis. The concentration of
sodium ions, Na + , in the extracellular fluid and blood plasma,
is maintained near 150 mOsm. Loss of Na + , for example, when
little or none is ingested for several days, leads to a number of
homeostatic responses. These include secretion of the min-
eralocorticoid hormone aldosterone from the adrenal cortex,
activation of the peptide hormone angiotensin II from its
circulating precursor, and stimulation of neural baroreceptors
in the vasculature. The synergistic action of these three sensory
feedback signals in the brain produces a specific appetite for
sodium, that is, a marked increase in the palatability of any
sodium-containing food or fluid and an increase in its intake.
This appetite is innate: it occurs during the first occurrence of
sodium depletion, and the taste of salt is recognized without
prior experience. Only a few other similar molecules, such as
potassium, are partially confused with sodium. The occurrence
of sodium appetite sensitizes the system in a way that increases
'need-free' salt ingestion. Although sodium appetite is gener-
ally considered in the context of body fluid homeostasis, salt
preference increases food palatability, which may lead to
chronic increases in the intake of salty foods.
In contrast to the sodium appetite, most specific appetites
are learned. Vitamin B appetite is a good example. When rats
are fed a diet containing no vitamin B, deficiency signs begin to
appear in about a week. If the rats are then offered a choice
of a few new and differently flavored diets, they consume
significant amounts of all of them. If just one of the new
diets contains vitamin B, they learn in the course of a few
days to eat mostly that diet. They do not recognize the flavor
of vitamin B; rather, the enriched diet must have a salient
flavor. If the experimenter switches the vitamin B to a different
diet after the rats have learned to choose the first vitamin
B-containing diet, the rats continue to eat the originally
enriched diet until deficiency signs reappear. Learning capacity
is also limited. If deficient rats are offered about ten diets, with
only one enriched with vitamin B, they cannot learn to select it.
Most micronutrient appetites appear to function like vitamin B
appetite. Such specific appetites, however, do not exist for
all micronutrients. Finally, in the cases of sodium appetite,
vitamin B appetite, and most other micronutrient regulations,
only deficits and not surfeits lead to regulatory behavior.
Energy Homeostasis
Food is the only source of metabolizeable energy. Both the
supply of readily available energy metabolites and the amount
of stored energy are physiologically regulated. These regula-
tions together are referred to as 'energy homeostasis.'
Energy balance refers to the relation between the metabolize-
able energy content of the food eaten and the energy expended
(as physical work, heat, or in excretions) over a specific time.
During positive energy balance, excess energy is stored as chem-
ical energy in the body, and during negative energy balance,
stored energy is utilized. Body weight is a relatively accurate
indicator of an individual's history of energy balance. Young,
growing individuals are normally chronically in positive energy
balance, as energy is transformed into increased lean body mass.
In adulthood, however, increases in body weight usually reflect
increases in adiposity, that is, energy stored as fat, that is,
Uiacylglycerols. These relationships are described by the energy
balance equation:
Energy stored = Energy ingested - Energy expended.
Fat is stored in the adipose tissue. There is also ectopic
storage of triacylglycerol in liver, muscle, and other tissues in
overweight people. Although ectopic fat has important health
consequences, the amount is much less than the amount of fat
in the adipose tissue. Similarly, the few thousand calories of
energy stored in the form of carbohydrate, that is, glycogen, is
quantitatively negligible in comparison to fat stores.
The gold standard for measuring adipose tissue mass is
computed tomography. In humans, much handier methods
are waist circumference and body mass index (BMI):
BMI = weight (kg)/height 2 (m 2 ).
BMI correlates well with body adiposity for most people,
although the relation breaks down in more muscular and taller
people, especially women, and in children. BMI scores predict
different percent body adiposity in males and females, in
accordance with the higher normal fat content of females
after puberty. The standard threshold BMI for overweight and
obesity are 25 and 30. These days, about two-thirds of adult
Regulated (constant)
variable
Food selection,
amount eaten
1 90 Appetite
US American adults are overweight or obese. Waist circumfer-
ence is increasingly used because it correlates better than BMI
with abdominal adiposity (i.e., the sum of abdominal subcu-
taneous and intra-abdominal adipose tissues) and has a better
predictive value for risk factors of metabolic disease. The
threshold waist circumferences for increased risk are 102 cm
in men and 88 cm in women.
Despite the high prevalence of overweight and obesity,
three lines of evidence indicate that the brain senses energy
stored and, depending on its level, appropriately adjusts eating
and energy expended. (1) The relative stability of body weight
over longer periods. Although people gain several kilograms
from young adulthood to middle age, the cumulative error
in energy balance is small - if there were only a 1% constant
error in energy balance over 25 years, weight gain would be over
1000 kg. (2) Dynamic compensatory responses that experimen-
tally induced overweight (or underweight) elicit. For example,
after experimentally induced obesity, animals usually decrease
eating and increase energy expenditure and return to near the
initial levels of body weight and adiposity. Why the compensa-
tion is often incomplete is not understood. (3) Our increasing
understanding of the extensive neuronal networks in the brain
that contribute to the regulation of adiposity, is introduced
below. Why does this system not function better? It seems that
the increased availability of highly palatable foods and the
decreased amount of physical activity in our society overwhelm
the regulatory system and cause many people to take in more
energy than they need and become overweight or obese.
Finally, it is generally agreed that starvation, rather than
obesity, was the more important challenge during the mamma-
lian evolution. Accordingly, it should not come as a surprise that
the physiological adaptations and compensatory responses eli-
cited by starvation, including the drive to eat more, are more
powerful than the responses elicited by overweight.
Within-Meal Controls of Eating
Flavor Hedonics
The effects of food hedonics on eating are often considered the
antithesis of homeostatic influences. In fact, this is not entirely
true, as shown by the example of increased sodium palatability
in sodium-deficient animals or people. Indeed, in many ways
we seem to be physiologically programmed to like what we
need. Unfortunately, in the environments in which most of us
live, energy-dense foods are available in abundance, so that
flavor hedonics opposes the regulation of adiposity.
The flavor of preferred foods is a biological pleasure, as
basic as the pleasure of sexual behavior. In fact, eating and
sex activate extensively overlapping neural networks in the
brain, as, unfortunately, do drugs of abuse. The neural evalua-
tion of the pleasantness of flavor stimuli can either stimulate
(for preferred flavors) or inhibit (for nonpreferred flavors)
eating during a meal. Thus, flavor hedonics is an important
within-meal control of eating.
The perception of flavor arises from olfactory, gustatory,
tactile, and thermal food stimuli affecting receptors in the
nasal and oropharyngeal cavities. Flavor has two roles in
eating. (1) Flavor stimuli provide information guiding food
selection, that is, identification of the type (e.g., it's sweet ) and
intensity (it's as sweet as candy) of food stimuli, independent of
the stimulus' hedonic qualities. (2) Flavor stimuli produce
hedonic (used especially when conscious experience is
measured) or palatability (used for behavioral as well as sub-
jective measures) judgments, that is, the pleasant (or unpleas-
ant) subjective experiences of food stimuli (e.g., I like sweet)
that stimulate or inhibit eating. Although some preferences
(e.g., sweet in any intensity, salt in moderate intensities) and
aversions (e.g., bitter, sour) seem to be innate, the vast majority
of preferences and aversions are learned. Some such learning
is linked to the physiological consequences of eating - for
example, hungry rats learn to prefer food with higher energy
contents. But much learning occurs independent of nutritional
or physiological consequences. Emotional, cognitive, and cul-
tural associations, and even just repeated exposure, produce
learned flavor preferences. Repeated exposure is a recom-
mended method to induce children to eat usually nonpreferred
vegetables or other foods the parents may be interested in, such
as hot chili (capsaicin). Most people learn to like certain spe-
cific foods, for example, chocolate or pizza, very much, and
their eating is markedly influenced by stable, strong 'cravings'
for such foods. Because flavor preferences dramatically affect
patients' success in adhering to therapeutic dietary regimens,
the origins and plasticity of flavor preferences are important
areas for behavioral and physiological research.
One mechanism for learned preferences seems to be asso-
ciations between flavors and the postingestive consequences
of the food. A dramatic example of the ability of such learning
to overshadow the strong innate hedonic response to sweet
was described by Anthony Sclafani of the City University of
New York, Brooklyn, NY, USA. Hungry rats were offered either
a noncaloric bitter solution or a noncaloric sweet solution
(0.2% saccharin) to drink on alternate training days. Food,
but not water, was available. On training days with the bitter
solution, glucose was intragastrically infused in proportion to
the amount drunk, and on training days with the sweet solu-
tion, water was intragastrically infused. After 4 days training
with each solution, the rats were offered the bitter and the
sweet fluid simultaneously, and no intragastric infusions were
done. Rats selected 80% bitter solution and only 20% sweet
solution, the opposite of the normal preferences.
It is important to note that, as the data just described
indicate, hedonic judgments are responses of the brain, not
properties of food. A flavor that is judged pleasant in one
situation (e.g., garlic added to salad dressing) might be judged
unpleasant in another (e.g., garlic added to chocolate cake).
The neural processes producing flavor hedonics are thought to
be mainly in the telencephalon (cerebral hemispheres). They
include both subcortical areas (e.g., parts of the striatum,
including the anterior caudate, ventral pallidum, and nucleus
accumbens (NAc), and parts of the amygdala (central nucleus
and basolateral)) and, at least in primates, several cortical areas
(e.g., cingulate cortex, insular cortex, and orbitofrontal cortex
(OFC)). The OFC is especially interesting. It receives relatively
direct inputs from both olfactory and gustatory receptors, and,
as increasing numbers of functional imaging studies indicate, it
is involved in the evaluation of the pleasantness of flavor and
other stimuli. For example, as shown in Figure 2, it has been
reported that the sight or flavor of chocolate produce greater
activation of the OFC in people who describe themselves as
Appetite 191
Figure 2 Demonstration that persons who are chocolate cravers (crav)
increase activation of the OFC to the sight (chocpic) or flavor of chocolate
(choc) more than persons who are not chocolate cravers (noncrav). Data
are percent changes in blood oxygenation level (BOLD) measured for
1 0 s after presentation of the chocolate or picture of chocolate. Reprinted
with permission from Rolls ET and McCabe C (2007) Enhanced
affective brain representations of chocolate in cravers vs. non-cravers.
European Journal of Neuroscience 26: 1067-1076.
chocolate cravers than in people who do not. Finally, the
neurochemicals most strongly implicated in hedonics include
dopamine (especially dopamine acting on D2 dopamine
receptors in the NAc), endogenous opiates, and serotonin.
Finally, hedonic judgments are conscious perceptual states,
and eating is a behavior. What, then, is the relationship
between subjective states and behavior? Since acceptance of
the principle of evolution by natural selection, the most parsi-
monious view is that subjective states evolved as causal agents,
that is, conscious experience can cause behavior. This has been
especially forcefully argued by the philosopher John Searle.
Part of his view is that the belief that the (subjective) mind
and the (physical) brain are fundamentally different kinds of
things is illusory. How neural activity produces consciousness,
and how consciousness affects neural activity, however, remain
beyond the scope of today's technologies. Nevertheless, func-
tional brain imaging methods are yielding more and more
information about the relationships among neural processes
in the brain, conscious states, and eating.
Gastric Mechanoreceptors
The stomach is richly innervated with mechanoreceptors, that
is, sensory nerve endings that respond to stretching the smooth
muscle of the stomach wall. These sensory nerves project to the
brain via both vagal and spinal visceral (splanchnic) afferents.
Studies in animals and humans indicate that the increase in
gastric volume that occurs during large meals is an adequate
stimulus for these mechanoreceptors and that the resulting
neural signals contribute to satiation. The chemical nature of
the gastric contents, in contrast, is not sensed. (More nutrient-
dense gastric loads empty into the intestines more slowly than
less nutrient-dense loads, however, to produce longer-lasting
volume signals.)
Figure 3 Demonstration that learned associations between flavor
stimuli and gastric and postgastric signals control meal size. Rats
consumed daily test meals of a sucrose solution during which pyloric
cuffs were closed, so that ingested food could not pass from the stomach
to the duodenum, in two conditions: first, with the gastric drainage
cannulas closed (closed symbols), and second, with the gastric drainage
cannulas open, so that the stomach remained empty (open symbols).
Three phenomena are visible. First, meal size increased about twofold on
the first test with the gastric cannulas open (day 3). This reflects the
absence of normally operating unlearned controls of eating. Second,
meal sizes in both conditions increased progressively during days 4-20,
when two tests with the gastric cannulas closed were done between each
test with the cannulas open. These progressive increases reflect the
gradual weakening of learned associations between the flavor of sucrose
and the absent gastric or postgastric food stimuli. Third, meal size
increased even more rapidly in tests after day 20, when the gastric
cannulas were open in every test. Because the only difference between
the two conditions was that gastric filling occurred when the gastric
cannulas were closed (recall that the pyloric cuffs were closed in all
tests), this indicates that learned flavor-gastric volume signals are potent
controls of meal size. Reprinted with permission from Davis JC,
Smith GP, and Meissner J (1993) Postpyloric stimuli are necessary
for the normal control of meal size in real feeding and sham
feeding rats. American Journal of Physiology 265: R888-R895.
Neural recording studies in animals indicate that some of the
vagal afferents respond synergistically to gastric volume and
other meal-control signals, such as cholecystokinin (CCK). In
humans, the volume of the antrum, that part of the stomach
nearest the pyloric sphincter, which separates the stomach from
the small intestine, seems most important for the feeling of
fullness. Functional brain imaging studies suggest that fullness
perceptions arising from increased gastric volume involve neural
networks in the amygdala and the insular cortex, two areas also
implicated in flavor hedonics.
The normal influence of gastric volume on eating is in large
part learned. This has been elegantly demonstrated in rats that
were surgically prepared so as to break the normal association
between the flavor of the food eaten and gastric volume during
test meals. As shown in Figure 3, this disconnection caused a
progressive increase in meal size over test days, indicating the
gradual deterioration of learned satiation signals.
Cholecystokinin
The gut peptide hormone CCK is an important mediator of the
satiating effects of nutrients in the small intestine. CCK is
1 92 Appetite
secreted from glandular cells dispersed in the epithelial cell
layer of the small intestine, especially its initial part, the duo-
denum. As is the case for many other endocrine cells of the
gastrointestinal tract, one side of the cell is exposed to the
gastrointestinal tract lumen and expresses receptors that stim-
ulate secretion from the opposite side of cells. CCK is secreted
from the other side of the cell into the extracellular space of the
lamina propria, from which it diffuses into the blood. The
digestion of fats and proteins stimulate CCK secretion within
minutes of meal onset. CCK satiating was first described in a
series of classic studies starting in the 1970s by Gerard
P. Smith, James Gibbs, and their colleagues at Cornell Univer-
sity Medical College. By now, tests of a number of stringent
endocrine criteria indicate that one of the normal effects of
CCK is to elicit satiation both in rats and humans. Some of the
most important results are (1) infusion of CCK in amounts
that reproduce levels reached during meals are sufficient to
inhibit eating in humans; (2) infusion of selective CCK-1
receptor antagonists just before meals blocks the satiating
effect of intraduodenal infusions offat; and (3) CCK-1 receptor
antagonist infusion during meals increases meal size (and the
perception of hunger in humans). Figure 4 shows some of
these data. The amount of evidence implicating CCK in normal
satiation is greater than for any other meal-control signal.
Whether long-term treatment with CCK or CCK agonists can
be used effectively to control eating therapeutically, however,
remains unclear.
Figure 4 Demonstration of the satiating action of CCK in human
subjects. Left: Intravenous (IV) infusions of the CCK-1 receptor
antagonist loxiglumide (LOX) or a control saline solution (SAL) and
intraduodenal (ID) infusions of a fat emulsion (FAT) or SAL were done
while normal-weight male volunteers ate a lunch of orange juice, ham
sandwiches, chocolate mousse, and coffee. ID fat infusions reduced meal
size, and this satiating effect of fat was blocked by CCK-1 receptor
antagonism, indicating that endogenous CCK was necessary of it. Right:
Linder similar conditions, except without ID infusions and a banana shake
preload before the IV infusion, IV infusion of a larger loxiglumide dose
increased meal size, indicating that endogenous CCK was involved in
normal satiation under these conditions. Meal sizes are expressed in
calorie equivalents because of the different foods eaten. ‘Statistically
significant effect. Reprinted with permission from (left) Matzinger D,
Gutzwiler J-P, Drewe J, etal. (1999) Inhibition of food intake in response
to intestinal lipid is mediated by cholecystokinin in humans. American
Journal of Physiology 277 : R1718-R1724; (right) BeglingerC, Degen L,
Matzinger D, D’Amato M, and Drewe J (2001) Loxiglumide, a CCK-A
receptor antagonist, stimulates calorie intake and hunger feelings in
humans. American Journal of Physiology 280: R1 1 49-R1 1 54.
CCK's satiating action may be in part endocrine (i.e., reach
receptors via the circulation) and, at least in rats, in part para-
crine mode (i.e., locally, in the laminal propria of the small
intestine, before entering the blood). In rats, the CCK-1 recep-
tors that mediate the paracrine action appear to be located on
gastrointestinal smooth muscle in the pyloric area and on
sensory fibers of the vagus nerve that innervate the same
region, and those that mediate the endocrine action appear
to be in the hepatic-portal vein or liver. The signals generated
by all these CCK- 1 receptors elicit afferent signals in the vagus
nerve, which projects from the gut directly to the brain, to a
region known as the nucleus of the solitary tract (NTS) in the
hindbrain. A wealth of data, some reviewed below, indicate
that the NTS is an important processing station in the control
of eating.
Glucagon-Like Peptide 1 and Peptide YY
Glucagon-like peptide 1 (GLP-1) and peptide YY (PYY) are also
gut peptide hormones. They are cosecreted by a single endo-
crine cell type, located mainly in the ileum, during and after
meals. Glucose produced during the digestion of carbohy-
drates is the most potent stimulus for GLP- 1 and PYY secretion.
These peptides may contribute to both satiation and postpran-
dial satiety. In rats, GLP-1 may act locally, on receptors on vagal
neurons in the lamina propria of the small intestine, to produce
a vagal signal to the brain. Whether its across-meal action is also
vagally mediated or is the result of stimulation of GLP-1 recep-
tors in the brain is unknown. The brain also contains neurons
that use GLP-1 as a neurotransmitter. These neurons have been
implicated in inhibition of appetite during visceral illness. The
site of action of PYY is not yet clear; the NPY-Y2 receptors that
mediate its actions of are expressed in the hypothalamus, the
hindbrain, and abdominal vagal afferents.
As for CCK, infusion of GLP-1 in amounts that produce
levels reached during meals are sufficient to inhibit eating in
humans. In contrast, pharmacological amounts of PYY, which
often elicit illness, seem to be required to inhibit eating. Note,
however, that intravenous administration may not mimic the
potentially paracrine effect of PYY on vagal afferents. The effects
of GLP- 1 or PYY antagonism have not been tested in humans as
yet. GLP-1 agonists are in use to facilitate insulin secretion in
patients with type-2 diabetes mellitus (T2DM), who are gener-
ally obese, and may also be useful for appetite control.
Amylin
The peptide hormone amylin (or islet amyloid polypeptide,
IAPP) is synthesized mainly by pancreatic P-cells and cose-
creted with insulin beginning in the first minutes of meals.
Meal-related amylin secretion may function as a within-meal
satiation signal, and basal amylin levels may function as an
across-meal control of meal size (basal levels). Thus, in rats,
acute injection of an amylin receptor antagonist, just prior to a
meal, increased the size of that meal, and continuous infusion
of the antagonist over 2 weeks chronically increased meal size
and body weight. Amylin acts on receptors in the area post-
rema, a part of the brain with little or no blood-brain barrier
that is adjacent to and heavily interconnected with the NTS.
Amylin's satiating effect has not yet been investigated in detail
Appetite 1 93
in humans, although pharmacological use of amylin agonists
is attracting interest both as an aid to weight loss and for the
control of diabetes.
Ghrelin
Ghrelin is a peptide hormone synthesized and secreted mainly
by endocrine cells in the gastric wall. Ghrelin has attracted
great interest as a potential hunger signal because it is the
only gut peptide whose secretion is stimulated during fasting
and inhibited by eating and because it is the only gut peptide
whose administration stimulates eating, which has been
shown in animals and humans. This appears to be an endo-
crine effect of ghrelin acting on receptors in the arcuate nucleus
(Arc) of the hypothalamus. The physiological status of ghrelin
is not yet well established.
Across-Meal Controls of Eating
Metabolic Signals
Neural sensitivity to metabolic processes in three organs, the
intestines, the liver, and the brain, appears to contribute to the
control of eating. Systemic administration of metabolic fuels
often reduces eating, and pharmacological inhibition of glu-
cose or fatty-acid utilization often increases it. These appear to
be across-meal signals, that is, the changes seem to develop
over durations that are longer than individual meals. Changes
in glucose, however, may lead to within-meal as well as across-
meal effects. Finally, whether metabolic processes are normally
involved in the control of eating or contribute only in meta-
bolic emergencies, such as severe hypoglycemia or starvation,
remains unknown.
Glucose
Glucose is the most important metabolic fuel and has been
much studied. Several demonstrations that intravenous glucose
infusions inhibit eating and that coinfusion of insulin increases
the effect suggest that the metabolic utilization of glucose
contributes to the conttol of eating. Mice with a transgenic,
relatively brain-specific deficiency in the glucose transporter-2
(GLUT-2) eat more than wild-type mice and show abnormal
patterns of hypothalamic neuropeptide expression during
the fasted-to-fed transition, suggesting the absence of GLUT-2
compromised the function of glucose sensors that contribute
to the hypothalamic control of eating (see next section).
The liver also expresses GLUT-2, and meal-related increases
in hepatic-portal vein glucose concentration may contribute to
satiation. For example, intrameal hepatic-portal vein infusions
of small amounts of glucose or of glucose plus insulin selec-
tively reduced the sizes of spontaneous meals in rats. Electro-
physiological and anatomical data indicate that the glucose
receptors stimulated are likely to be on vagal afferents termi-
nating in the wall of the hepatic-portal vein.
Glucose-sensing neurons, that is, neurons that change
electrophysiological activity depending on changes in glucose
level or utilization within the normal range, are present in
several brain areas, from the hindbrain to the hypothalamus.
Many of these neurons express GLLIT-2, sulfonylurea receptor- 1
(whose activation stimulates insulin secretion), and glucokinase
(the rate limiting enzyme in glycolysis), thus making them
excellent candidates as glucose-utilization sensors. There is evi-
dence that brain glucose-sensing neurons conttibute both to the
homeostatic regulation of blood glucose level and to the conttol
of eating.
Finally, a small, transient decrease in blood glucose level
occurs prior to spontaneous meals in rats and humans under
some conditions. This decrease may be a pattern whose recog-
nition conuibutes to meal initiation. The decrease seems not to
reflect a decrease in glucose utilization because it is very small
and because glucose level usually returns to normal before the
meal is initiated. One possibility is that the brain periodically
initiates a pulse of insulin secretion via vagal afferents in order
to test the body's ability to compensate and initiates eating if
the compensation is not rapid enough.
Fatty acids
Fatty-acid utilization has also been implicated in eating. Acute
pharmacologic inhibition of fatty-acid oxidation (FAO) has
been shown to stimulate eating in animals and humans.
Although the liver is a primary site of FAO, several results
suggest that it is not the site where FAO sensing contributes
to the control of eating. Rather, recent evidence suggests that
the intestinal wall may be the site. That is, depending on
nutritional status, intestinal epithelial cells may absorb fatty
acids either from the blood or (when medium or short chain
fatty acids are ingested) from the intestinal lumen and oxidize
them to cover their own energy needs, and the level of oxida-
tion may be sensed by vagal afferents. Fatty-acid levels or
metabolism can also be sensed centrally, in the hypothalamus,
and this also appears to affect eating, at least during pharma-
cological manipulations.
Intracellular energy How sensors
The intracellular signaling mechanisms linking metabolism
and eating have also been investigated. Reduction of cellular-
energy availability, whether due to decreases in fatty-acid or
glucose utilization, increases the AMP/ATP ratio and activates
(via phosphorylation) the intracellular-energy sensor AMP
kinase (AMPK), which exists in the periphery and the brain.
Local pharmacological activation of AMPK in the hypothala-
mus increases eating, suggesting that changes in cellular-energy
status are sensed. Similarly, pharmacological manipulation
of another cellular-energy status sensor, the molecule mam-
malian target of rapamycin (mTOR), also affects eating. mTOR
appears to be expressed by neuropeptide Y (NPY)/agouti-
related peptide (AgRP) neurons in the Arc of the hypothala-
mus, which, as explained in the next section, are supposed to
play a crucial role in homeostatic eating. mTOR activity is
decreased by metabolic-fuel deficiency and is increased by
fuel repletion, that is, the opposite pattern as AMPK activity.
In addition, activation of AMPK inhibits mTOR activity in the
some brain areas. These data suggest that activation of the two
fuel-sensing molecules should have opposite effects on neural
activity, but this has not yet been shown.
Insulin
Insulin, besides its many metabolic actions in the periphery,
seems to have a direct inhibitory action on eating. This action is
1 94 Appetite
the result of active transport of insulin from the blood into the
brain, where it acts on receptors in the Arc of the hypothala-
mus. It has been suggested that insulin may function as an
'adiposity signal,' that is, a signal providing the brain with
information about body fat that enables it to mount compen-
satory responses when body fat is increased or decreased.
A variety of indirect evidence supports this view. Insulin
administration into the brain reduces eating and increases
energy expenditure, and insulin antagonist infusion has oppo-
site effects. In addition, cross-sectional studies indicate that in
humans or animals who are maintaining a steady weight,
whether normal or elevated, basal (i.e., fasting) plasma insulin
levels are correlated with weight or adiposity. Finally, specific
transgenic knockout of brain insulin receptors rendered mice
obese. Although these data indicate that insulin is necessary
for normal energy homeostasis, they do not provide direct
evidence that insulin signals change in adiposity to the brain.
In fact, basal plasma insulin is not correlated with either body
fat or the magnitude of compensatory eating or energy expen-
diture responses in rats that were dynamically regulating their
weight following experimentally induced obesity or under-
weight. Thus, insulin's role in the control of eating remains
in doubt.
Leptin
Leptin is a peptide hormone secreted by the adipocytes them-
selves. Its position in the physiology of eating is similar to
insulin's. Cross-sectional studies indicate that basal plasma
leptin levels are correlated with weight or adiposity in humans
or animals who are maintaining a steady weight. Administra-
tion of leptin or leptin antagonism decrease and increase
eating, respectively, and genetic defects in the leptin signaling
pathway produce dramatic obesity phenotypes. But direct sup-
port for a role as an adiposity signal is lacking. For example,
basal plasma leptin was not correlated with either body fat of
the magnitude of compensatory eating and energy expenditure
responses in rats that were dynamically regulating their weight
following experimentally induced obesity.
Several data suggest that leptin's physiological role may be
in starvation or weight loss rather than in obesity. For example,
underweight animals become progressively more sensitive to
leptin administration, whereas overweight animals become
less sensitive. Rudolph Leibel, Michael Rosenbaum, and their
colleagues at Columbia University, New York, NY, USA,
have performed several fascinating studies of human subjects
recovering from a ~10% loss in body weight, which leads to
decreased plasma leptin levels, decreased energy expenditure,
and, presumably, increased hunger. In one study, these inves-
tigators showed that leptin infusions that reinstated normal-
weight leptin levels normalized energy expenditure. In a
follow-up study, brain responses to the sight of food items or
nonfood items were measured with functional magnetic reso-
nance imaging (fMRI) at normal weight and after weight loss.
Weight loss caused selective changes in the neural responses to
the sight of food items in several brain areas thought to be
involved in the control of eating, and normalization of leptin
levels reversed many of these. Such data suggest leptin is a
necessary part of the brain's complex response to the homeo-
static challenge of weight loss.
Brain Mechanisms
Eating is mediated by a complex, anatomically diffuse neural
network that is organized hierarchically, redundantly, and
recurrently. Three somewhat autonomous subnetworks have
been identified: a telencephalic network for hedonic proces-
sing, a hypothalamic network especially involved in homeo-
static mechanisms, and a hindbrain network. Some of the
principal structures in each of these networks are depicted in
Figure 5. The telencephalic network was introduced in the
section 'Flavor Hedonics.' The others are briefly sketched here.
Hypothalamus
Several subareas, or nuclei, of the hypothalamus have been
implicated in the control of eating, and an intrahypothalamic
neural network is thought to orchestrate the eating, neuro-
endocrine and autonomic responses contributing to energy
homeostasis. The Arc appears to be an important sensory
node in this network (Figure 6). As mentioned already, Arc
neurons are sensitive to hormones including leptin, insulin,
and to metabolites. One population of these neurons syn-
thesizes and uses the neurotransmitter NPY and the neuro-
modulator AgRP, and another population synthesizes and
uses the neurotransmitters a-melanocyte-stimulating hormone
(a-MSH; these neurons are often called POMC neurons after
the precursor of a-MSH, proopiomelanocortin) and cocaine-
and amphetamine-related transcript (CART). NPY and AgRP
stimulate eating when injected into the hypothalami of ani-
mals and are therefore called orexigenic peptides, whereas
a-MSH and CART injections inhibit eating, so they are called
anorexigenic peptides. Three main targets of projections of
these two classes of Arc neurons are the nearby paraventricular
nucleus (PVN), lateral hypothalamic area (LHA), and perifor-
nical area (PFA). Arc NPY/AgRP neurons inhibit PVN neurons
that express neurotransmitters including oxytocin (OT) and
corticotropin-releasing hormone (CRH), and other anorexigenic
peptides, and stimulate LHA neurons that express melanin-
concentrating hormone (MCH) and PFA neurons that express
orexin-A and -B (Or-A, Or-B), which are orexigenic. In contrast,
Arc a-MSH neurons inhibit PVN OT and CRH neurons and
stimulate LFLA/PFA MCH, Or-A, and Or-B neurons. Additional
neural elements of this network include inputs from serotonin
(or 5-hydroxytrypamine, 5-HT) neurons projecting from the
raphe nuclei in the midbrain and electrophysiologically inhibi-
tory interconnections mediated by y-aminobutyric acid (GABA)
neurons, both local interneurons and projection neurons from
the NAc. The latter represents one likely connection between the
hypothalamus and telencephalic networks processing flavor
hedonics.
This summary is based on a huge number of experiments.
Nevertheless, because available methodologies do not permit
truly crucial physiological experiments, the extent to which this
neural network, or parts of it, controls normal eating is uncer-
tain. Whether or not particular neurochemical signaling systems
operate in most conditions, however, does not preclude the
development of pharmacotherapies based on them (or on the
gut signaling molecules discussed above). Because of the many
health problems related to the obesity epidemic, pharmaceutical
companies have mounted huge efforts in this direction.
Appetite 1 95
Raphe N
Figure 5 Schematic diagram of the roughly localizing three partially independent subnetworks controlling eating in the telencephalon (cerebral
hemispheres), hypothalamus, and hindbrain. Several important brain areas in each region are listed. The view of the brain is the medial aspect of the
right side of the brain after a midsagittal section, as can be done by cutting tissue with a knife or by computed tomographic imaging. The front is to
the left. AP, area postrema; Arc, arcuate nucleus; BLA, basolateral amygdala; CeA, central nucleus of the amygdala; Cing, cingulate cortex; DMH,
dorsomedial hypothalamus; DMV, dorsal motor nucleus of the vagus; HN, hypoglossal nucleus; LHA, lateral hypothalamic area; MRT, medullary
reticular formation; MV, motor nucleus of the trigeminal nerve; NAc, nucleus accumbens; NTS, nucleus of the solitary tract; OFC, orbitofrontal cortex;
PBN, parabrachial nucleus; PFA, perifornical area; PVN, paraventricular nucleus; Raphe N, midbrain raphe nuclei (dorsal and magnus); VLM,
ventromedial medulla; VMFH, ventromedial hypothalamus; VP, ventral pallidum.
Hindbrain
The hindbrain (midbrain, pons, and medulla) is the target of
both neural and endocrine signals involved in the control
of eating. In addition, it contains the interneuronal networks
that function as 'central-pattern generators' coordinating the
acts of eating and the cell bodies of the motor neurons that
project to the muscles involved. The results of neural proces-
sing in the telencephalon and hypothalamus must be conveyed
to these hindbrain motor networks to have any influence on
eating. These descending projections do not appear to target
the motor networks directly, but to project to interneuronal
networks that process peripheral inputs to the hindbrain, such
as vagally mediated meal-control signals. The final integrated
output is then directed to the motor networks. The PVN is the
source of a number of projections from the hypothalamus to
the hindbrain. For example, PVN OT and CRH neurons project
to the NTS and nearby sensory and motor areas.
Hindbrain neurons that express the amine neurotransmit-
ters serotonin, dopamine, and the catecholamines norepi-
nephrine and epinephrine are another source of projections
to the forebrain involved in the control of eating. Serotonin
and dopamine have been implicated in both flavor hedonics
and satiation. The catecholamines, which are also expressed by
neurons in the forebrain, have been implicated in normal
satiation and in anorexia.
A series of elegant studies by Harvey Grill and his colleagues
at the University of Pennsylvania, Philadelphia, PA, USA,
revealed several unsuspected aspects of the hindbrain's contri-
bution to eating. Much of this work used chronic decerebrate
rats, which are rats with transections of the neuroaxis that
separate the hindbrain and forebrain (i.e., telencephalon, thal-
amus, and hypothalamus). This work indicates that the hind-
brain is sufficient for nearly normal effects of gustatory and
gastrointestinal feedback signals on meal size. Although these
rats do not initiate meals unless food is placed into the mouth,
when this is done by intraoral infusion, they take well-defined
meals terminated by passive refusal of more food, ingest more
of sweeter solutions, ingest less after CCK injections, etc. These
investigators have also shown that leptin and a number of
other endocrine signals can act directly in the hindbrain to
control eating. How these capacities of the hindbrain are
integrated with forebrain neural processing is a focus of current
research.
Intercellular Signaling
Multiple neurotransmitters and neurotransmitter receptor sub-
types in addition to those already mentioned have been impli-
cated in the control of eating. These are listed in Table 2. The
assessment of their physiological roles in eating is much more
difficult than the assessment of peripheral endocrine signals
discussed previously, so that progress in this area is slow. Again,
however, it is important to realize that pharmacological treat-
ments for eating can be based on neurotransmitter systems even
if they are normally only rarely involved in the control of eating.
Intracellular Signaling
Some intracellular signaling molecules involved in the brain
neural networks controlling eating have been identified. These
1 96 Appetite
Figure 6 Schematic diagram of the core ‘anorexigenic’ and ‘orexigenic’
neural networks in the hypothalamus, as described in the text. Dashed
lines indicate the Arc, PVN, and LHA/PFA. Within these areas, neural cell
bodies are represented by circles, neurotransmitter or endocrine
receptors by squiggles, axons by straight solid lines, and axon terminals
by ovals. Excitatory synapses are denoted by a plus sign in the axon
terminal, and inhibitory synapses by a negative sign. Arc neurons are
receptive to a variety of metabolites, hormones, and neurotransmitters.
A medially located population of neurons expresses the signaling
molecules NPY and AgRP and has excitatory projections to neurons
in the LHA/PFA that express Or-A and other orexigenic molecules; the
NPY/AgRP neurons also have inhibitory connections with neurons in
the PVN that express OT and other anorexigenic molecules. A more
laterally situated population of Arc neurons that express a-MSH has a
reciprocal connectivity: it excites PVN neurons and inhibits LHA/PFA
neurons. Finally, the PVN and LHA/PFA send and receive projections from
a variety of hindbrain and telencephalic areas. The diagram shows only
one side of the brain; the dotted line on the left represents the midline.
include adenosine 44/42 mitogen-activated protein kinase
(MAPK or ERK1/2), cAMP response element-binding protein
(CREB), signal transducer and activator of transcription 3
(STAT3), and suppressor of cytokine signaling 3 (SOCS3). In
addition, the immediate early gene products c-Fos and c-Jun,
which are activated by neural activity, are useful markers of
brain areas activated in particular functional states. The roles of
these various molecules are not yet well established. Activation
(i.e., phosphorylation) of MAPK is especially interesting
because it has been reported that pharmacological inhibition
of MAPK affects eating, indicating that it has a necessary role
under some circumstances.
Sex Differences
In addition to their functions in reproductive physiology,
gonadal steroid hormones, especially the estrogens and andro-
gens, act in the brain to affect eating. The clearest phenome-
non, which occurs in women as well as in many animal
species, is that eating decreases during the periovulatory
phase of the ovarian cycle. In rats and mice, this has been
shown to be due to an inhibitory effect of endogenous estro-
gens on eating leading to reduced meal size. An important
mechanism underlying this is that estrogens increase the satiat-
ing potency of CCK. This increase is due to the activation of
estrogen receptors on neurons in the NTS about a day before
ovulation, as has been shown with local administration of
estradiol (the major estrogen) or estrogen antagonists. Three
other effects of estrogens have been identified in rats and mice
that may also contribute: first, estrogens decrease the eating-
stimulatory effect of ghrelin; second, estrogens increase the
satiating potency of pancreatic glucagon; and, third, at least
under some test conditions, estrogens decrease the eating-
stimulatory effect of sweet taste.
In rats and mice, interruption of ovarian cycling increases
meal size and leads to increased food intake and adiposity.
This phenomenon has also been linked to the effect of estro-
gens to increase CCK satiation. On average, women increase
adiposity during the menopausal transition, but whether this is
due to increased eating is unknown (there is also a shift in
regional adipose tissue deposition, which may be due to the
decrease in circulating leptin levels after menopause).
The estrogenic inhibition of eating is mediated by a partic-
ular estrogen receptor, estrogen receptor- alpha (ER-a). Girls
with variants of the ER-a gene that lead to partial loss of func-
tion have been found to develop an increased risk to become
overweight or obese beginning at puberty. This strongly sug-
gests that estrogens control eating similarly in women as in
rats and mice. It has been reported that estrogens increase the
eating-inhibitory effect of exogenous leptin and decrease
the eating-inhibitory effect of exogenous insulin, but whether
these are endogenous physiological mechanisms is not yet clear.
There are also sex differences in flavor hedonics. In rats,
females prefer sweet solutions less than males, and in humans,
women designate less sweet foods maximally pleasant than
men. Estrogens appear to reduce the hedonic evaluation of
sweets both in female rats and women. Estrogen-independent
mechanisms may also contribute. In humans, there is a sex
difference in the numbers of sweet- and bitter-sensitive taste
receptors which is associated with differences in flavor hedonics.
Men and women also tend to have very different sorts of food
cravings, although the extent to which this reflects biological sex
differences rather than cultural influences is unknown.
Finally, how sex differences in the physiology of eating are
related to women's increased vulnerability to psychiatric eating
disorders are important, but little understood, issues.
Anorexia
Anorexia is decreased appetite or decreased eating caused by
illness, aversion, or abnormal interference with the normal
physiological controls of eating.
Illness Anorexia
Anorexia is a common element of systemic infections and
other immune challenges. Anorexia during the 'acute-phase
response' of the innate immune system, like fever, is unpleas-
ant, but can facilitate recovery. More chronic illness anorexia,
such as that which accompanies cancer, however, is a maladap-
tive response that can increase illness severity.
Appetite 1 97
A number of proinflammatory cytokines (interleukin- 1,
interleukin-6, tumor necrosis factor-a, etc.), prostaglandin-E 2 ,
and other immune signaling molecules are involved in acute
illness anorexia. These appear to converge on the same neural
networks that mediate normal eating. For example, serotoner-
gic neurons in the dorsal raphe nuclei, which project into the
hypothalamus and telencephalon, have been implicated in
both the normal control of eating and in illness anorexia.
Acute illness leads to a unique form of learning, the
conditioned taste aversion. If an unfamiliar food item is eaten
in the hours before an acute illness occurs, especially upper
intestinal malaise, animals and humans develop a very strong
aversion to the flavor. This is an unusual form of Pavlovian
conditioning in that it occurs with only one association between
the flavor stimulus and illness, it occurs with very long intervals
between these stimuli, and it is unusually resistant to forgetting.
Stress Anorexia
The classical hypothalamic-pituitary-adrenal axis stress
response involves a three-step endocrine cascade: secretion of
CRH from the hypothalamus into the hypophyseal portal circu-
lation, secretion of adrenocorticotropic hormone (ACTH) from
the anterior pituitary into the systemic circulation, and secretion
of glucocorticoid hormones, especially cortisol in humans and
corticosterone in rats, from the adrenal cortex. Stress also stimu-
lates secretion of arginine vasopressin and OT from the posterior
pituitary and activates the sympathetic nervous system, leading
to the secretion of epinephrine and norepinephrine from the
adrenal medulla. All of these hormones appear to contribute to
stress-induced anorexia, although their specific roles in the
many types of stressors that are associated with anorexia are
not fully clear.
CRH is secreted into the hypophyseal circulation by neu-
rons in the parvocellular part of the PVN. CRH neurons also
project to other brain sites, where CRH is released as a neuro-
transmitter, as mentioned above. These neurons can be acti-
vated in the absence of HPA axis activation, suggesting that
CRH has a role in normal eating as well as anorexia. Similarly,
as OT also acts as a neurotransmitter, and oxytocinergic projec-
tions from the PVN to the NTS also appear to be involved in
the control of normal eating.
Glucocorticoid hormones play a crucial role in obesity -
most experimental obesity syndromes are reversed by
adrenalectomy and reinstated by glucocorticoid treatment.
Pathological increases in CRH or ACTH lead to Cushing's
syndrome, which includes abdominal obesity. Both overeating
and direct metabolic effects of glucocorticoids appear to con-
tribute to the obesity of Cushing's syndrome.
Psychiatric Eating Disorders
Anorexia Nervosa
Anorexia nervosa is a grave disorder characterized by danger-
ously low body weight (often defined as BMI < 17), an intense
fear of weight gain, refusal to eat and, in women, amenorrhea.
A substantial number of people who develop anorexia nervosa
die from it. The name, however, is a misnomer. That is,
although people with anorexia refuse to eat, it is because of
suppression of appetite, not lack of appetite. This is evidenced,
for example, by findings that people with anorexia nervosa
often consume huge amounts of noncaloric sweetener. The
prevalence of anorexia nervosa is approximately threefold
higher in women than in men. Furthermore, the incidence of
anorexia nervosa is highest around puberty, when the estro-
genic inhibition of eating appears. Unfortunately, therapeutic
manipulation of the physiological controls of eating to allevi-
ate anorexia nervosa remains a distant goal.
Bulimia Nervosa and Binge-Eating Disorder
Bulimia nervosa and binge-eating disorder are characterized by
binge eating (i.e., ingestion of abnormally large amounts of
food associated with a perceived loss of control of eating); in
bulimia nervosa, binges are followed by inappropriate com-
pensatory behaviors, such as self-induced vomiting or laxative
abuse, with maintenance of normal weight. In binge-eating
disorder, there is not compensation, so obesity develops.
There is considerable evidence that some satiation signals are
less potent in patients with bulimia nervosa. For example, food
preloads decrease meal size less in patients with bulimia ner-
vosa than in controls, patients with bulimia must eat larger
amounts of food to produce equivalent self-reports of fullness,
and food-stimulated CCK release is less in patients with
bulimia nervosa than controls. There is also evidence that
serotonergic processing of satiation signals in the brain is
altered in patients with bulimia nervosa. All these abnormal-
ities appear to resolve as bingeing decreases, however, suggest-
ing that they are not the initial cause of bulimia. It is likely
that they facilitate the maintenance of bulimia nervosa once it
has begun and impede recovery from it. Thus, these are attrac-
tive candidates for development of physiologically or pharma-
cologically based treatments. The risk of both disorders is
substantially higher in women (approximately threefold for
bulimia nervosa and approximately twofold for binge-eating
disorder). Binge-eating disorder, which was recognized more
recently, urgently requires investigation as it has a strong statis-
tical association with morbid obesity (i.e., BMI > 40).
Animal Models of Psychiatric Eating Disorders
A number of animal models have been developed in the
attempt to understand normal genetic, neurochemical, physi-
ological, and behavioral factors that are involved in the cause
or maintenance of psychiatric eating disorders. Activity-based
anorexia (ABA) is the most interesting animal model of
anorexia nervosa. ABA is produced by limiting food access to
l-2hday _1 and providing free access to a running wheel.
When this is done, rats become progressively anorectic until
death. The neurochemical changes produced by ABA have been
extensively investigated and have provided several clues as to
potential pharmacological treatments for anorexia nervosa.
Animal models of binge eating, of which there are several,
also seem to hold great promise. These models are based on
increased intake of palatable foods stimulated by restricted
access, stress, and other factors thought to be involved in the
development of the human syndrome.
1 98 Appetite
Genetics of Eating
Heritability of Obesity
Body weight is a highly heritable trait. The heritability (h 2 , the
percent of variation in a population phenotype that is due to
genetic variation) of body weight or BMI is generally ~75%,
which is similar to that of height. A recent large (>90 000
subjects) genome-wide association study revealed eight genes
that contribute to this high heritability. Some of these are
related to eating, such as the gene for the a-MSH receptor, the
melanocortin-4 receptor (MC4R). About 100 MC4R variants
have been associated with obesity. People bearing such gene
variants report constant hunger and apparently lack a normal
sense of satiation. MC4R variants are quite common, occurring
in about 2-6% of obese people, depending on the popula-
tion studied, and one variant had a minor allele frequency of
about 25% in a cross-sectional study of US American women
of all weights.
A common question is, how can the recent increase in the
prevalence of obesity, which clearly occurred without any
change in the human genome, be reconciled with the high
heritability of body weight? The problem evaporates when
one recalls that h 2 measures heritability in a single environ-
ment. If the environment changes, so will h 2 . Thus, individual
estimates of h 2 in the present high obesity-risk environment
and in the former low obesity-risk environment might each be
~75%, but this would be markedly reduced if h 2 were estimated
in both environments simultaneously. Second, although the
genome has not changed much in recent times, the particular
genes that are expressed and their degree of expression probably
have. That is, different genes are likely to contribute effectively to
our phenotypes in the present environment more than in for-
mer environments. Genes whose expression is related to sugar
and fat intake, for example, would certainly be expressed more
now than in former days, when sugar and fat intakes were lower.
Genes and Eating
Fascinating links between genetic variation and eating have
begun to emerge. These new findings suggest, among other
things, that, at least in some individuals, overeating and obesity
may be caused by inborn genetic differences, no more under the
person's control than common inherited variations in color
vision. Here I describe two interesting examples in addition to
MC4R, described above, that support this conclusion.
First, Mariken de Krom, Roger Adan, and their colleagues at
the University Medical Center, Utrecht, NL, identified women
who were obese and who overate, as estimated using photos of
food portions, and then searched for the presence of known
variants in the cck, lep, and lepr genes. Women with one particu-
lar lep variant had markedly increased risk for extremely frequent
meals, whereas those with a particular cck variant had increased
risk for extremely large meals. These data indicate that specific
high-risk eating traits can have genetic bases. Furthermore, the
variants identified were surprisingly common.
Second, Eric Stice and his colleagues at the Oregon Research
Institute, Eugene, OR, USA, combined genetic and functional
imaging approaches to link dopamine neurotransmission with
food hedonics, eating, and body weight. Subjects were people
with or without a variant of a gene (rsl800497) that reduces
dopamine D2 receptor function. Activation of the striatum, a
telencephalic area in which dopamine neurotransmission con-
tributes to the processing of food hedonics, was measured with
fMRI while subjects ate a chocolate drink or a tasteless solu-
tion. They found that eating chocolate activated the striatum
less in subjects with the gene variant in comparison to subjects
with other forms of the gene and that the degree of activation
of the striatum was significantly associated with BMI. Their
interpretation was that the gene variant reduced the hedonic
impact of foods, leading people to eat more in order to get the
desired or normal hedonic experience (a similar phenomenon
is well known in drug- reward studies).
See also: Behavioral Medicine; Bulimia Nervosa; Homeostasis;
Hormones and Behavior.
Further Reading
Berridge KC and Kringelbach ML (2008) Affective neuroscience of pleasure: Reward in
humans and animals. Psychopharmacology (Berlin) 199: 457-480.
Berthoud H-R and Morrison C (2008) The brain, appetite, and obesity. Annual Review of
Psychology 59: 55-92.
Cummings DE and Overduin J (2007) Gastrointestinal regulation of food intake. Journal
of Clinical Investigation 117: 13-23.
de Krom M, Bauer F, Collier D, Adan RA, and LaFleur SE (2009) Genetic variation and
effects on human eating behavior. Annual Review of Nutrition 29: 1-22.
Geary N and Moran TH (2009) Basic science of appetite. In: Sadock BJ, Sadock VA, and
Ruiz P (eds.) Comprehensive Textbook ot Psychiatry, 9th edn, pp. 375-386.
Philadelphia, PA: Wolters Kluwer/Lippincott Williams & Wilkens.
Grill H (2006) Distributed neural control of energy balance: Contributions from
hindbrain and bypothalamus. Obesity 14(Suppl): 216S-221S.
Hillebrand JJG and Geary N (2010) Do leptin and insulin signal adiposity? In:
Langhans W and Geary N (eds.) Frontiers in Eating and Weight Regulation,
pp. 111-122. Basel: Karger.
Kas MJH, Kaye WH, Foulds Mathes W, and Bulik C (2009) Interspecies genetics of
eating disorder traits. American Journal of Medical Genetics 150B: 318-327.
Langhans W (2010) The enterocyte as an energy flow sensor in the control of eating.
In: Langhans W and Geary N (eds.) Frontiers in Eating and Weight Regulation,
pp. 75-83. Basel: Karger.
Langhans W and Geary N (2010) Overview of the physiological control of eating.
In: Langhans W and Geary N (eds.) Frontiers in Eating and Weight Regulation,
pp. 9-53. Basel: Karger.
Leibel RL (2008) Moleclular physiology of weight regulation in mice and humans.
International Journal of Obesity (London) 32(Suppl 7): S98-S108.
Lutz TA (2010) Roles of amylin in satiation, adiposity and brain development.
In: Langhans W and Geary N (eds.) Frontiers in Eating and Weight Regulation,
pp. 64-74. Basel: Karger.
Moran TH (2010) Hypothalamic nutrient sensing and energy balance. In: Langhans W
and Geary N (eds.) Frontiers in Eating and Weight Regulation, pp. 94-1 01 .
Basel: Karger.
Morris MJ, Na ES, and Johnson AK (2008) Salt craving: The psychobiology of
pathogenic sodium appetite. Physiology and Behavior 94: 709-721.
Morton GJ, Cummings DE, Baskin DG, Barsh GS, and Schwartz MW (2006) Central
nervous system control of food intake and body weight. Nature 443: 289-295.
Searle J (2004) Mind: A Brief Introduction. New York: Oxford University Press.
Smith GP (2008) Critical introduction to obesity. In: Blass EM (ed.) Obesity, pp. 1-17.
Sunderland, MA: Sinauer.
Stice E, Spoor S, Bohon C, and Small DM (2008) Relation between obesity and
blunted striatal response to food is moderated by TaqIA A1 allele. Science
322: 449-452.
Asperger’s Syndrome and Nonverbal Learning Disorder
J M Gillis, M Sevlever, and M Roth, Auburn University, Auburn, AL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Adaptive skills Living skills needed to live, work, and play
in the community.
Crossmodal integration Synthesizing information from
different sensory modalities (tactile, gustatory, visual,
auditory, and olfactory).
Pragmatics The social use of language; both verbal and
nonverbal.
Semantics The knowledge and usage of words; one's lexicon
of vocabulary.
Social reciprocity The back and forth of interactions such as
playing, gesturing, attention, and conversations.
Introduction
Asperger's syndrome (AS; sometimes referred to as Asperger's
disorder) and nonverbal learning disorder (NLD; sometimes
referred to as nonverbal learning disability or nonverbal
learning disability syndrome) are first identified in childhood
and are considered neurological disorders, meaning that the
way the brain processes information is affected. These condi-
tions share a substantial number of symptoms and characteris-
tics; however, there are certain features that distinguish these
conditions from one another. Understanding the overlapping
and distinctive genetic, neurological, cognitive, and behavioral
characteristics of AS and NLD is the focus of current researchers
across multiple scientific disciplines. There is conflict as to
whether these are two separate disorders; that is, whether they
are similar enough such that they belong to the same diagnostic
continuum. There are no clear definitions of NLD, as compared
to AS, which results in relatively more familiarity with and
information available on AS compared to NLD.
The diagnostic classification systems commonly used by
practitioners (such as psychologists, medical doctors, etc.) to
diagnose psychological and psychiatric disorders include the
Diagnostic and Statistical Manual of Mental Disorders, 4th
edition, Text Revision (DSM-IV TR) and the International
Classification of Diseases, 10th edition (ICD-10). AS was first
introduced into these classification systems in 1993 (ICD-9)
and 1994 (DSM-IV). Despite NLD's absence in these two diag-
nostic systems, children continue to be diagnosed with NLD,
especially within school settings, and there is increasing research
on NLD. Practitioners and researchers currently conceptualize
these two conditions as belonging to different diagnostic cate-
gories; AS is considered a pervasive developmental disorder,
whereas NLD is classified as a learning disability. This is due in
part to the available research on these disorders. Examination of
the diagnostic criteria can be helpful in understanding the
overlap and distinguishing characteristics of each condition.
Diagnosis of Asperger’s Syndrome
AS was first described by Hans Asperger, an Austrian psycholo-
gist, in 1944. Asperger's early descriptions of AS continue
to correspond with current definitions of the disorder. Accord-
ing to Asperger, children with AS display a pattern of social,
communicative, and behavioral difficulties, including a lack of
empathy, difficulty forming friendships, a tendency to engage
in one-sided conversations, intense and limited interests, and
poor motor coordination skills. To date, there is substantial
overlap between the diagnostic descriptions of AS in both the
DSM-IV TR and ICD-10.
Within the DSM-IV TR, AS is subsumed under the category
of pervasive developmental disorders (PDD), which also
includes autistic disorder (more commonly known as Autism),
Rett's syndrome, child disintegrative disorder, and pervasive
developmental disorder, not otherwise specified (PDD-NOS).
Broadly, the criteria for a diagnosis of AS identified by the
DSM-IV TR and ICD-10 include impairments in social interac-
tion (often including social communication impairments)
as well as the presence of restricted, repetitive, or stereotyped
interests, activities, or behaviors. AS is distinguished from
Autism in the area of language development. Specifically, indi-
viduals with AS do not present with a delay in or absence of
language development before the age of 3, which is a require-
ment for the diagnosis of Autism. There is ongoing research
to increase knowledge about important similarities and differ-
ences between AS, autistic disorder, and PDD-NOS. For exam-
ple, it is often difficult to differentiate persons with AS from
individuals with Autism who demonstrate average to high
average intellectual abilities. As such, there is significant debate
among researchers and practitioners as to whether HFA and AS
are distinct psychological disorders or manifestations of the
same disorder across a spectrum of severity.
As part of the established research literature and the process
of updating diagnostic classification systems, proposed revi-
sions for new DSM-V (to be released in 2013) suggest remov-
ing the diagnosis of AS and subsuming three of the pervasive
developmental disorders (i.e., autistic disorder, Asperger's dis-
order, and pervasive developmental disorder-not otherwise
specified) under the broader category of autism spectrum dis-
order. Under these revisions, the level of severity of an autism
spectrum disorder would be indicated based on the extent of
social deficits, the presence of a language delay, and the level
of intellectual functioning. Despite issues related to the diag-
nosis of AS and whether or not it will be recognized as a
separate disorder in the future release of DSM-V, inclusion of
AS in the DSM-IV TR and ICD-10 has led to exponential
growth in research and knowledge related to the disorder and
enhanced our understanding of autism spectrum disorders
199
200 Asperger’s Syndrome and Nonverbal Learning Disorder
more generally. Interestingly, this is one reason why awareness
and research on NLD may be more limited at the present time.
Diagnosis of NLD
In contrast with AS, NLD are not represented in the DSM-IV or
ICD-10, which can lead to confusion regarding how NLD is
defined in research and clinical or school settings. Learning
disability is a broad diagnostic category in the DSM-IV TR and
only encompasses disorders that describe learning difficulties
in the areas of reading, math, and writing (not NLD). The term
NLD was first used by Helmer Myklebust in 1975 and is often
described as a subtype of learning disabilities within the field
of neuropsychology. NLD is used to refer to a profile of neuro-
psychological strengths and weaknesses, and these neuropsy-
chological impairments are similar to AS in that they are
associated with poor understanding of social situations. In
the mid-1990s, Dr. Byron Rourke developed a model that
includes the primary strengths and deficits associated with
NLD. The major strengths that many individuals with NLD
possess include strong auditory perception skills, verbal skills,
and memory skills related to verbal information. The three
primary deficits of NLD include tactile perception, motor coor-
dination, and visual perception. According to Rourke's model,
these deficits lead to marked impairments in social perception
and awareness (similar to those seen in AS, but often to a lesser
degree of severity). For example, due to difficulties in organiz-
ing and interpreting visual information, individuals with NLD
are often unable to interpret nonverbal cues in their environ-
ment, and thus may have difficulty interpreting social gestures
and facial expressions. Children with NLD are often incorrectly
diagnosed with attention deficit hyperactivity disorder due to
their difficulty focusing on tasks and hyperactivity; however, as
individuals with NLD age, they often become much less physi-
cally active (hypoactive).
Common Characteristics of AS and NLD
NLD and AS are described as neurological disorders, meaning
that specific brain processes are impaired or dysfunctional and
that these processes in turn affect one's behavior and how one
interacts with the world. Even though there are characteristics
that distinguish AS from NLD, these conditions do share a
significant number of common features, which impact aca-
demic performance, executive function, tactile and visual per-
ception, motor function, as well as social skills.
Academic Performance
Because individuals with AS and NLD present with relatively
strong verbal skills, they may appear to be more competent in
academics even though there is not commensurate perfor-
mance in the classroom. High expectations can be placed on
both groups; however, due to some of the characteristics of
these disorders, academic problems can arise. For example, due
to difficulty with fine motor skills, handwriting is problematic
in AS and NLD. Additionally, sometimes math is identified
as a weakness since mathematical concepts are more abstract,
nonverbal, and spatial in nature. At a younger age, children
with NLD or AS may be able to memorize formulas and com-
plete simple calculations; however, their performance deterio-
rates when reasoning skills and word problems are introduced.
Individuals with AS and NLD also demonstrate several
academic strengths. For one, strengths in recalling factual
information are common due to well-developed rote memory
skills. Additionally, word recognition and basic reading skills
(i.e., decoding words, phonemic awareness) are well developed.
Individuals with NLD and some with AS are also known to have
strong spelling, short-term memory, and auditory-perceptual
skills. However, comprehension of abstract and inferentially
based information is usually impaired in both groups (i.e.,
reading comprehension, word problems, idioms, analogies,
and metaphors). As more abstract understanding is required
of students, and the value of memorizing factual information
decreases, the academic success of individuals often declines.
Executive Functioning
Many skills associated with executive functioning (i.e., the abil-
ity to plan, organize, shift attention, and set goals) are found to
be impaired in AS and NLD, and may result in academic
difficulties. Difficulty in organizing information can make
planning, carrying out work, and breaking down tasks into
components challenging. Executive functioning deficits also
make cause and effect relationships difficult to understand.
Furthermore, poor problem solving is common in NLD and
AS; tasks are approached with inflexibility, inhibiting the abil-
ity to generate alternative options. Additionally, individuals
with AS and NLD have problems generalizing newly learned
techniques and behaviors to different tasks and settings.
Some have suggested that individuals with AS have diffi-
culty discerning relevant from irrelevant stimuli. This is known
as the weak central coherence theory. Central coherence can be
thought as the ability to process different levels of information
together to generate overall meaning and understanding in an
immediate context. Thus, this theory suggests, individuals with
AS tend to attend to detail rather than the 'big picture.' Despite
the ability of individuals with NLD to also recall specific details
of verbal information rather than drawing inferences (e.g.,
seeing the forest through the trees), more research related to
weak central coherence is needed to better understand this
phenomenon in NLD.
Tactile and Visual Perception
Individuals with NLD are known to have deficits in tactile
and visual perception, discrimination, and attention. Visual-
spatial organization impairments are a hallmark of this disor-
der. Although individuals with NLD may be able to perform
simple tasks requiring these skills (i.e., matching items of a
similar shape), they may have more difficulty as tasks become
more complex (i.e., finding a figure embedded in a picture or
putting blocks in a form board while blindfolded). Due to
tactile and visual perception deficits, individuals with NLD
are less likely to physically explore their environment. Addi-
tional neuropsychological characteristics of NLD include pro-
blems with time perception; individuals with NLD often have
difficulty placing themselves in space and orientation. Tactile
Asperger’s Syndrome and Nonverbal Learning Disorder 201
and sensory disturbances are also well noted in AS. Individuals
with AS are known to be hypo- or hypersensitive to visual,
auditory, gustatory, olfactory, and tactile stimuli. Thus, they
often experience sensory information at a higher or lower
degree than typically developing individuals. For example,
difficulty in modulating sensory information can lead to indi-
viduals with AS avoiding situations where the stimuli may be
encountered (e.g., avoiding crowded rooms in an attempt to
escape loud noises). Although hypersensitive individuals may
attempt to avoid certain types of sensory information, hypo-
sensitive individuals with AS may engage in repetitive behavior
(e.g., humming, hand flapping), in order to experience sensory
stimulation.
Motor Functioning
Motor problems are characteristic of both AS and NLD. Deficits
in fine motor (small muscle movements) and gross motor
(large muscle movements) skills can cause individuals with
AS and NLD to appear 'clumsy' and 'awkward.' Additionally,
problems with coordination and balance are known to occur.
Simple motor tasks can be performed; however, more complex
tasks are challenging. As noted earlier, motor problems can
make handwriting difficult. Complex motor tasks are also
difficult for individuals with AS and NLD, such as playing
sports. Visual-motor coordination or synthesizing visual infor-
mation with motor output is known to be impaired in NLD
and some individuals with AS. This deficit in crossmodal inte-
gration (synthesizing the information from different senses)
can make tasks such as completing a puzzle challenging.
Social Skills
Individuals with AS and NLD often demonstrate poor social
competence due to their inability to perceive and interpret the
social interactions, social communication, and behavior of
others. Individuals with AS and NLD may appear 'socially
awkward,' 'socially inappropriate,' or 'odd' to those around
them due to the combination of deficient social skills, impaired
social perception, and poor social communication skills. As a
result, many individuals with NLD and AS experience difficulty
in forming close friendships or relationships, even when hav-
ing these relationships are important to them.
Conversations can be characterized by a lack of social reci-
procity, commonly known as the back and forth of social
interactions. Attempts by another person to respond or change
topics may be unsuccessful due to problems with turn taking
(i.e., people are often interrupted). When conversations are
one sided, the individual with AS or NLD often dominates
the dialog in a pedantic manner (which can appear as if
he/she is showing off). Thus, it can appear that they are 'talking
at you' rather than 'talking to you.' Likewise, language can
be overly formal and full of jargon. In addition, language
impairments are observed both in AS and in NLD that interfere
with social interactions. Notably, problems with semantics
(the knowledge and use of different words) and pragmatics
(the social use of language) are present. This can lead to diffi-
culties interpreting the subtleties of communication, such as
humor, sarcasm, and metaphors. Additionally, individuals
with AS and NLD demonstrate impairments with prosody,
the patterns of stress and intonation, which cause them to
sound monotonous or make it difficult for others to under-
stand their emotional state based on the tone of their voice. In
AS, speech may be at a rapid rate or higher volume. Since it is
difficult for individuals to use language in a social context,
often learned and rote, or memorized, responses are provided
even when individuals realize that their repertoire of responses
are not appropriate in a number of social situations. Children
with AS and NLD may be more successful interacting with
adults or even younger peers, as these social situations are
generally more structured, predictable, and forgiving.
Individuals with AS and NLD also have difficulty with the
nonverbal aspects of the social environment (e.g., reading
facial expressions, body language, and gestures). In NLD and
AS, these problems can be attributed to the visual-perceptual
deficits that are characteristic of the disorder. Another explana-
tion of these deficits is an impairment in theory of mind (the
ability to recognize and understand the thoughts, desires,
beliefs, and intentions of others); however, the research is
mixed and this remains to be clarified. An inability to read
the intentions of others might lead individuals with AS and
NLD to inappropriately plan their own behaviors in social
situations. For both populations, problems with social judg-
ment, inferring intentions of others, and reading nonverbal
communication lead to an increased vulnerability to peer rejec-
tion and bullying. Although social skills may improve over
time from childhood, they often continue to remain problem-
atic into adulthood.
Difficulty in navigating the social world and repeated
unsuccessful attempts to form peer relationships can cause
low feelings of self-worth and self-esteem in individuals with
AS and NLD. Among the most common comorbid (i.e., co-
occurring) psychiatric conditions for individuals with AS and
NLD are internalizing problems (e.g., depression and anxiety).
In AS, increased self-awareness of differences between them-
selves and their peers contributes to the development and
maintenance of internalizing problems.
Individuals with AS and NLD have difficulty adjusting to
novel situations and environments. In NLD, these issues may
arise due to deficits in nonverbal problem solving, flexibility,
and the ability to learn from environmental feedback. They
may approach novel situations with a reliance on rote beha-
viors, which in many cases may be inappropriate. Conversely,
problems recognizing familiar situations can lead to approach-
ing the same situation with different responses. Some research-
ers hypothesize that behavior problems emerge as a result of
perceiving the world as unpredictable or 'fast-paced.' Thus, due
to deficits in social perception, individuals may feel constantly
bombarded by unpredictable and confusing information.
Additionally, a lack of emotional regulation, the ability to main-
tain and control emotional states, can contribute to behavior
problems. Since individuals commonly have preference for rou-
tine, they often exhibit problem behaviors when these routines
are altered or broken. Moreover, impairments in identifying and
interpreting emotions can lead to frequent distressing events.
For example, 'tantrums,' 'meltdowns,' and withdrawal from
situations can often occur when children become upset.
Other behavior characteristics of AS and NLD include diffi-
culty with adaptive skills, which are skills required to live,
work, and play in the community. Examples of adaptive skills
202 Asperger’s Syndrome and Nonverbal Learning Disorder
that may be deficient in AS and NLD include dressing, diet,
hygiene, household chores, managing money, and shopping.
Deficits in adaptive skills may continue into adulthood if not
taught properly in childhood and adolescence. Furthermore,
in NLD, individuals may show hyperactivity in childhood;
however, this behavior decreases with age.
Distinguishing Characteristics of AS and NLD
The characteristics that distinguish AS from NLD are debated
by researchers and practitioners due to different theoretical
orientations, limited research, and conceptualizations or defi-
nitions of the disorders. Given the relatively recent research on
both disorders, there is not yet a definitive answer. However,
some differences that have yielded more support include cog-
nitive profile of nonverbal compared to verbal intelligence,
severity of social impairment, and circumscribed interests.
One of the most significant features of NLD is the disparity
between verbal intelligence (language-based abilities) and
nonverbal intelligence (visual-spatial abilities), with verbal
intelligence being relatively stronger. This is not surprising
given the visual-spatial deficits characteristic of NLD. Even
though there is conflicting evidence and opinions, it appears
that not all individuals with AS present with a similar cognitive
profile (a higher verbal intelligence score relative to nonverbal
intelligence); however, when individuals with AS demonstrate
a similar cognitive profile as NLD, the individual is sometimes
given a diagnosis of NLD in addition to AS.
Social impairment is a defining characteristic of both AS
and NLD. In general, individuals with AS and NLD have
deficits in social behaviors and interactions, as well as poor
social communication skills. What distinguishes the two dis-
orders is level of severity: individuals with NLD show similar
social difficulties as AS, but to a much lesser degree. A social
communication difficulty often observed in AS occurs when
circumscribed interests are discussed to a long and inappropri-
ate extent ( and at times in a pedantic manner) . A circumscribed
interest is a topic that is of intense focus. Often, a large amount
of information on this topic has been acquired by the individ-
ual with AS, regardless of age, meaning that young children
in addition to adults with AS exhibit circumscribed interests
that might interfere with social interactions. These interests
can range considerably from a well-known topic such as a
popular television show to a more obscure topic such as loca-
tions corresponding to zip codes around the world. The
research is mixed on whether circumscribed interests are char-
acteristic of NLD.
Etiology
At present, the exact cause or etiology of AS and NLD is
unknown. There is ongoing research examining the role of
genetics in both, but no consistent findings have yet emerged.
There has been and continues to be extensive efforts to deter-
mine the neurological characteristics of AS and NLD. Distur-
bances in the development of, or damage to, white matter in
the right hemisphere of the brain have been found to be the
main neurologic characteristic that may be responsible for the
manifestation of NLD. The right hemisphere of the brain
houses the ability to process information simultaneously and
the ability to synthesize novel information and has important
neural connections to different areas of the brain. It has been
suggested that the perceptual and motor deficits characteristic
of NLD are more marked on the left side of the body because of
this right hemispheric dysfunction (information to and from
the brain travel contralaterally to the other side of the body).
It has been reported that some conditions that have deleterious
affects on the brain's white matter have led to NLD symptoms
(e.g., callosal agenesis, hydrocephalus, multiple sclerosis, sei-
zures, head injuries, tumors, etc.). Conversely, there is not yet a
clear understanding of the brain structure in AS despite the
efforts of researchers. Abnormalities in the prefrontal cortex,
temporal, striatal, and cerebellar regions of the brain and
gray matter have all been identified by researchers as related to
a potential cause of AS; however, there is not yet a consensus due
to a lack of consistent findings. Thus, disruption to several areas
of the brain, rather than a focal lesion, is thought to contribute
to the development of AS. Even with the use of advanced brain
imaging technology, many studies have resulted in inconsistent
results, and different areas of the brain have been implicated.
Nevertheless, strong evidence exists to suggest AS is at least
partially genetically determined and caused by dismption to
brain regions important for social and emotional functioning.
Prevalence
Prevalence is a term used to refer to the proportion of the
population diagnosed with a certain disease at a specific time.
Prevalence estimates can vary dramatically across studies due
to differences in how the studies were conducted or the year in
which they were conducted. This is especially true for AS since
the prevalence of AS appears to be rising over the years. The
prevalence of AS has ranged from 0.3 per 10 000 children to
8.4 per 10 000 children. AS is now reported as occurring in
20 per 10 000 children. This is equal to 1 in 500 children in the
United States. As NLD is not an official diagnostic category
within the DSM-IV or ICD-10, the prevalence of NLD has
been difficult to estimate. Approximately 3% of the total
United States' population is diagnosed with a learning disabil-
ity, and of those diagnosed with a learning disability, ~5-10%
are diagnosed with NLD. Some estimates suggest that about
2.7 million people have NLD in the United States (or 1% ofthe
population). With regard to differences in prevalence between
genders, AS is much more common in males than females:
For every nine males diagnosed with AS, one female is diag-
nosed with the disorder. Currently, it is suggested that NLD
occurs in males and females at similar rates (i.e., no gender
differences are reported).
Treatment Considerations
Although NLD is not currently part of the DSM-IV or ICD-10,
practitioners do diagnose NLD and individuals with NLD
are often eligible to receive special education services as are
individuals with AS. Currently, there is no biological or phar-
macological intervention that improves either disorder in its
Asperger’s Syndrome and Nonverbal Learning Disorder 203
entirety. Rather, medication is sometimes prescribed to man-
age specific symptoms (e.g., hyperactivity, concentration diffi-
culties, anxiety, etc.) experienced by individuals with AS or
NLD. Although there is no cure available for AS or NLD,
there are several interventions that can help individuals with
AS or NLD learn new skills, overcome challenges of their
disorder, and adapt to different environments in order to
successfully engage in these environments using educational,
behavioral, or cognitive-behavioral approaches. Given that the
consequences of having AS or NLD surround academic and
social functioning, the focus of treatment within the school
setting might include addressing poor organizational skills,
information processing deficits, difficulty in integrating and
adjusting to new information or routines, stress/anxiety in
social situations, poor social competence, deficits in compre-
hension and abstract reasoning, attention and/or concentra-
tion difficulties, and time management.
See also: Nonverbal Communication; Reading and Phonological
Processing; Word Retrieval.
Further Reading
Collins DW and Rourke BP (2003) Learning-disabled brains: A review of the literature.
Journal of Clinical and Experimental Neuropsychology 25: 101 1-1034.
Gunter HL, Ghaziuddin M, and Ellis HD (2002) Asperger syndrome: Tests of right
hemisphere functioning and interhemispheric communication. Journal of Autism
and Developmental Disorders 32: 263-281 .
Nyden A, Niklasson L, Stahlberg 0, et al. (2010) Adults with Asperger syndrome with
and without a cognitive profile associated with ‘non-verbal learning disability.'
Research in Autism Spectrum Disorders 4: 612-618.
Rourke BP (ed.) (1995) Syndrome of Nonverbal Learning Disabilities:
Neurodevelopmental Manifestations. New York: Guilford Press.
Tsatsanis KD (2004) Heterogeneity in learning style in Asperger syndrome and
high-functioning autism. Topics in Language Disorders 24(4): 260-270.
Tsatsanis KD and Rourke BP (2003) Syndrome of nonverbal learning disabilities: Effects
on learning. In: Fine AH and Kotkin R (eds.) Therapists Guide to Learning and
Attention Disorders, pp. 109-145. New York: Academic Press.
Tsatsanis KD and Rourke BP (2008) Syndrome of nonverbal learning disabilities in
adults. In: Wolf L, Schreiber H, and Wasserstein J (eds.) Adult Learning Disorders:
Contemporary Issues, pp. 159-190. New York: Psychology Press.
Volkmar FR, Paul R, Klin A, and Cohen DJ (eds.) (2005) Handbook of Autism and
Pervasive Developmental Disorders, 3rd edn., vols. 1 and 2. New York: Wiley.
Relevant Websites
http://www.nld-bprourke.ca/-Byron Rourke, PhD, on NLD.
http://www.ninds.nih.gov/disorders/asperger/detail_asperger.htm - Information page
for Asperger’s Disorder on the website of the National Institute of Neurological
Disorders.
http://www.nlda.org - Nonverbal Learning Disorders Association.
http://childstudycenter.yale.edu/index.aspx-Yale University Child Study Center.
Associative Learning
E Fantino and S Stolarz-Fantino University of California, San Diego, La Jolla, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Autoshaping The tendency for an organism to make contact
with a stimulus (CS) that reliably predicts food (US), as
when a pigeon begins to peck a lighted key that appears
shortly before food is presented.
Classical conditioning The process by which a conditioned
response - for example, salivation - comes to be elicited by a
previously neutral conditioned stimulus (CS) - for example,
a tone - through the association of the CS with an
unconditioned stimulus - for example, food.
Conditioned suppression When a previously neutral
stimulus (CS) is paired with an aversive stimulus (US),
presentation of the CS will disrupt ongoing operant
behavior.
Eliciting stimuli Events that invariably produce reflex
actions (or respondent behavior). For example, a puff of air
in the eye produces an eyeblink.
Extinction In classical conditioning, when the conditioned
stimulus (CS) is presented several times without the
unconditioned stimulus, the conditioned response ceases to
occur when the CS is presented.
Habituation A decline in the magnitude of an
unconditioned response after it has been repeatedly elicited
by an unconditioned stimulus.
Observing responses Responses that produce stimuli
correlated with schedules of reinforcement, but that have no
effect on the occurrence of reinforcement.
Rescorla-Wagner model A theory that clarifies the specific
conditions under which a conditioned stimulus (CS)
acquires control over a conditioned response through
pairing of the CS with an unconditioned stimulus.
Respondent behavior Behavior that is automatically
produced by a specific eliciting stimulus.
Spontaneous recovery In classical conditioning, the return
of a conditioned response after a delay following an
extinction period; in habituation, the resumption of
responding to a habituated stimulus following a rest
period.
Basic Concepts
Innate, automatic reactions that occur in all organisms are
called reflex actions or respondent behavior, and the events
that unvaryingly produce them are known as eliciting stimuli.
The experimental procedure known as Pavlovian conditioning
was developed from a study of reflexes carried out by the
Russian physiologist, Ivan Pavlov (1849-1936). Pavlov had
developed a technique to study digestion whereby he could
measure the fluids secreted in a dog's mouth and stomach. His
aim was to study the precise relationship between a stimulus
(food) and a response (salivation). But as his research on
digestion progressed, Pavlov encountered the unexpected: his
dogs were salivating prematurely, before they even saw the
food. In fact they would salivate as soon as Pavlov entered
the room. Pavlov then began to study this phenomenon which
he called 'conditioning.' Since Pavlov's time, the conditioning
process has been intensively investigated and applied experi-
mentally to behavior in addition to reflex actions. The subject
of Pavlov's initial research, however, is viewed as a distinctive
form of conditioning dealing mainly with reflexive behavior.
Thus, it is sometimes called classical conditioning. It is a funda-
mental form of associative learning - that is, the learning of a
relationship between stimuli. We turn first to the basic concepts
of classical conditioning.
Paradigm
The central principle of conditioning is the modification of an
organism's response by environmental events. Pavlov found
that after pairing a formerly neutral stimulus with a positive
stimulus such as food, the organism would come to react to the
previously neutral stimulus. Pavlov grasped the significance of
this observation and redirected his research effort to its study.
Perhaps a great deal of learned behavior might be based on
modification of innate reflexes, much as the learned behavior
of salivating to the sight of the experimenter developed from
the initial food-salivation reflex. Thus, Pavlov developed pro-
cedures for studying classical conditioning that remain in use
today. Pavlov's discovery also opened up a new way of study-
ing behavior. Pavlov described his dogs' reflexes to experimen-
tal stimuli without speculating about the dog's thoughts and
feelings. Thus he helped usher in the idea that some behavior
may be described and understood in terms of observable and
measurable reflexes, helping to lay the foundation for modern
behaviorism.
Figure 1 depicts the sequence of events in the experiments
conducted by Pavlov. A hungry dog was restrained in an appa-
ratus designed to isolate the dog insofar as possible from dis-
tracting stimuli. A tube leading from the dog's mouth carried
saliva directly from the salivary gland to a receptacle where it
could be measured. When food powder, the eliciting stimulus
or unconditioned stimulus (US) was placed in the animal's
mouth, the dog salivated reflexively. Salivation was the uncon-
ditioned response (UR) to food in the mouth. Pavlov then
struck a tuning fork one-half second before each presentation
of the food. After several pairings of tone and food, he would
occasionally sound the tuning fork without presenting food.
On these food-omission trials, the tone alone was sufficient to
elicit salivation to almost the same degree as when food was
actually placed in the dog's mouth. Thus, the originally neutral
stimulus, the tone, had become a conditioned stimulus (CS),
204
Associative Learning 205
ucs „ cs
(meat powder) (bell)
' , Association
► Conditioned
►- Innate
UCR (salivation) =CR (salivation)
Figure 1 Pavlov’s classical conditioning paradigm. See text for
details. Reproduced from Fantino E and Logan CA (1979) The
Experimental Analysis of Behavior: A Biological Perspective.
San Francisco: W.H. Freeman.
capable of eliciting a conditioned response (CR), salivation,
which it had not previously elicited.
It should be noted that research since Pavlov's era has shown
that the CR is not identical to the UR: classical conditioning is
not a simple process of substitution in which one stimulus (the
CS) merely replaces another stimulus (the US). Even when the
CR and UR appear to be the same (as in the salivation example),
they differ in detectable ways (e.g., in terms of latency, duration,
or magnitude of the response). In other situations, there is no
obvious relationship between the CR and UR. For example, in
one type of experiment, a rat is periodically subjected to the
occurrence of a light for 5 s (the CS), followed by a moderately
intense electric shock (the US) delivered to the rats' feet for 2 s.
The rat's normal response (the UR) to electric shock is to jump
and/or run around the cage. But after a few pairings of the light
and shock the rat begins to develop a CR to the light that is the
antithesis of running or jumping: it demonstrates rigid immo-
bility ('freezing') at the onset of the light. Freezing (the CR)
actually precludes simultaneous expression of the jumping or
running (the UR).
Parameters of conditioning
A crucial variable in classical conditioning is the time interval
between the CS and US. Conditioning works best when the CS
precedes the US by a Vi s or more whether or not the CS remains
on until US onset (delayed conditioning) or whether it turns off
some time before US onset (trace conditioning). Conditioning
works less well, or typically not at all, when the CS and US are
contemporaneous (simultaneous conditioning) or when the US
precedes the CS (backward conditioning). As with operant con-
ditioning if the CS is presented several times without the US,
extinction (EXT) occurs, that is the CR ceases to occur upon CS
presentation. But as with operant conditioning, if sufficient time
passes the CS will again elicit the CR (spontaneous recovery).
Relation of classical to operant conditioning
Whereas the principal relation in classical conditioning is that
between the CS and US (a stimulus-stimulus association), the
central relation in operant conditioning is that between
the response and the reinforcer contingent upon the response
(a response-stimulus association). The critical difference is
in the contingency: in classical conditioning, the US occurs
whether or not the CS elicits the CR; in operant conditioning,
the reinforcing stimulus occurs if and only if the response has
been emitted. That is, the reinforcer is contingent upon the
response in operant conditioning but the US is not contingent
upon the CR in classical conditioning.
Theory of associative learning
Elegant theories of the processes underlying classical condi-
tioning have been proposed. The most influential is that of
Robert Rescorla and Alan Wagner, developed while both were
at Yale University. Their theory clarifies the specific conditions
under which associative strength accrues to a stimulus when
CS-US pairings have occurred. Specifically, under what condi-
tions does contiguity of CS and US result in formation of an
association? The model rests on the assumption that the effect
of a reinforcer (US) depends not only on that reinforcer itself
but also upon the relationship between that reinforcement and
the reinforcement anticipated by the organism. In other words,
the organism's existing expectations form a context that mod-
ulates the effectiveness of the US. The following are the essen-
tial assumptions of the Rescorla-Wagner model:
1. Any given US can support only a given amount of condi-
tioning. After this point further US presentations will pro-
duce no additional behavior change. This parameter, the
asymptote of learning, differs for each distinct US, effec-
tively placing a ceiling on the magnitude of changes in
associative strength.
2. The rate of learning supported by different USs also differs
with qualitative changes in the US.
3. When conditional stimuli (CSs) occur in compound, the
total associative strength of the compound is equal to
the sum of the associative strengths of each element of the
compound.
4. Changes in associative strength occurring with additional
experience are determined by the difference between the
current strength of the compound and the asymptote of
learning (first assumption above) defined by a particular
US. Only a certain amount of change can occur. Thus, incre-
ments in associative strength with each successive CS-US
pairing will become progressively smaller, since each addi-
tional change insures that less change is potentially possible.
This final assumption most succinctly describes how the
associative strength of a CS will be altered during the course
of conditioning.
While this contemporary account of classical conditioning
is in accord with Pavlov's assignment of an informational, or
signaling, function to the CS, agreement on the means whereby
that function develops has changed dramatically. Mere conti-
guity of CS and US should no longer be thought of as necessary
or sufficient for conditioning. Instead, conditioning must be
considered in a larger, correlational context: only when a CS
and US are positively correlated will the CS acquire associative
strength. Thus, a CS in close temporal contiguity with a US (say
2 s) may not acquire associative strength if the US is as likely or
more likely to follow the absence of the CS within an equiva-
lent (2 s) period. At the same time, a CS that is less contiguous
with the US (say 5 s) may acquire associative strength if its
occurrence increases the likelihood that the US will follow.
This correlational view of conditioning, inspired by the
research and theory of Rescorla and Wagner, has a parallel
development in operant conditioning, particularly in the area
of conditioned reinforcement. We will see further evidence for
this contextual and correlational view of associative strength
throughout this article.
206 Associative Learning
Habituation
Many behaviors are modified simply as a result of repeated
exposure to a stimulus. When these repeated stimulations
result in a decrease in responsiveness, habituation has occurred.
Habituation is an important behavioral adjustment. An interest-
ing example of habituation occurs in relation to an organism's
ability to pay attention to new or unexpected stimuli in its
environment. Known as the orienting reaction, this behavior is
significant in that it helps the organism detect and prepare for
new and potentially dangerous events. It would be inefficient,
however, if the organism made an orienting reaction to every
occurrence of a previously novel stimulus. Instead, habituation
tends to occur; repetition of the once-novel stimulus without
any consequence results in a decline (even disappearance) of
the organism's responsiveness to that stimulus. It is a form
of learning in that the organism has, with repeated exposure to
the stimulus, learned not to respond to it. The decline in
response is temporary, however; after an interval during which
the habituating stimulus does not occur (from several minutes
to several days, depending on the particular situation), the
response to the stimulus returns. Habituation is an adaptive
phenomenon that occurs among a wide range of organisms
from those as relatively simple as primitive worms to more
complex organisms such as primates. Habituation can occur in
organisms without central nervous systems, for example sea
anemones. Sea anemones close and retract their tentacles when
water strikes them from above, a protective response that pre-
vents intense stimuli from harming the anemone. But sea ane-
mones also employ their tentacles to gather food in the form of
organisms residing in the ocean. Were anemones living in tide
pools to contract each time that they were struck by a wave or
each time a strong current of water passed over them, they would
lose the food gathering advantage of having their tentacles ready
to capture organisms that might be passing overhead. Thus, it is
adaptive for the sea anemone to cease responding (habituate) to
repeated presentations of water striking them from above. That
they in fact do so has been demonstrated in the laboratory; after a
number of repeated stimulations from a stream of water pre-
sented from above, the sea anemones stop contracting. The
recovery from habituation, noted above, has also been demon-
strated with these sea anemones: when water is not presented for
24 h, the next water stimulation produces a response almost as
intense as that elicited initially.
In addition to reinstating the original strength of a habitu-
ated response by allowing time to pass, an alternative tech-
nique is that of dishabituation, in which a novel stimulus is
presented. For example, if the habituated sea anemone is sti-
mulated with a glass rod instead of a water stream, there is an
immediate loss of habituation to the water stream: the next
several presentations of the water stream stimulation produce
contraction. This renewed responsiveness is short-lived, how-
ever; after a few water streams, the anemone again habituates.
Habituation occurs more rapidly on a second series of
stimulus presentations than on the first such series and occurs
still more rapidly on a third. This progressively more rapid
habituation suggests that habituation may not only represent
the simplest form of learning but also a primitive form of
memory.
Habituation is a form of simple nonassociative learning:
the organism learns that the repeatedly presented stimulus has
no significant consequence. This is distinct from associative
learning phenomena in which the association between stimuli
(classical conditioning) or response and reinforcement (oper-
ant conditioning) is centrally important. Habituation can be
considered as learning when not to respond. This can be vital
to the organism. For example, some insectivorous birds spend
90% of each winter day eating insects (catching them at a rate
of one every 2.5 s!) in order to sustain themselves. For these
birds, too much time responding to unimportant stimuli may
result in starvation. Research with human infants and adoles-
cents suggests that the ability to habituate to unimportant,
repetitive stimuli early in life maybe a predictor of later mental
ability and health.
Complex Phenomena
There are a plethora of phenomena based on associative
learning that have important implications for understanding
the behavior of organisms. In this section, we discuss research
involving biological constraints, taste-aversion learning, and
conclude with autoshaping. At this point, we will be ready to
address the important theoretical question: What are the nec-
essary and sufficient conditions for the occurrence of associa-
tive learning? We will do this by drawing on critical research in
the areas of autoshaping, conditioned suppression, and on the
determinants of observing and information seeking.
Biological Constraints
For decades, an implicit assumption of researchers in condi-
tioning was that a principle of equipotentiality held, at least to
an approximation. According to this principle, learning repre-
sents a general process in which the elements of the learning
situation, for example, the CS, the US, and the CR (or, in
operant conditioning, the stimulus, response, and reinforcer
or punisher) can be arbitrarily chosen for conditioning. The
specific quality of the elements was deemed unimportant.
According to this view, these elements are arbitrary and
completely interchangeable. They do not bear an innate fixed
connection to one another. Thus, two potent aversive stimuli
such as nausea-producing irradiation and electric shock should
be equally effective in supporting associative strength regard-
less of the stimulus with which they are associated (e.g., a
gustatory or auditory stimulus). We shall see in the next
section (on taste-aversion learning) the inadequacy of this
prediction. We will illustrate the limitations of the equipoten-
tiality principle here with an example from the extensive litera-
ture on avoidance learning. Then we will summarize the
contemporary view of the biological constraints position.
Avoidance learning
Avoidance is a type of negative reinforcement in which behavior
is strengthened because it avoids an aversive event such as electric
shock. In a typical avoidance acquisition procedure, a CS (say a
light) is followed a few seconds later by a US (say an electric
shock). After a few CS-US pairings, the previously neutral CS
acquires associative strength and the onset of the CS becomes
Associative Learning 207
aversive. In avoidance learning, the organism (most commonly a
rat) is given the opportunity to make a response that will simul-
taneously turn off the CS and avoid the US. Hence this response
is termed 'the avoidance response.' When the response is not
made, the US occurs and the same response can terminate
(escape) the US. Now according to the equipotentiality princi-
ple, the nature of the response should not matter. But it turns out
that the selection of the response has a profound effect on the
ability of the organism to acquire the avoidance response and
therefore avoid the shocks. If the response involves jumping up
to a platform, a rat will acquire the response in a handful of trials
(sometimes as few as one or two). If the response involves
running to another part of the apparatus, the rat's acquisition is
also fairly rapid. But if the response is a simple lever press, the rat
may not acquire the response for several hundred trials,
enduring painful shocks along the way. Thus, with respect to
the acquisition of the avoidance response, the critical relation is
between the required response and the reinforcer. How might we
best understand why this is so? One answer is that provided by
those who stress that there are biological constraints on the
acquisition process.
Consider the rat's normal response to an aversive, say pain-
ful, stimulus. Likely responses include freezing, running away,
and jumping. These are known as species-specific defense
reactions (SSDR). Thus running and jumping, being natural
responses to electric shock, are readily associated with avoid-
ance and are acquired swiftly. A more measured response
such as a chain pull is not a natural response to shock and is
therefore acquired more slowly. But by selecting a lever-press
response, researchers were unwittingly selecting a response that
was almost designed to fail. First of all, lever pressing is not
a natural response to shock. Second, the fact that the rat
can escape the shock once it begins, lever pressing keeps the
rat in the vicinity of the lever (actually crouching over it) when
the CS turns on. Freezing, a natural response to shock, pre-
dominates and freezing is incompatible with lever pressing.
In fact, rats show good acquisition of the lever-press response
when they are required to make the ostensibly more complex
response of running to another part of the chamber to press the
lever. Running, a natural response to shock, disrupts freezing.
At the end of the run, the rat presses the lever, avoiding shock.
Interestingly, the same lever-press response that is resistant
to acquisition when the reinforcer is avoidance of shock is
acquired rapidly when it produces food, another example of
how the nature of the response-reinforcer association is critical
in determining the extent and speed of acquisition of associa-
tive strength.
Challenge and limitations of the biological constraints
position
The focus on biological constraints on associative learning has
leveled two classes of criticism against traditional theories of
reinforcement and of associative learning. The first criticism is
that laboratory research on learning is artificial. Thus, the
principles of the acquisition and maintenance of behavior
that results from such research lacks generality for any natural
settings. Technically principles derived from such studies are
said to lack external validity. The second challenge is the claim
that traditional theories incorrectly assume that any emitted
response and any contingent reinforcer may be associated with
equal effectiveness (technically, the principle of equipotential-
ity, introduced above).
The first challenge does not negate the value of laboratory
research on the principles of associative strength and behavior
change. The natural environment imposes temporal con-
straints on accessibility to vital resources. The study of sche-
dules of reinforcement under rigorously defined laboratory
conditions may help determine how natural restrictions result
in changes in associative strength and in behavior change.
The focus on biological constraints is important in that it
correctly suggests that different patterns of responding may
emerge depending upon the nature of the species being studied
and on the particular selection of stimuli, responses, and rein-
forcers. But if we are to comprehend how the effects of response
and temporal partitioning are altered by biological constraints
on learning, we must also understand how they operate when
constraints are absent. Because the biological factors are demon-
strably potent, an arbitrary laboratory situation in which the
effects of these factors are minimized becomes a definitive
way to analyze the effects of temporal factors in isolation. The
emerging principles must, at some point, be integrated with the
constraints presumed to operate in the wild. To summarize,
laboratory studies have value for both because they reveal fun-
damental ways by which behavior change is effected by limiting
access to vital resources and because these effects can be
assessed most clearly under conditions in which the intrusion
of biological constraints are minimized.
The second challenge argues against the equipotentiality
position. This challenge is well taken in the sense that most
experimenters in the area appeared to harbor the belief that the
particular selection of stimulus, response, reinforcing event,
and species was relatively unimportant. However this assump-
tion was never made explicit by any of the great learning
theorists: in fact Edward Thorndike, B. F. Skinner, and others
explicitly stated that the principle was invalid. Indeed, well
before the issue of biological constraints on learning became
popular, Skinner acknowledged that a thorough appreciation of
the determinants of behavior change could not be accomplished
simply from the study of representative stimuli, responses, and
reinforcers. Recognition of this fact led Skinner to select stimuli,
responses, and reinforcers that would likely minimize intrusion
by complicating biological factors. He hoped that the resultant
behavior would be orderly and that its form would also describe
the behavior maintained by other stimuli, responses, reinforcing
events, and species. As Barry Schwartz has argued, the central
problem centers around appreciating which features of a phe-
nomenon are attributable to general associative principles and
which are attributable to situation-specific variables.
We conclude by noting that both the complexity and
diversity of behavior require that both phylogenetic and onto-
genetic principles are part of any complete explanation of
behavior. Whereas phylogenic factors may predominate
where insect behavior is concerned, ontogenetic (learned) fac-
tors may assume a more central role in the explanation of the
more generally flexible behavior of humans. The real challenge
is to achieve the correct mix of phylogenetic and ontogenetic
factors that best accounts for the behavior of individual mem-
bers of a particular species in a given circumstance.
208 Associative Learning
Taste-Aversion Learning
When discussing avoidance learning we showed that the ease
or difficulty of associative learning in avoidance depends criti-
cally on the interaction between the selected response and the
reinforcer. Research on taste-aversion learning instead empha-
sizes stimulus-reinforcer or stimulus-punisher interactions.
The bulk of the pioneering research on this topic, con-
ducted by John Garcia and his colleagues, was concerned
with an analysis of the conditions under which rats would
learn to avoid ingestible foods. In one classic experiment,
Garcia and Sam Revusky studied the differential suppression
of drinking as a function of the type of CS and US used. Four
groups of rats were served. For two groups, the CS was 'bright-
noisy (bubbly) water': when the rats drank from the water tube
gurgling sounds and flashing lights were presented. For the
other two groups, the water was flavored with sweet-tasting
but nonnutritive saccharin. For one group of the bright-noisy
rats and one group of saccharin rats drinking was followed
immediately with electric shock. Rats in the other two groups
received the sensation of nausea (produced by X-ray irradia-
tion). The results are shown in Figure 2: suppression of drink-
ing occurred only for rats that had the taste-nausea association
or the visual/auditory-shock association. Shocks following the
saccharin-flavored water and nausea following the bright-noisy
water were ineffective in suppressing drinking. Thus, the same
punisher (whether it was toxicosis or electric shock) was either
extremely effective or ineffective in suppressing drinking
depending on the stimulus modality with which it was asso-
ciated. This general finding has been replicated dozens of times
with various species. Other studies have shown that rats are
capable of associating novel exteroceptive stimuli with illness.
Thus, it may not be necessary to interpret the taste-aversion
findings in terms of natural predispositions to associate certain
stimuli with sickness. From the ecological perspective, it also
makes sense that the organism has a flexible response to
novelty.
Bright-noisy water Q Saccharin
Figure 2 Differential suppression of drinking as a function of the
type of cue and consequence used. See text for details. Reproduced from
Revusky S and Garcia J (1970) Learned associations over long delays.
In: Bower GH (ed.) The Psychology of Learning and Motivation, vol. 4,
pp. 1-84. New York: Academic Press.
Autoshaping
A hungry pigeon is placed in a standard pigeon chamber for the
first time. A circular key on the wall of this box is repeatedly
illuminated once every minute for a period of 6 s. When the
light turns off, the pigeon is presented with food. In this
procedure, food is delivered independently of the pigeon's
behavior vis-a-vis the circular key. That is, the food appears
even if the pigeon never approaches the key. Nonetheless, after
very few light-food pairings, the pigeon is likely to begin
pecking the key as soon as it is illuminated. This procedure
is called autoshaping, since the pigeon's key pecking is ac-
quired without the conventional shaping by differential rein-
forcement and successive approximations commonly used in
animal training. All that appears necessary is the simple associ-
ation between the light and the food.
The mechanism for autoshaping is well understood. Once
the pigeon pecks the lit key, it produces immediate reinforce-
ment of the key pecking. As for the first peck, that too is
unsurprising: the lit key is a highly salient stimulus and the
pecking response is a high probability one for the pigeon. More-
over, after repeated pairings with food, stimuli become likely to
promote food-related behaviors such as pecking, a phenomenon
that is the basis for instinctive drift, in which natural, prepotent
responses appear spontaneously in a food-related behavioral
sequence. For example, in 1937, the legendary American behav-
iorist B. F. Skinner found that a hungry rat that had been trained
to pick up a marble and deposit it down a chute in order to
obtain food began holding on to the marble instead. More
darkly, a porpoise at Marineland in Florida died playing a game
of 'catch,' when it swallowed a baseball that had been associated
with food. In the case of autoshaping, however, the key peck is
actually reinforced and is neither puzzling nor maladaptive.
But a more intriguing finding illustrates the robustness of
autoshaping, the power of the association between the stimu-
lus and food and a case where the key pecking is highly
maladaptive. David Williams, Alan Silberberg, and their col-
leagues largely reported these findings at the University of
Pennsylvania. The phenomenon has been called automainte-
nance, negative automaintenance, and food avoidance by
different investigators. This procedure differs from basic auto-
shaping in that pecks at the lighted key actually prevent the
next scheduled presentation of food. In other words, when the
key light is pecked instead of producing immediate food (auto-
shaping), the light turns off but no food is presented. The only
way to obtain food is by not pecking at the lit key. This proves
extremely difficult for most pigeons and rats. Some pigeons
acquire the pecking response to such a degree that they rarely
obtain food (hence food avoidance). What is the mecha-
nism for this highly maladaptive behavior? Automaintenance
appears to depend in part on basic associative learning (fueled
by light-food pairings) and instinctive drift, both processes
promoting pecking to the light, as well as to conditioned
reinforcement associated with the offset of the key light. As in
autoshaping, pecking the key does have an effect: it turns off
the key light (although unlike autoshaping, food is not pre-
sented). On trials when the organism does not peck for 6 s, the
key light is turned off and food is simultaneously provided.
Thus, the food is always preceded by the termination of the
light. Steven Hursh and his colleagues at the University of
Associative Learning 209
California, San Diego showed that light termination developed
into a conditioned reinforcer after repeated pairings of light
termination and food. Together these sources of strength (clas-
sical conditioning, instinctive drift, and conditioned reinforce-
ment) appear to account for the puzzle of automaintenance.
Necessary and Sufficient Conditions for Learning
An important study by Elkan Gamzu and Barry Schwartz at the
University of Pennsylvania addresses the necessary and suffi-
cient conditions for the generation of autoshaped responses.
This study also makes a central point about the necessary and
sufficient conditions for the occurrence of conditioning (in this
case autoshaping). In one condition of their study, pigeons
were exposed to two alternating schedules of reinforcement,
each associated with a different stimulus (technically, a multi-
ple schedule). On one schedule, food was never provided. On
the other alternating schedule, food occurred on the average
every 33 s and occurred whether or not the pigeon had
responded. In other words, these food presentations were
response-independent. This schedule is technically a variable-
time 33-s schedule (VT 33 s). From our discussion of autoshap-
ing, we would expect that the pigeon might begin responding
to the stimulus associated with the VT 33-s schedule even
though such responding was superfluous and that no respond-
ing would occur in the presence of the EXT stimulus. This is
indeed what was found. In another condition, however, the
EXT component was replaced with another VT 33-s schedule.
Thus, in this condition, two VT 33-s schedules alternated,
identical except for the different key colors illuminated during
their operation. Would autoshaped responding occur to these
two stimuli? The answer was 'no,' suggesting that a differential
correlation with reinforcement is necessary for the generation
of autoshaped responding. Rephrasing in simple terms, when
both stimuli were equally good, neither elicited autoshaping.
Recall that in the basic autoshaping procedure, the stimulus
follows a period of darkness in which no reinforcement is
provided; thus the light onset is correlated with reinforcement
and elicits autoshaped responding.
Finally, consider a condition in which the VT 33-s schedule
alternated with a VT 1 00-s schedule (in which food is provided,
on the average, once every 100 s). We can guess (correctly as it
turns out) that the VT 33-s schedule generates autoshaped
responding in this condition since it is differentially correlated
with (a higher rate of) food presentations. The interesting ques-
tion is whether or not the VT 100-s schedule generates auto-
shaped responding. From what we have suggested above, the VT
100-s schedule should behave more like EXT than a positive
stimulus since it is not differentially correlated with a higher rate
of food presentation. The results supported this view.
Based on these results, we may reach the tentative conclu-
sion that in order to promote associative learning a stimulus
must be differentially correlated (or be 'predictive of') rein-
forcement. The next two sets of studies confirm this conclusion
in areas other than autoshaping.
Conditioned Suppression
Conditioned suppression is a quantitative measure of anxiety
that has been used extensively in nonhuman laboratories, and,
in modified form, with humans as well. In this procedure, a
hungry organism that is emitting a standard operant response
such as a lever press, reinforced by food presentations on an
intermittent basis, is presented at specific intervals with an
originally neutral stimulus (such as a light or tone) and imme-
diately afterward is administered electric shock independent of
its behavior. The effect of pairing a neutral stimulus with shock
and superimposing the two on a schedule of food reinforce-
ment is to lower the rate of responding in the presence of the
stimulus (the light or tone is a CS; the electric shock the US)
when compared with the rate of responding in the absence of
the stimulus (i.e., the rate when the stimulus is not present).
This suppression of response is called conditioned suppres-
sion, or a conditioned emotional response. This paradigm
has proven effective in the study of anxiety and conditioned
suppression with many species of animals including rats,
pigeons, monkeys, and humans in order to assess the condi-
tions under which a stimulus paired with an aversive event will
suppress or dismpt ongoing behavior.
An additional reason for introducing conditioned suppres-
sion is that it was utilized in another critical study evaluating
the necessary and sufficient conditions for associative learning.
Traditional accounts of conditioning had generally assumed
that simple contiguity is the central factor in the reinforcement
process. Today we are aware of the central role of the signaling
or informative (or predictive) value of the CS. We have already
encountered this when discussing research from the autoshap-
ing domain. Mere temporal contiguity between the CS and US
is not enough for conditioning to occur. An obvious example is
the fact that simultaneous conditioning is ineffective, though
in that case it might be argued that the salient US overshadows
the simultaneously presented CS. As the autoshaping results
suggest, contiguity is effective for conditioning only if a corre-
lation exists between the CS and US. A correlation exists if the
CS reliably predicts occurrence of the US, for example, when
the occurrence of the US is contingent upon (depends on) the
prior occurrence of the CS. Robert Rescorla used the conditioned
suppression paradigm to demonstrate this. His rats, responding
on an intermittent schedule for food, were also exposed to
superimposed CS and US pairings. Unlike the usual conditioned
suppression procedure, the US also occurred at times in the
absence of a prior CS, and the US sometimes failed to occur
following a CS. Thus, Rescorla could independently manipulate
the probability of the US following a CS as well as the probabil-
ity of the US in the absence of the CS. Across conditions, he
could observe how the degree of conditioning was affected by
manipulating the probability of the US in the absence of the
CS while holding constant the probability of the US following
the CS. No conditioning was found when the probability
of the US was equal following the presence or absence of
the CS, that is, when the US was equally likely to occur with or
without the CS. The CS no longer predicted (was not correlated
with) the US and no longer gained associative strength (here,
no longer suppressed ongoing response for food). The conclu-
sion: pairing without correlation is insufficient for conditioning.
Observing
Research on observing explores the question of what types of
information reinforces (strengthens) behavior. The procedures
210 Associative Learning
and results are presented at length when discussing conditioned
reinforcement (operant conditioning). Thus, here we need only
to summarize the rationale and the implications of the results
for the issue of the necessary and sufficient conditions for asso-
ciative learning. Observing responses are those which produce
stimuli correlated with schedules of reinforcement, but that
have no effect on the occurrence of reinforcement. For example,
two equally probable schedules of reinforcement differing
only in frequency of reinforcement - say, variable time (VT)
and EXT - may alternate unpredictably. Effective observing
responses would produce stimuli identifying the schedule in
effect. Will organisms respond to produce these stimuli even
though they have no influence on the course of reinforcement?
Many studies with a wide array of species (including humans)
have answered this question in the affirmative. What is the
mechanism at play here since it is not an enhanced rate of
primary reinforcement? Does a stimulus maintain observing
responses because it is correlated with the occurrence of pri-
mary reinforcement (the conditioned-reinforcement hypothe-
sis)? For example, according to the delay-reduction theory, a
stimulus will be a conditioned reinforcer when its onset is
correlated with a reduction in time to primary reinforcement.
This prediction is also consistent with other major theories
of conditioned reinforcement. Alternatively, does a stimulus
reinforce observing responses because it predicts or informs
about the availability of reinforcement (the information or
uncertainty-reduction hypothesis)? The critical test for distin-
guishing between these views is whether or not an S-, or bad
news, is reinforcing, since bad news is not a conditioned rein-
forcer but is highly informative. The overwhelming preponder-
ance of evidence shows that bad news does not reinforce
(maintain) an observing response. Only good news reinforces
observing. Additional research has refined this position: a stim-
ulus maintains observing (shows associative strength) only
when it is differentially correlated with positive reinforcement
(good news). Thus, an FI 5-min schedule will maintain observ-
ing when it alternates with EXT but will not maintain observing
when it alternates with a richer schedule (say FI 1-min).
following conclusion: a stimulus typically acquires associative
strength if and only if it is differentially correlated with rein-
forcement. The role of contextual variables is central: depending
on the context, the same stimulus-reinforcer relation (whether
in classical or operant conditioning) may be sufficient to gen-
erate powerful conditioning (or associative strength) or none
at all.
See also: Associative Learning; Operant Conditioning.
Further Reading
Balsam PD, Drew MR, and Yang C (2002) Timing at the start of associative learning.
Learning and Motivation 33: 141-155.
Fantino E, Preston RA, and Dunn R (1993) Delay reduction: Current status. Journal of
the Experimental Analysis of Behavior 60: 159-169.
Fantino E and Silberberg A (2010) Revisiting the role of bad news in maintaining
human observing behavior. Journal of the Experimental Analysis of Behavior
93(2): 157-170.
Gottfried JA, O'Doherty J, and Dolan RJ (2002) Appetitive and aversive olfactory
learning in humans studied using event-related functional magnetic resonance
imaging. Journal of Neuroscience 22: 10829-10837.
Mazur JE (2006) Learning and Behavior. Upper Saddle River, NJ: Pearson Prentice Hall.
Pavlov IP (1927) Conditioned Reflexes. London: Oxford.
Pierce WD and Cheney CD (2008) Behavior Analysis and Learning, 4th edn. New York:
Psychology Press.
Rescorla RA (1968) Probability of shock in the presence and absence of CS in fear
conditioning. Journal of Comparative and Physiological Psychology 66: 1-5.
Rescorla RA and Wagner AR (1972) A theory of Pavlovian conditioning: Variations in
the effectiveness of reinforcement and nonreinforcement. In: Black AH and
Prokasy WF (eds.) Classical Conditioning II: Current Research and Theory,
pp. 64-99. New York: Appleton-Century-Crofts.
Revusky SH and Bedarf EW (1967) Association of illness with prior ingestion of novel
foods. Science 155: 219-220.
Sokolowska M, Siegel S, and Kim JA (2002) Intraadministration associations:
Conditional hyperalgesia elicited by morphine onset cues. Journal of Experimental
Psychology: Animal Behavior Processes 28: 309-320.
Stiers M and Silberberg A (1974) Lever-contact responses in rats: Automaintenance
with and without a negative response-reinforcer contingency. Journal of the
Experimental Analysis of Behavior 22: 497-506.
Summary
The results from all three of these areas of research (autoshap-
ing, conditioned suppression, and observing) support the
Attention
C Spence, University of Oxford, Oxford, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Alerting A short-lasting increase in neural activation that is
often hard to distinguish from attention. However, while
alerting, just like attention, is typically characterized by a
speeding-up of participants' behavioral responses, it often
leads to more error-prone performance (thus contrasting
with attention).
Attention Refers to those (neural) mechanisms giving rise
to a temporary enhancement (or prioritization) of the
processing of certain stimuli relative to others.
Attentional bottleneck The hypothetical point in
information processing at which the selection of certain
information, over other information, takes place. For many
years, researchers have argued over whether the bottleneck
(i.e., selection) takes place 'early' or 'late' in human
information processing.
Divided attention The ability to process multiple streams of
information, or to perform multiple tasks, at the same time.
Typically, those conditions that make dividing one's
attention easier will make it harder to focus one's attention,
and vice versa.
Focused attention The ability to process a particular subset
of the available incoming sensory information while
simultaneously ignoring other distracting (or currently
task-irrelevant) information.
Perceptual/attentional load According to the theory of
perceptual load, our attentional resources are always fully
engaged with the processing of any incoming sensory
information. Thus, when a person's primary task is not
overly demanding, any spare attentional resources will be
available for the processing of other stimuli.
Prior entry Refers to Titchener's (1907) claim that
attended stimuli are perceived earlier than unattended
stimuli.
Spatial cuing The name given to a popular behavioral
paradigm introduced by Mike Posner in the 1970s to study
spatial attention. In a typical study, a cue stimulus is
presented from one or other side of central fixation,
followed a short time later by a target stimulus on either the
same or opposite side. Participants normally respond more
rapidly and/or accurately to targets presented on the cued
side. The claim is that this facilitation of performance reflects
the beneficial effects of spatial attention on information
processing.
Sustained attention Otherwise known as 'vigilance' - the
ability to maintain one's attentional focus on a given task or
source of information, often over prolonged periods of time.
Vigilance tasks are often repetitive and/or boring, and
perhaps for this reason are less popular than they once were.
Working memory The cognitive system thought to contain
storage buffers and a central executive that actively
maintains goal-relevant information in the service of
complex cognition. The central executive, the least
well-understood part of the working memory system,
is typically conceptualized as controlling attention and
the flow of information to/from the buffers.
Everyone knows what attention is. It is the taking possession by the
mind, in dear and vivid form, of one out of what seem several
simultaneously possible objects or trains of thought.
(William James, The Principles of Psychology, vol. 1, pp. 403-404)
Introduction
Research on the topic of selective attention really developed
in the years after the end of the Second World War, when
scientists began to attempt to understand why it was that fighter
pilots sometimes missed perfectly audible signals that had been
presented over their headsets. Perhaps unsurprisingly, such
applied problems led to the development of the selective (or
dichotic) listening task, in which psychologists attempted to
simulate the kinds of multispeaker environments faced by the
pilots. In a typical study, participants would be required to
shadow the speech stream presented to one ear, and then they
would be tested for their recall of the information presented to
the other 'unattended' ear (or channel). The majority of research
efforts in the intervening years have, however, tended to focus on
understanding the mechanism underlying selective attention in
the visual modality. That said, the role of selective attention
in the other senses as well, such as touch, is now being studied.
Everyone agrees that the human brain has only a limited
capacity to process incoming sensory information, and hence
selection must occur in order to prevent overload. The key
question motivating much of the early research on attention
was where exactly selection (or the bottleneck) occurred
and just how much processing there was of the information
that was not currently task relevant. Some researchers, such as
Donald Broadbent, argued that the bottleneck in human
information processing occurred early in the process, whereas
others, such as Deutsch and Deutsch, argued that all incoming
stimuli were processed to a fairly high (possibly semantic) level,
just short of awareness. If one were to try and summarize this
large body of research, one would have to say that different
researchers using different experimental paradigms were able
to provide evidence consistent with both early and late selection.
Part of the problem in this area was that different researchers also
used different measures of awareness (everything from verbal
report through to galvanic skin responses and implicit behavioral
measures). By the early 1990s, the field had stagnated. However,
211
212 Attention
two changes occurred that revitalized research in this area. First,
the development of neuroimaging meant that for the very first
time researchers were able to peer inside the heads of their
experimental participants in order to see where exactly atten-
tional selection was occurring in the human brain. Second,
Lavie and Tsai put forward the perceptual load theory, which
provided a means of explaining the apparently inconsistent pat-
tern of selection sometimes occurring early and at other times
appearing to occur much later in information processing.
The Theory of Perceptual Load
According to the theory of perceptual load, our attentional
resources are always fully engaged with the processing of any
incoming sensory information (i.e., we cannot store them up
for use at a later point in time). Thus, under those conditions
in which a person's primary task is not overly demanding, any
spare attentional resources they have will be available for the
processing of other stimuli (e.g., the irrelevant stream of speech
presented in the 'unattended' ear in a dichotic listening study).
Lavie has argued that under such 'low-load' conditions, late
selection can be observed. However, under those conditions
in which the load of a participant's primary task is increased
(such as when the complexity, or rate of presentation, of the
to-be-shadowed message increases, or when the number of
stimuli presented simultaneously in a visual display goes up),
a participant will need to devote more resources to processing
it. Hence, there will be fewer resources left for the processing of
any other incoming information. Lavie argues that under such
'high-load' conditions, attentional selection will, on the con-
trary, likely occur relatively early in information processing.
Perceptual load theory has provided a popular explana-
tion for why attentional selection sometimes occurs early
in information processing, whereas at other times (and/or in
other studies) it occurs much later. As such, this intuitive
theory appears to offer a means of resolving the long-standing
(and, for many years, seemingly intractable) debate amongst
researchers over early and late attentional selection. In the
years since Lavie and Tsai first proposed their theory, a large
and ever-growing number of studies have been published in
support of the theory's main tenets. Empirical support has
come from a number of sources, including both behavioral
and neuroimaging research. What is more, relatively few results
have been reported that cannot be accounted for by the theory.
That said, one limitation of the theory is that no objective
measure of 'load' has yet been provided.
Thus far, the majority of studies of perceptual load have
tended to restrict themselves to investigating those conditions
in which participants are presented with (and hence have to
select between) the information presented in just one sensory
modality. Most researchers have focused on conditions of unim-
odal visual selection, though studies demonstrating the applica-
bility of the theory to those conditions requiring unimodal
auditory selection have now started to appear in the literature.
As yet, there has been relatively little research investigating the
validity of perceptual load theory in accounting for participants'
behavior under conditions requiring unimodal tactile selection.
However, that said, there is no obvious reason to believe that
selective attention should operate any differently within the
tactile modality than it does within either of the other spatial
modalities (i.e., vision or audition).
Having demonstrated attentional selection under a variety
of different unimodal conditions, the next question to be
addressed by researchers was whether the attentional resources
that lie at the heart of perceptual load theory are unitary, or
whether there are instead separate resources for the processing
of stimuli in each sensory modality (as originally suggested by
Christopher Wickens' 'multiple resource theory,' back in the
early 1980s). While the results of the early crossmodal percep-
tual load studies came to the conclusion that resources were
indeed modality-specific, the story has, in recent years, become
much less clear. While some researchers have demonstrated that
varying the perceptual load in one sensory modality (e.g.,
vision) affects the level of processing of stimuli presented simul-
taneously in another sensory modality (e.g., audition), other
researchers have failed to observe any such crossmodal effect.
Working Memory and Selective Attention
What has emerged very clearly though in recent years is the
close link that exists between selective attention and working
memory (WM). It appears that whenever a participant's WM is
compromised, his or her ability to selectively attend to one
source (or stream) of sensory information and to ignore
another (task-irrelevant) source (or stream) of information
will be impaired. In laboratory studies, WM is typically loaded
by giving a participant a list of stimuli to remember (such as a
random list of digits) while he or she simultaneously performs
a selective attention task. Many studies have now demon-
strated that loading WM effectively impairs a participant's abil-
ity to selectively attend to a particular source of information.
Intriguingly, it would appear that individual differences in WM
capacity can also help to explain the existence of striking dif-
ferences in people's selective attention abilities. So, for exam-
ple, it has been known for more than 50 years now that only
around one-third of participants will report hearing their own
name if it happens to be presented in the 'unattended' ear in a
dichotic listening experiment. Counterintuitively, it turns out
that those individuals who have a lower WM capacity are
significantly more likely to hear their name. One suggestion
here is that these individuals are simply more likely to forget
which ear it is that they are supposed to be shadowing, and
hence will be more likely to inappropriately shift their atten-
tion to the other ear (the one containing their name).
Neuroimaging studies have now started to highlight the
neural underpinnings of this interaction between selective
attention and WM. For example, in one recent study, loading
participants' WM (by giving them a digit string to remember)
resulted in the suppression of neural activity in the right tem-
poroparietal junction, a brain region that constitutes a core
structure in the ventrofrontoparietal attention network.
Inattentional Blindness, Change Blindness, and
Change Detection
One intriguing visual analog of auditory shadowing experi-
ments has come from a spate of studies over the last decade
Attention 213
or so that have revealed the existence of severe limitations in
our awareness of multiple simultaneously presented visual
stimuli. In what is perhaps the most famous example (of
what has come to be known as inattentional blindness),
researchers found that participants would often fail to notice
a person dressed up in a gorilla suit thumping his chest in the
middle of a monitor, if their attention happened to be focused
on a group of people dressed in white passing a basketball.
It turns out that the more demanding the participant's task -
such as, for example, having to keep separate running totals
of the number of times that the white team throw the ball and
the number of occasions on which they bounce the ball - the
more likely it is that participants will miss the gorilla, consis-
tent with the predictions of perceptual load theory. Surpris-
ingly, the results of many such studies now demonstrate that
we frequently miss (i.e., fail to notice) even highly salient
visual stimuli, if our attention happens to be directed else-
where (i.e., to another object or stimulus). Such findings
show that our subjectively strong belief that we see everything
before our eyes is nothing more than an illusion.
A related phenomenon, known as change blindness, occurs
when a scene and the same scene with a major (i.e., otherwise
clearly noticeable) change in it are presented for brief periods
of time separated by a blank interval. It turns out that until
attention alights on the changed element it will not be report-
able. The argument here is that the blank display inserted
between the two scenes acts as a global visual transient that
obscures the localized spatial transients in the scene that would
normally capture our visual attention (as when an item sud-
denly appears, disappears, or moves). Even when these transi-
ents are visible (i.e., when one scene directly transitions to the
other), people may not perceive the scene change if muds-
plashes (i.e., localized changes in a visual display, such as a
bunch of gray circles projected at random points on the screen)
happen to be presented at the same time.
Following the original research conducted in the visual
modality, it was demonstrated by subsequent studies that a
similar phenomenon, known as 'change deafness,' also affects
auditory perception. For example, in one study, participants
were presented with an auditory scene containing between four
and eight auditory stimuli (including sounds such as those of a
trumpet, a piano, a female voice, and a clucking hen) for 5 s.
Each sound occupied a different position in the virtual auditory
display presented over headphones. Next, white noise was pre-
sented for 500 ms, followed by a second auditory display for
a further 5 s. The two auditory displays were identical except
that, on half of the trials, one of the auditory stimuli was
removed from the second display (sometimes to be replaced
by another item). On each trial, the participants simply had to
decide whether one of the auditory stimuli had been removed
(or replaced) in the second display. Performance dropped from
near perfect when the first display contained four auditory
objects, down to chance-level performance when it contained
eight. By contrast, in the control condition, in which participants
were informed which of the stimuli might change, performance
was virtually perfect regardless of the set size, thus demonstrat-
ing the attentional nature of the effect (and ruling out an
account of the poor performance observed in terms of masking) .
Studies of change numbness have revealed even more
severe limitations in the perception of tactile stimuli presented
over the body surface or fingertips. So, for example, in a typical
study of the tactile analog of change blindness, participants are
presented with patterns consisting of vibrotactile stimuli being
delivered to two or three locations across the body surface. The
mask typically consists of the brief stimulation of a number
(e.g., seven) of widely distributed vibrators. A second pattern
of two or three vibrotactile stimuli is then presented, and
participants have to decide whether there had been a change
(which occurs on half of the trials), either in the total number
of stimuli presented or in the location of the stimuli on the
body surface. The results of several such studies have shown
surprisingly poor change detection performance when the
two vibrotactile patterns are separated by a mask. By contrast,
the performance deficit was far less severe when the interval
between the two patterns was empty (i.e., when no mask was
presented). Interestingly, researchers have even shown that the
presentation of visual mudsplashes can crossmodally give rise
to tactile change blindness as well.
Taken together, studies of inattentional blindness, change
blindness, change deafness, and change numbness support
the conclusion that people may actually be unaware of much
of the sensory information present in complex multisensory
scenes. It would seem that we are only aware of the stimuli
that we happen to be attending to voluntarily, or which capture
our attention automatically. The results of the many studies
published in this area therefore highlight the critical role that
attention plays in awareness.
Spatial Attention
Researchers studying attention distinguish between two types
of selection - exogenous and endogenous: Exogenous (or invol-
untary) attentional orienting occurs whenever attention is
reflexively shifted to the location of a sudden and unexpected
peripheral event, such as when a person calls your name at a
cocktail party, or when a fly suddenly, and unexpectedly, lands
on your arm. By contrast, endogenous attention is typically
involved in the voluntary orienting of a person's attention to
a particular event or spatial location, such as when you choose
to attend to a particular person at the noisy party, or when you
decide to concentrate on the feel of the MP3 player that you are
holding in your hand. Orthogonal to this distinction between
endogenous and exogenous (or voluntary and automatic)
spatial attention, is the distinction between overt and covert
attentional orienting: Overt orienting refers to any shift of the
sensory receptors (as in eye, head, or hand movements) that
helps to bring a stimulus onto the most sensitive parts of
the receptor sheet (think, the fovea or the fingertips). Covert
orienting, which is of most interest to cognitive psychologists
studying the topic of selective attention, refers to an internal
shift of attention, as when what we attend to visually changes
in the absence of any overt movement of the eyes. Many
researchers believe that there is a close link between these two
forms of spatial orienting, as, for example, captured by the
premotor theory of attention. According to the premotor theory,
the same frontal-parietal brain circuits control both action -
that is, motor behavior toward specific spatial locations - and
attention, when those circuits happen to be activated less
strongly. Although the premotor theory of attention has
214 Attention
perhaps not received the credit it is due, it is certainly true that
the superior colliculus, a subcortical structure implicated in the
control of overt eye and head movements, plays an important
role in the covert shifting of spatial attention.
Exogenous Spatial Attention
While it is certainly true that attention can be directed, either
exogenously or endogenously, to a sensory modality, or to any
one of a range of other stimulus features (such as color), by far
the most popular topic in attention research in recent decades
has been the spatial distribution (or orienting) of attention.
The majority of studies in this area have adopted the spatial
cuing paradigm first popularized by Mike Posner back in the
1970s. In a typical exogenous cuing study, a cue stimulus is
presented in one modality shortly before a target stimulus,
presented on either the same or opposite side (and in either
the same or different sensory modality) . Participants normally
have to make either a speeded detection or discrimination
response to the target (on occasion, they may be asked to try
and detect a near-threshold masked target). The key feature of
such studies is that the cue is nonpredictive with regards to the
likely location of the subsequently presented target. Often, par-
ticipants are explicitly instructed to try and ignore the cue as
much as possible. However, the results of many such studies
conducted over the last 30 years or so have demonstrated that
often people simply cannot ignore such cues. Instead, they
respond more rapidly, and sometimes more accurately, to targets
presented at the cued location as compared to the uncued one.
These short-lasting exogenous spatial cuing effects typically last
for only a few 100 ms after the presentation of the cue. Often,
especially in detection tasks, facilitation is replaced by a longer
lasting inhibitory effect, known as 'inhibition of return' (IOR).
The idea here is that an inhibitory tag is attached to any location
where attention has been directed (either overtly or covertly),
but where no target has been found. Such inhibitory tagging is
thought to help prevent us from perseverating on uninformative,
or irrelevant, locations (e.g., as in visual search tasks; see below).
The presentation of spatially nonpredictive auditory, visual,
and tactile cues has been shown to elicit exogenous spatial
cuing effects that affect the subsequent processing of auditory,
visual, and tactile targets presented from more or less the same
location. Researchers using signal detection analysis have
demonstrated a genuine enhancement of perceptual sensitivity
for stimuli presented at the cued (or attended) location. What
is more, the same stimuli reach consciousness sooner when
attended than when attention happens to be directed else-
where, a phenomenon known as 'prior entry.'
Whether peripheral cues always capture our attention auto-
matically, or whether we have to be 'set' appropriately in order
for them to elicit a shift of attention has been a topic of heated
debate in the attention literature in recent years. The evidence
generated by this debate has led many researchers to suggest that
even the putative involuntary spatial orienting of our attention
may, in fact, be subject to at least some degree of top-down
(or strategic) control - that is, it depends on our having the right
'attentional set.' In fact, the interplay between the bottom-up
and top-down control of our attention is one that has pervaded
the literature on attention over the last three decades or so.
Endogenous Spatial Attention
Turning now to the deployment of endogenous spatial
attention, it has been demonstrated that if people deliberately
direct their attention to a particular spatial location, perfor-
mance will normally be enhanced for targets subsequently
presented at that location, as compared to targets presented
elsewhere. Just as for the case of exogenous orienting, endoge-
nous (or voluntary) attention facilitates the processing of audi-
tory, visual, and tactile stimuli presented at the attended
location. Several different methods have been used to direct a
participant's spatial attention endogenously to one side or the
other. In early studies, peripheral spatial cues were often made
informative with regard to the likely location of the target,
as when the target appears on the cued side on 75% of the
trials, and on the uncued side on the remaining 25% of trials.
However, under such conditions, both exogenous and endog-
enous spatial attention are likely directed to the cued side. In
many subsequent studies, therefore, a central cue (such as
an arrow pointing to the left or right) has been used to direct
attention to one or other side on a trial-by-trial basis instead.
That said, it has been argued that even central arrow cues
may direct attention exogenously (in a manner similar to
centrally presented faces where gaze is deviated to one or the
other side). Hence, currently, one of the best means of direct-
ing a participant's attention in a purely endogenous manner
is by means of blocked cuing, where, for example, the majority
of targets appear on the left, say, and only a small number of
unexpected targets appear on the other side. The latter
approach ensures that any performance facilitation that hap-
pens to be observed can unequivocally be attributed to the
consequences of endogenous attentional orienting.
Endogenous Versus Exogenous Spatial Orienting:
One Mechanism or Two?
There has been a great deal of debate over the years as to
whether exogenous and endogenous orienting simply reflect
different means of engaging the same attentional resources, or
whether instead they have fundamentally different properties/
consequences for human perception/performance. Recently,
Prinzmetal and his colleagues have reported on a very large
number of studies demonstrating that while endogenous spa-
tial attentional orienting typically enhances perceptual sensi-
tivity, exogenous orienting primarily influences which spatial
location information is sampled from first. As a consequence,
exogenous orienting primarily affects the speed (but not neces-
sarily the sensitivity) or a participant's performance.
Spatial attention has been conceptualized as taking a variety
of forms: Early on, researchers typically envisioned spatial
attention as a spotlight (the assumption being that it had a
fixed size). Subsequently, the analogy of a zoom lens was
proposed as an alternative - the idea being that the size of
the attentional spotlight was variable, and that there was a
trade-off between the size of the spotlight and the 'density' of
attentional resources directed toward a particular region of
space. When researchers discovered that attention could be
directed to objects, as well as spatial locations, a debate started
concerning the relation (or interplay) between space- and
Attention 215
object-based attention. While the majority of this research has,
once again, focused on the visual modality, researchers have,
more recently, highlighted the existence of object-based atten-
tion in the auditory modality as well.
Crossmodal Spatial Attention
The question of how attention might be configured crossmod-
ally has been a topic of intense scrutiny over the last decade or
so. It now appears that there are extensive crossmodal links in
spatial attention, such that visually attending to a particular
location will lead to a shift of auditory and tactile attention to
the same spatial location (or at the very least in the same
direction). Similarly, the exogenous capture of spatial attention
by a sudden auditory, visual, or tactile event leads to a shift of
attention that will facilitate participants' ability to respond
to any target subsequently presented at the cued location
(regardless of the modality in which the target happens to be
presented). What is more, these crossmodal links in spatial
attention update following posture change, such as when the
eyes move with respect to the head, or when the arms are
moved (often crossed) with respect to the body. This means
that attention is always directed in such a way that the proces-
sing of all sensory stimuli originating from a given environ-
mental location will be facilitated, regardless of how our
various sensors (e.g., the eyes, ears, and skin) happen to be
configured spatially.
The Attentive Brain
Neuropsychological studies of patients suffering from spatial
neglect/extinction following brain damage have been par-
ticularly informative in terms of furthering our understanding
the neural substrates of attention. Neglect and extinction typi-
cally occur following right hemisphere damage to the parietal
cortex. Neglect patients tend to be unaware of stimuli pre-
sented in the contralesional field (i.e., the left side of space
following right parietal damage). Extinction patients can per-
ceive contralesional stimuli when presented in isolation, but
the simultaneous occurrence of an ipsilesional stimulus will
typically extinguish the contralesional stimulus from the
patient's awareness. Research on such patients has, for exam-
ple, further helped to highlight the importance of the distinc-
tion between space- and object-based attention.
The last few years have seen an explosion of cognitive
neuroscience research investigating the neural mechanisms
underlying attentional selection. Researchers have used a vari-
ety of brain imaging techniques, from PET and fMRI through to
electrophysiology and single-cell neurophysiological record-
ings, in order to better understand how, and where, attention
is operationalized in the brain. While a detailed discussion of
this literature is beyond the scope of this article, it is worth
noting that the spatial distribution of attention in neurologically
normal individuals appears to be controlled by a large-scale
frontoparietal neural network. Perhaps unsurprisingly, the
exogenous and endogenous orienting of attention appears
to rely on somewhat different neural circuitry. Consistent
with the behavioral data, the neuroimaging and transcranial
magnetic stimulation (TMS: a means of temporarily lesioning
a normal participant's brain) data support the view that
exogenous spatial attention (and IOR) may be controlled by
a supramodal orienting system, while endogenous spatial
attention is controlled by a system that approximates to the
separable-but-linked account originally proposed by Spence
and Driver in 1996.
Feature Integration Theory
Around 1980, Anne Treisman published what has subse-
quently turned out to be a very influential model of selec-
tive attention, known as feature integration theory (FIT). The
model helps to account for why visual search through cluttered
displays is sometimes difficult (as when we search for our
misplaced car keys in our untidy office), while at other times
it appears easy (as when a single hand is raised in a crowded
auditorium). According to Treisman, attention is required to
bind the different features of an object (such as its color, shape,
orientation, and motion) into a coherent whole. By contrast,
unique features (such as a pink circle in a sea of green circles)
'pop-out,' thus automatically capturing our attention. Across a
series of now classic experiments, Treisman demonstrated that
the search for a visual target defined by a unique feature was
equally fast regardless of the number of stimuli that happened
to be presented in the display. By contrast, search for a con-
junction of features (such as for a red square presented in
amongst green squares and red and green circles) tended to
be slow and effortful (or attention-demanding). Hence, search
latencies would increase linearly as a function of the set size.
One prediction of FIT is that when attention is overloaded,
either because attention has been spread over too wide an area,
or because the stimuli are presented for too brief a time (and
are, anyway, quite possibly masked), then illusory conjunc-
tions may occur (whereby, e.g., the color of one object may
be bound together with the shape of another). Whether such
errors reflect perceptual mis-combinations rather than just
errors in memory recall remains a contentious issue. Research-
ers have demonstrated that brain-damage, or repetitive TMS
(rTMS) applied to right parietal sites, can impair a person's
ability to conjoin features correctly.
Over the intervening years, FIT has generated a huge
amount of interest from the psychological research commu-
nity. Perhaps unsurprisingly, then, a number of findings that
are apparently inconsistent with the theory, at least as origi-
nally stated, have now been published. In particular, the evi-
dence that is currently available shows that attention is
sometimes required to detect (or at the very least, to respond
to) targets defined by a unique feature. What is more, many
targets that are defined by a conjunction of features appear
to pop-out, when, according to Treisman, they should not.
However, that said, the general approach still lives through
various other models - that can be seen as an extension of
Treisman's FIT - such as, for example, Wolfe's 'guided search'
model. It is, though, important to bear in mind that the atten-
tional spotlight thought to be directed to the location of an
unexpected peripheral stimulus in the Posnerian cuing task
appears to have somewhat different properties than the feature-
integrating spotlight made famous by Treisman's FIT. While FIT
216 Attention
was originally formulated to account solely for participants'
visual search performance, the last few years have seen research-
ers extending the approach to account for the binding of
auditory, tactile, and even multisensory object features.
Temporal Attention
While the majority of studies related to attention have tended
to focus on the spatial aspects of selection, a growing body of
research has, in recent years, started to look at the temporal
limits on information processing instead, as when participants
are offered rapidly presented sequences of visual, auditory,
or tactile stimuli. So, for example, using such an approach,
researchers have highlighted the existence of a temporal atten-
tional deficit known as the 'attentional blink' (AB). This robust
performance impairment is typically observed under those
conditions in which two targets, each requiring a behavioral
response (often unspeeded), are presented in close succession
within a rapidly presented sequence of distractor stimuli. The
rate of stimulus presentation in such studies is often very high
(~8-10 items per second). If the second target is presented a
couple of stream positions after the first target, a participant's
ability to identify it will often be severely impaired. (Curiously,
if the second target is presented directly after the first target
then discrimination performance is often preserved - a phe-
nomenon known as Lag-1 sparing.) As the interval between the
first and second target increases, performance typically recovers
back to baseline level. It is almost as if there is an attentional
gate that is closed after the arrival of the first target in order to
prevent it from being overwritten by subsequent items in the
stream. Interestingly, certain patients suffering from clinical
extinction have been shown to exhibit a significantly larger
AB than neurologically normal individuals.
Although originally discovered in the visual modality, the
AB has now been demonstrated to occur in the auditory and
tactile modalities as well. There is even evidence to suggest
the existence of a crossmodal visual-tactile AB. That said, the
evidence concerning the existence of an audiovisual AB is,
despite the continued efforts of researchers, much more mixed.
Researchers have also identified a number of other temporal
limitations to our ability to process rapidly presented stimuli.
For example, Hal Pashler has conducted very many experi-
ments documenting what appears to be a response selection
bottleneck that limits our ability to make speeded responses to
different stimuli when two targets are presented in rapid suc-
cession. Many neuroscientists have spent the last decade look-
ing for the neural correlates of this psychological refractory
period (PRP) bottleneck in the human brain.
Over the last decade or so, researchers have also started to
study the temporal orienting of attention to a specific 'point in
time.' Many of the studies in this area have simply adapted the
spatial cuing paradigm (originally developed to study spatial
orienting) in order to study the consequences of temporal expec-
tancies on the efficiency of human information processing. So,
for example, temporal expectations may be generated in the
mind of an observer by making it highly likely that the target
will appear a short time after a cue stimulus has been presented
(and much less likely that it appears at a later point in time).
Target performance in such trials (or blocks) is then compared
to that seen when the probabilities are reversed (such that the
target is now much more likely to appear in the later interval
than in the earlier one). The results of many such studies now
show that stimuli presented at an expected point in time are
processed preferentially as compared to when the same stimuli
are presented unexpectedly at another point in time.
Attention: The Future
As will have become clear from this brief review, the last few
years have seen tremendous advances in our understanding of
attention at both the behavioral and neural levels. The field has
certainly moved a long way forward since the late Stuart
Sutherland uttered the following put-down in a commentary
in the science magazine Nature in 1998: "after many thousands
of experiments, we know only marginally more about attention
than about the interior of a black hole. " The old box-and-arrow
diagrams, once popular in cognitive psychology textbooks
(approximately the 1960s), are now increasingly being
replaced by neurophysiologically plausible models of the man-
ner in which, and the location where, attention operates in the
human (and animal) brain. These models are also increasingly
able to explain the deficits in attention (such as neglect and
extinction) sometimes observed in patients following brain-
damage. One of the other trends in attention research that has
been noticeable in recent years is the increasing shift from the
traditional approach of considering selection among sensory
inputs from just a single modality (most often vision) to a
rapid growth of interest in the nature of the crossmodal con-
straints on human information processing.
In the years to come, it is hoped that our growing under-
standing of attention, fine-tuned by sophisticated behavioral
experiments paired with the latest cognitive neuroscience
research, will allow researchers to better understand the real-
world limitations on human information processing: every-
thing from the problems associated with trying to divide
one's attention between eye and ear talking on the mobile
phone while driving, through to the difficulties associated
with trying to detect infrequent targets (be it a weapon in the
luggage at the baggage-scanning machine or the precancerous
growths on a radiological image) while engaged in a repetitive
visual search task.
See also: Spatial Perception; Touch; Visual Neglect; Visual
Perception.
Further Reading
Corbetta M and Shulman GL (2002) Control of goal-directed and stimulus-driven
attention in the brain. Nature Reviews Neuroscience 3: 201-215.
Driver J (2001) A selective review of selective attention research from the past century.
British Journal of Psychology 92: 53-78.
Ho C and Spence C (2008) The Multisensory Driver: implications for Ergonomic Car
Interface Design. Aldershot: Ashgate.
Klein R (2000) Inhibition of return. Trends in Cognitive Sciences 4: 138-147.
Klein RM and Shore Dl (2000) Relationships among modes of visual orienting.
In: Monsell S and Driver J (eds.) Control of Cognitive Processes: Attention and
Performance XVIII, pp. 195-208. Cambridge, MA: MIT Press.
Attention 217
Lavie N (2005) Distracted and confused? Selective attention under load. Trends in
Cognitive Sciences 9: 75-82.
Nobre AC and Coull JT (eds.) (2010) Attention and Time. Oxford: Oxford University
Press.
Posner Ml (1978) Chronometric Explorations of Mind. Hillsdale, NJ: Erlbaum.
Posner Ml (ed.) (2004) Cognitive Neuroscience of Attention. New York: Guilford.
Prinzmetal W, McCool C, and Park S (2005) Attention: Reaction time and accuracy reveal
different mechanisms. Journal of Experimental Psychology: General 134: 73-92.
Quinlan PT (2003) Visual feature integration theory: Past, present, and future.
Psychological Bulletin 129: 643-673.
Simons DJ and Ambinder MS (2005) Change blindness: Theory and conseguences.
Current Directions in Psychological Science 1 4: 44-48.
Spence C and Driver J (eds.) (2004) Crossmodal Space and Crossmodai Attention.
Oxford, UK: Oxford University Press.
Spence C and Parise C (2010) Prior entry. Consciousness and Cognition
19: 364-379.
Wright RD and Ward LM (2008) Orienting of Attention. Oxford: Oxford University Press.
Relevant Websites
http://www.scholarpedia.org/article/Attention.
http://www.scholarpedia.org/article/Crossmodal_attention.
http://en.wikipedia.org/wiki/Attention.
Attention Deficit Hyperactivity Disorder
M Simonescu, K M Antshel, and S V Faraone, State University of New York - Upstate Medical University, New York, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Dopamine and norepinephrine Catecholamine
neurotransmitters that regulate brain activity.
Executive functions Executive functions represent an
umbrella construct that includes a collection of interrelated
functions responsible for problem-solving behavior.
Executive functions include the ability to formulate goals,
to focus and sustain attention, to inhibit impulsive
responses, to generate effective and efficient problem
solving, organizational, and learning strategies, and to
monitor and correct one's performance when necessary
(e.g., self- regulation).
Off label use Legal use of an available medication in a
manner or for a condition not approved by the Food and
Drug Administration.
Prefrontal cortex The most highly developed part of the
human brain, which is responsible for executive functions
and higher levels of thinking.
Psychiatric comorbidity The presence of one or more
psychiatric disorders in addition to ADHD.
Historical Overview
One of the earliest descriptions of a clinical phenotype that
we now associate with attention deficit/hyperactivity disorder
(ADHD) can be found as early as 1902 in the works of Sir
George Frederick Still, a British pediatrician. In a series of lec-
tures, he presented a number of cases with 'a quite abnormal
incapacity for sustained attention,' deficits in 'volitional inhibi-
tion,' and a 'defect in the moral control over their own behav-
ior.' Over the decades, this clinical constellation has been
labeled 'minimal brain dysfunction,' 'hyperkinetic impulse dis-
order,' attention deficit disorder (ADD) with or without hyper-
activity, and finally, in 1987, attention deficit/hyperactivity
disorder (ADHD).
Introduction
The operational definition of ADHD is straightforward: the
presence of significant and developmentally inappropriate
symptoms of inattention, impulsivity and/or hyperactivity
in more than one setting, for more than 6 months and with
an onset before age 7. These symptoms must be worse than
expected for the patient's age. Children with these symptoms
are at increased risk of academic underachievement and a
variety of interpersonal difficulties, most notably disruptive
behaviors. In addition to their core symptoms of inattention
and hyperactivity/impulsivity, individuals with ADHD tend to
(a) have problem-solving and social skill deficits, which can
lead to social marginalization, (b) have more accidents and
visits to local emergency rooms, (c) be more likely to smoke
and use alcohol and illicit drugs, (d) be less likely to finish
high school and college, (e) have children earlier, have more
partners, and spend less time in relationships, (f) have a
harder time maintaining a job, and (g) have a higher rate of
motor vehicle accidents. Although not all those with ADHD
show all of these features, a defining feature of ADHD is that it
causes serious problems in the patient's life. In fact, the pres-
ence of serious problems is essential to make an ADHD
diagnosis.
Epidemiology
The prevalence of ADHD in children and adolescents is around
5-10% with most results being between 4.2% and 6.3%,
although a recent epidemiological study suggests that the prev-
alence may be as high as 9.5%. Boys are diagnosed more
frequently with ADHD than girls although this gender differ-
ence is less extreme in community samples. The most signifi-
cant risk factor for ADHD in a child is presence of parental
ADHD (up to eightfold) followed by low birth weight, and
in utero exposure to neurotoxic substances such as alcohol,
pesticides, and tobacco.
Etiology
ADHD is a complex trait that likely involves an interplay of
numerous genetic and environmental effects. ADHD is one
of the most heritable psychiatric conditions, as heritable as
schizophrenia or bipolar disorder. Certain environments (ges-
tational nicotine exposure, consumption of food additives,
etc.) appear to contribute to ADHD risk in the presence of
some but not other gene variants. In addition to genetics
and environmental risks, the neurotransmitters, dopamine,
and norepinephrine have received attention as being relevant
to understanding ADHD possibly because medications which
alter the transmission of these chemicals in the brain are effec-
tive for treating ADHD.
Neurologically, the prefrontal cortex appears relevant to
understanding ADHD. The functioning of prefrontal cortex
requires dopamine and norepinephrine and plays a role in
cognitive functions such as executive functions. Executive func-
tions are an umbrella term which refers to goal-oriented
problem-solving. Included in the definition of executive func-
tions are abilities such as self-regulation of attention, behavior,
and emotion. Low levels of dopamine in the prefrontal cortex
can cause the prefrontal cortex to have difficulties maintaining
effective problem-solving behaviors.
The prefrontal cortex has many reciprocal connections with
other brain regions including the striatum (caudate nucleus,
218
Attention Deficit Hyperactivity Disorder 219
putamen), cerebellum, and parietal cortex. Research has indi-
cated that some of these brain regions are slightly smaller or
have decreased activation in ADHD.
Assessment Issues
The symptoms of ADHD are currently defined by the American
Psychiatric Association's (APA) Diagnostic and Statistical Man-
ual of Mental Disorders, fourth edition, text revision (DSM-
IV-TR). To be diagnosed with ADHD, an individual must
have persistent and impairing symptoms of inattention, impul-
sivity, and/or hyperactivity. ADHD is diagnosed when six or
more of the nine symptoms of inattention or hyperactivity-
impulsivity occur for at least 6 months, occur in two or more
settings, and impair the individuals' functioning (see Table 1
for DSM-IV-TR symptoms).
The DSM-IV-TR is currently being revised to update the
diagnosis based on recent research findings. For example,
the DSM-IV-TR requires that some impairing ADHD symp-
toms be present before age 7. The age of onset criterion (before
age 7) is highly controversial and is likely to be dropped from
the next iteration of the DSM. In its place is likely an age of
onset before early adolescence (~ age 12). The decision to
change the age of onset criterion is based primarily on a lack
of evidence differentiating children with onset before and after
age 7. This and other changes are anticipated for the fifth
edition of Diagnostic and Statistical Manual of Mental Disorders
(DSM-5).
Diagnosing ADHD in adolescents and adults is often
more challenging than diagnosing ADHD in children. For
Table 1 DSM-IV-TR criterion A symptoms
Inattention
1. Often does not pay close attention to details or makes careless
mistakes in schoolwork, work, or other activities
2. Often has trouble keeping attention on tasks or play activities
3. Often does not seem to listen when spoken to directly
4. Often does not follow instructions and fails to finish schoolwork,
chores, or duties in the workplace (not due to oppositional behavior or
failure to understand instructions)
5. Often has trouble organizing activities
6. Often avoids, dislikes, or does not want to perform tasks that take a
lot of mental effort for a long period of time (such as schoolwork or
homework)
7. Often loses items needed for tasks and activities (e.g., toys, school
assignments, pencils, books, or tools)
8. Is often easily distracted
9. Is often forgetful in daily activities
Hyperactivity/impulsivity
1 . Often fidgets with hands or feet or squirms in seat
2. Often gets up from seat when remaining in seat is expected
3. Often runs about or climbs when and where it is not appropriate
(adolescents or adults may feel very restless)
4. Often has trouble playing or enjoying leisure activities quietly
5. Is often ‘on the go’ or often acts as if ‘driven by a motor’
6. Often talks excessively
7. Often blurts out answers before questions have been finished
8. Often has trouble waiting for one’s turn
9. Often interrupts or intrudes on others (e.g., butts into conversations
or games)
example, children are often more carefully supervised by par-
ents and teachers (usually only one) who are able to observe
much of the child's functioning. However, in adolescence,
many more teachers are involved and outside of school, ado-
lescents have less contact with parents. For instance, parents
may be able to provide information regarding homework
completion and school grades but may not have as much
information on peer interactions and social relationships. In
adults, most are not involved in school and thus, obtaining
information from others who know the adult (e.g., spouse,
parent, employer) is important yet can be difficult to obtain.
This may be even more important as significant discrep-
ancies may be present between self report of symptoms and
the reports from others. Many times, the self-report will un-
derestimate the level of symptoms and the amount of
impairment.
Three subtypes of ADHD now exist in the DSM-IV-TR:
Predominantly inattentive subtype (only has threshold levels
of impairing inattentive symptoms; occurs in 30-40% of all
ADHD), predominantly hyperactive/impulsive subtype (only
has threshold levels of impairing hyperactive/impulsive symp-
toms; occurs in <5% of all ADHD, nearly always in preschool
children), and the combined subtype (has threshold levels of
impairing inattentive and hyperactive/impulsive symptoms;
occurs in 50-60% of all ADHD).
For some patients, these ADHD impairing symptoms may
occur in most situations, including high-interest activities.
However, for most, these attention and inhibition difficulties
are manifested primarily in settings requiring self-restraint,
persistence, and a high level of concentration to relatively
uninteresting activities. In children with ADHD, these difficul-
ties negatively affect classroom or social functioning and may
result in limited learning, disruption of class activities, poor
work completion, impaired peer interactions, and increased
conflict with teachers/parents.
It is also possible that an adolescent or adult may be present
for an ADHD evaluation having never been formally diag-
nosed with ADHD. The age of referral and recognition of the
problem is older for those with the inattentive type, compared
with the combined type. Intelligence, family support, and
academic placement can be protective and in some cases
appear to delay the onset of evident symptoms or impairment
to much later than specified in the DSM. For example, some-
times highly intelligent individuals show no academic impair-
ments relative to peers until they complete high school,
college, or even later
Careful attention to potential comorbidities is critical in
ADHD, particularly learning disabilities as well as mood, anx-
iety, and substance use disorders. While psychological testing
can be helpful in determining strengths/weaknesses and asses-
sing for the presence of learning disabilities, psychological
testing is not necessary to either rule in or rule out the presence
of ADHD. Rather, an ADHD diagnosis is based on the careful
integration of clinical information from as wide a variety of
sources that are available. Rating scales and other clinical
instruments have been devised to assist with this. An ADHD
diagnosis is not based on brain scans or any other biological
test. Despite this lack of biologically sensitive and specific
tests, ADHD is a well-validated and highly reliable clinical
diagnosis.
220 Attention Deficit Hyperactivity Disorder
Developmental Course
Most often, ADHD symptoms and associated problems occur
as early as age 3 or 4 and the first manifestations tend to reflect
problems of hyperactivity and impulsivity (disinhibition).
Issues pertaining to inattention tend to be more easily noticed
later around age 5-7 when the child starts attending school.
In children with ADHD, difficulties exerting physical self-
control are central. Children with ADHD fidget, leave their seat
when seating is expected, have a hard time waiting for their
turn, are often on the go, and run, climb, or touch objects
inappropriately. Children with ADHD also have difficulties
exerting verbal self-control. Children with ADHD talk exces-
sively, even when restraint is expected, blurt out answers even
before the question was completed, and tend to interrupt
others or intrude in other's conversations. In adolescents and
adults with ADHD, these physical and verbal self-control diffi-
culties appear more as restlessness.
During school years, the problems of inattention and
school productivity tend to be almost universal. As the child
is progressing into adolescence, often the symptoms of hyper-
activity are less obvious but impairment related to impulsive
behaviors tends to stay significant. Academic problems that
might have been less noticeable or were treated effectively
during elementary school may become much more of a prob-
lem. This is possibly a function of the increased demands for
functional independence that are inherent in the middle- and
high-school curriculum. In middle and high school, adoles-
cents are expected to become more independent of adult super-
vision. Adolescents also change classes and are exposed to
multiple teachers, teaching styles, and homework demands.
Procrastination, forgetfulness, and poor persistence often
become the most academically impairing symptoms.
In adolescents, peer relationships rise in importance. Ado-
lescents with ADHD often struggle in peer relationships,
typically as a function of their impulsivity. Emotional impul-
siveness manifested as low frustration tolerance, quickness to
anger, being impatient, and being easily excitable or annoyed
is most predictive of social relationship problems for adoles-
cents and adults with ADHD.
In children, adolescents, and adults with ADHD, the core
ADHD symptoms can lead to a wide variety of other associated
features. Individuals with ADHD tend to have a hard time
regulating arousal levels to meet contextual demands, tend to
seek more immediate reinforcements, tend to discount more
steeply the value of future rewards, and have a poor sense of
time. In addition to the core symptoms, these associated fea-
tures contribute to functional impairments. In fact, the overall
relationship between symptoms and impairment in ADHD
seems to be such that as an individual ages, the core symptoms
of ADHD, especially hyperactive and impulsive symptoms,
decline while functional impairment persists or worsens
through adolescence and into adulthood. Due to the decline
of ADHD symptoms throughout adolescence, by adulthood,
many grown-up ADHD children will no longer show a high
level of ADHD symptoms. However, despite this decline of
symptoms, about two-thirds of children with ADHD continue
to have some impairing symptoms of ADHD as they enter
adult life. These impairing symptoms can cause serious pro-
blems if not treated.
Treatment Issues
The treatment of ADHD typically includes the following types
of interventions: (a) medication(s), (b) behavioral interven-
tions (also known as psychosocial treatments), and (c) a com-
bination of medication and behavioral interventions.
Stimulant Medications
In 1937, Dr. Charles Bradley from East Providence, RI,
reported in the American Journal of Psychiatry his findings
about behavioral changes in children when using an amphet-
amine drug called benzedrine. From his description of these
children, it is likely that most would have been diagnosed as
ADHD using modern criteria. His initial thought was that
Benzedrine would alleviate the headaches these youth experi-
enced after an experimental procedure. The headaches per-
sisted, but the children's teachers reported "a spectacular
improvement in school performance in half of the children."
Dr. Bradley also found that "a large proportion of the patients
became emotionally subdued without, however, losing interest
in their surroundings." These results represent historically one
of the first examples of using pharmacology to address behav-
ioral disorders.
Over the last 70 years, the use of amphetamine and other
stimulant medications for treating ADHD has grown consider-
ably. Stimulants are now used for treating symptoms of ADHD
across the life span from preschoolers to school-age children to
adolescents and later adults. The two stimulant medications
most often used in the treatment of ADHD are methylpheni-
date and amphetamine.
Methylphenidate, which has been used since the early
1960s, blocks reuptake of the neurotransmitters norepineph-
rine and dopamine by blocking the activity of their respective
transporters in the brain. As a result, the synaptic availability
of norepinephrine and dopamine is increased. Methylpheni-
date is available in multiple formulations including immedi-
ate and extended release. Methylphenidate has a short
elimination half-life (1.5-5 h) and this creates significant
problems in practical treatment. Therefore, extended release
formulations are more often prescribed because one daily
dose is often sufficient for treating symptoms throughout
the day. Amphetamine was discovered at the end of the
nineteenth century and is also used in the treatment of
ADHD. Amphetamine is also available as immediate or
extended release formulations.
Over 70% of individuals respond to initial trials of either
methyphenidate or amphetamine; of those who fail the initial
trial, an additional 70% generally responds to the other
class. Both medications usually require adjustments in doses
(including when doses are taken) in order to maximize
effectiveness.
Both amphetamine and methylphenidate are associated
with small but generally insignificant changes in pulse and
blood pressure. The more common side effects of stimulants
include delayed sleep, reduced appetite, headaches, and mild
abdominal pain growth delay. There is a good deal of individ-
ual variability in these side effects. While significant adverse
events may occur, when prescribed correctly, stimulants are
generally well tolerated.
Attention Deficit Hyperactivity Disorder 221
Both amphetamine and methylphenidate also have poten-
tial for misuse, abuse, and dependence. Misuse refers to situa-
tions in which the medication is used for reasons other than
treating ADHD. For example, some college students obtain
stimulant medication illegally and use it to stay up late to
study. Abuse and dependence refer to situations in which the
medication is used to get high, which could lead to addiction.
Although these are concerns, there is no convincing evidence
that the therapeutic use of stimulants causes abuse or depen-
dence. In fact, long-term follow-up studies of stimulant-treated
children with ADHD suggest the opposite effect. Treatment
with stimulant medication generally decreases the risk for
later drug abuse and dependence.
The rate of failure to adhere to treatment with stimulants
may be as high as 20-65%. There are multiple factors contri-
buting to poor adherence including perceived stigma, denial,
externalization of the problem, unwanted side effects (head-
aches, emotional blunting etc.), the cost of the medication, as
well as the fact that ADHD symptoms (e.g., forgetfulness) may
directly contribute to poor adherence.
Nonstimulant Medications
Stimulant medications are usually the first line of treatment for
ADHD due to their high efficacy and relatively long history of
use. However, there are alternatives, including three food and
drug administration (FDA)-approved nonstimulants: atomox-
etine, extended release guanfacine, and extended release cloni-
dine. Other medications such as bupropion, desipramine,
venlafaxine, and modafinil are sometimes used 'off label' in
the treatment of ADHD, especially in the adult population.
'Off label' means that the use of the medication for ADHD
has not been approved by the FDA.
Atomoxetine is a selective norepinephrine reuptake inhibi-
tor (SNRI) that increases the availability of synaptic norepi-
nephrine, especially in the prefrontal cortex. One of the
advantages of atomoxetine (and other nonstimulants) is the
absence of a potential for abuse, which makes it a preferred
agent in many situations where this can be a concern, espe-
cially in adults. Atomoxetine tends to be well tolerated; head-
aches, nausea, and loss of appetite are the most common side
effects. Some of the rare events in treatment include increased
suicidal thinking and behavior, unusual changes in behavior,
and liver problems, potentially severe. While numerous studies
found superiority of atomoxetine relative to placebo, atomox-
etine is less effective than stimulant medications for most
patients.
Guanfacine is an a-2 agonist with some selectivity for the
2A postsynaptic receptors. Guanfacine and other a-2 agonists
increase noradrenergic brain activity; yet actually serve to regu-
late glutaminergic transmissions. A similar agent, clonidine, is
also frequently used; yet clonidine is less selective at the level of
2A receptors. Both guanfacine and clonidine have similar side
effects including somnolence, dry mouth, hypotension, and
constipation. Both of these medications will likely increase in
use in coming years because long-acting versions of both were
recently approved by the FDA.
While the medication treatments for ADHD are effective,
'normalization' of functioning with medication treatments
only occurs in a minority of those with ADHD. For this
reason, most in the field recommend that after medication
treatment has been instituted, if any significant problems
remain, these should be addressed with behavioral (psychoso-
cial) interventions.
Psychosocial Interventions for Children
Behavioral parent training (BPT) programs are effective for
children with disruptive behaviors whether or not they have
co-occurring attentional/hyperactive difficulties. BPT techni-
ques generally consist of training parents in the use of reward
and punishment following appropriate and inappropriate
behaviors. Reward procedures have typically relied on praise,
privileges, or tokens while punishment methods have usually
been loss of positive attention, privileges, or tokens or formal
time out from reinforcement.
BPT programs generally consist of weekly training sessions,
either in groups of parents or with individuals, each focusing
on a discrete method. These methods can be grouped into
three basic types of procedures: (1) those that manipulate the
setting events that may precede or surround a child's tasks or
activities so as to increase positive or negative behavior (i.e.,
parental commands, task demands, teacher instructions, etc.);
(2) those which may restructure the tasks to be done (reduce
work quotas, insert more interesting task materials, etc.);
(3) and those that manipulate the nature of the consequences
for child behavior in that setting (i.e., attention, praise, token
reinforcement, punishment, etc.).
Teachers often receive explicit training in classroom behav-
ioral management during their training and education. Proper
classroom placement, adequate attention to concomitant
learning disabilities, and reasonable accommodations that
might include classroom seating, extended time on tests, and
modified work assignments are also commonly employed in
school settings. Accommodations (including extended time)
on standardized tests are also relatively common. Curriculum
modifications, strategy training, and other cognitive behavioral
approaches are less reliably associated with classroom behav-
ioral improvements in children.
In general, medication management along with individu-
ally tailored psychosocial treatments are the most commonly
employed interventions for managing children and adoles-
cents with ADHD.
Adult Psychosocial Interventions
As noted above, behavioral treatments for ADHD children
involve direct contact with parents and teachers and less direct
intervention with the child. Cognitive behavioral therapy
(CBT) is a more individual-delivered therapy which involves
less direct contact with parents and teachers and more contact
with patients. Unlike behavior therapy which focuses exclu-
sively on changing behaviors, CBT also focuses on changing a
person's cognitions or thoughts and how they can use their
thoughts to overcome real-world problems associated with
their ADHD. CBT for ADHD children historically included
many different treatment components such as self-instruction
training, problem solving training, and social skills training; as
psychosocial interventions. Research studies suggest that CBT
is not useful for treating children with ADHD. In fact, due to its
222 Attention Deficit Hyperactivity Disorder
limited efficacy in children, CBT was not included in the MTA
treatment protocols. In adults, however, CBT holds some
promise as a psychosocial component to an integrated treat-
ment plan.
There are several CBT protocols for managing adult ADHD.
Most protocols address the following topics: (1) education
about ADHD; (2) organizational/planning training usually
including teaching adults with ADHD how to use a calendar/
daily planner, how to use 'to-do' lists, and how to break down
a large task into more manageable chunks; (3) reducing dis-
tractibility through a variety of means, including managing
environmental distractions, reducing large tasks into smaller
chunks that correspond to the length of the adult's attention
span, and the use of alarms/timers to encourage sustained
focus; (4) reducing procrastination; (5) improving communi-
cation/social skills; (6) reducing anger and frustration through
managing maladaptive thoughts and dysfunctional beliefs.
Although there is relatively little research on the use of CBT
for treating adult ADHD, initial studies show that ADHD
adults treated with CBT and medication have better outcomes
than ADHD adults who are treated with medication alone.
Complementary and Alternative Treatments
The pharmacological treatment of ADHD is effective and safe
for the large majority of patients. Nevertheless, because many
parents object, in principle, to the idea of using psychiatric
medication with their children, they seek help from comple-
mentary and alternative medical therapies.
Many alternative approaches for treating ADHD have been
attempted. These use nutritional interventions; natural health
products such as essential fatty acids, vitamins, and minerals;
and other health supplements. Currently, there is no evidence
to support recommending such interventions systematically,
although anecdotal reports of improvement are numerous.
When alternative treatments have been studied systematically,
results are typically negative. For example, studies of zinc and
omega-3 fatty acid supplementation show they are not effective
for treating ADHD. Because most alternative treatments for
ADHD have not been adequately studied, parents and patients
should use them with extreme caution, keeping in mind that
any concerns about the side effects of medications should be
balanced by the risks associated with not treating ADHD as
effectively as possible.
One alternative tteatment that has gained some traction is
use of electroencephalogram (EEG) neurofeedback. The EEG
measures the electrical activity of the brain and is known in the
popular language as 'brain waves.' Attempts to describe patterns
of EEG activity in children with ADHD go back many decades.
These studies suggest that ADHD children show increased theta
and decreased beta and alpha activity. Neurofeedback has the
child performing an activity that requires sustained attention.
The child will receive negative feedback when the EEG reflects
an increased theta/beta ratio and positive feedback otherwise.
Initial studies suggest that this method is useful for some
ADHD children. However, large randomized, well-controlled
studies will likely clarify this issue. In scientific circles suspicion
persists as we lack currently a convincing model correlating the
EEG findings and the putative correction of the theta/beta ratio
with executive functions and attentional processes.
Challenges and Perspectives
Over the past several decades, scientists have made important
sttides in understanding the causes of ADHD, refining its diag-
nosis, and developing new and more effective treatments.
However, much more work remains to be done. One impor-
tant challenge, which is faced by nearly all psychiatric disor-
ders, is the lack of a biological test for the disorder. Although
we know that ADHD and non-ADHD patients differ on mea-
sures of brain structure and functioning and that some genetic
variants predispose to ADHD, these measures are not suffi-
ciently accurate to use as a diagnostic test.
Scientists also need to provide a richer description of the
causes of ADHD and the effects that these causes have on the
brains of ADHD patients. Our hope is that future discoveries
along these lines will lead to newer and more effective treat-
ments for ADHD. New treatments are needed for several rea-
sons. Current treatments for ADHD reduce symptoms and the
risk for impairing behaviors, but in most cases they do not
eliminate ADHD or 'cure' the disorder. Further studies of the
brains and genes of ADHD patients may lead us to such
improved tteatments.
Current medication tteatments for ADHD do a good job of
reducing the symptoms of inattention, hyperactivity, and
impulsivity. They are, however, less effective in treating other
features of ADHD such as executive dysfunction or problems
with the self-regulation of emotions. More work is needed
to develop new medicines for such problems or to combine
medicine and psychosocial treatments in a manner that
addresses the full range of symptoms and impairments caused
by ADHD.
Conclusion
In concluding, it is important to recognize the social context
for ADHD in the United States and many other countries. It is
not unusual to read on the internet or in newspapers, stories
that question the validity of ADHD as a true disorder or
describe current treatments as remarkably dangerous. Many
of these stories are simply wrong or, at best, exaggerations.
Parents and patients should rest assured that the disorder we
call ADHD has received substantial scientific scrutiny and can
be considered as valid as any other medical disorder that might
be diagnosed in their child. They should also rest assured that
medical tteatments approved by the FDA have been sufficiently
studied for safety and effectiveness.
People with ADHD should also understand that, although
this article has focused on the clinical features and impairments
associated with ADHD, having this disorder does not define the
patient as hopelessly impaired or 'abnormal.' All people have
sttengths and weaknesses. ADHD is a weakness that leads to
problems, but ADHD people often have many other strengths
that help them lead productive and satisfying lives.
See also: Academic Achievement; Anxiety Disorders; Attention;
Cognitive Behavior Therapy; Depression; Developmental
Psychopathology; Neuroexecutive Function.
Attention Deficit Hyperactivity Disorder 223
Further Reading
Arnold LE (1999) Treatment alternatives for attention-deficit/hyperactivity disorder
(ADHD). Journal of Attention Disorders 3: 30-48.
Barkley RA (2006) Attention Deficit Hyperactivity Disorder: A Handbook for Diagnosis
and Treatment, 3rd edn. New York: Guilford.
Barkley RA and Murphy KR (2006) Attention Deficit Hyperactivity Disorder: A Clinical
Workbook. New York: Guilford.
Barkley RA, Murphy KR, and Fischer M (2008) ADHD in Adults: What the Science Says.
New York: Guilford.
Biederman J, Faraone SV, Keenan K, et al. (1992) Further evidence for family-genetic
risk factors in attention deficit hyperactivity disorder. Patterns of comorbidity in
probands and relatives in psychiatrically and pediatrically referred samples.
Archives of General Psychiatry 49: 728-738.
Biederman J, Faraone S, Mick E, et al. (1996) Attention deficit hyperactivity disorder and
juvenile mania: An overlooked comorbidity? Journal of the American Academy of
Child and Adolescent Psychiatry 35: 997-1 008.
Bradley C (1937) The behavior of children receiving benzedrine. American Journal of
Psychiatry HA-. 577-585.
Cantwell DP (1972) Psychiatric illness in the families of hyperactive children. Archives
of General Psychiatry 27: 41 4-41 7.
Conners CK (1969) A teacher rating scale for use in drug studies with children. The
American Journal of Psychiatry 126: 884-888.
Faraone SV, Perlis RH, Doyle AE, et al. (2005) Molecular genetics of attention deficit/
hyperactivity disorder. Biological Psychiatry 57: 1313-1323.
Fischer M, Barkley RA, Fletcher KE, and Smallish L (1993) The adolescent outcome of
hyperactive children: Predictors of psychiatric, academic, social, and emotional
adjustment. Journal of the American Academy of Child and Adolescent Psychiatry
32: 324-332.
Gaub M and Carlson CL (1997) Gender differences in ADHD: A meta-analysis and
critical review. Journal of the American Academy of Child and Adolescent Psychiatry
36: 1036-1045.
Nigg J (2006) What Causes ADHD? Understanding What Goes Wrong and Why.
New York: Guilford.
Pliszka SR (2009) Treating ADHD and Comorbid Disorders: Psychosocial and
Psychopharmacological Interventions. New York: Guilford.
Valera EM, Faraone SV, Murray KE, and Seidman LJ (2007) Meta-analysis of structural
imaging findings in attention deficit/hyperactivity disorder. Biological Psychiatry
61: 1361-1369.
Young J (2007) ADHD Grown Up: A Guide to Adolescent and Adult ADHD. New York:
W. H. Norton.
Relevant Websites
http://www.aacap.org - American Academy of Child & Adolescent Psychiatry.
http://www.aap.org - American Academy of Pediatrics.
http://www.chadd.org - CHADD Organization.
http://www.ldanatl.org - Learning Disabilities Association of America (LDA).
http://www.nichcy.org - National Information Center for Children and Youth with
Disabilities.
Attitude Change
R E Petty, Ohio State University, Columbus, OH, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Attitude A global evaluation of a person, object, or issue
indicating the extent to which it is liked or disliked.
Attitude change Modifying one's evaluation to become
different from what it was.
Attitude strength The extent to which an attitude is
persistent, resistant, and influences thinking and behavior.
Central route Persuasion through effortful thought and
extensive elaboration that involves generating issue-relevant
thoughts.
Dissonance An attitude change technique that
involves having a person engage in inconsistent thought
or action.
Inoculation Providing arguments against one's attitude in
an attempt to strengthen it from future attacks.
Peripheral route Persuasion through less effortful
mechanisms including mental shortcuts, simple
associations, and the impact of contextual cues.
Resistance The extent to which attitudes remain firm when
confronted with an attempt to persuade.
Attitudes refer to people's global and relatively enduring (i.e.,
stored in long-term memory) evaluations of objects, issues, or
persons (e.g., I oppose capital punishment; I like the Presi-
dent). Numerous procedures have been developed to modify
these evaluations with some change techniques involving con-
siderable thinking about the attitude object and some requir-
ing little. Attitudes are one of the most studied and important
constructs in psychology because of the critical role of attitudes
in guiding everyday choices and behavior.
Overview of Attitudes and Attitude Change
Attitudes are based on some combination of cognitive, behav-
ioral, and affective (emotional) influences and are typically
measured by self-report scales such as the semantic differential,
where a person rates the target on bipolar evaluative dimensions
such as how good/bad or favorable/unfavorable it is. Increas-
ingly, in addition to assessing people's explicit or deliberative
attitudes, assessments are made of the evaluations that come to
mind automatically without much reflection (i.e., quick evalua-
tive associations) using implicit measures. Measures such as
Greenwald's Implicit Association Test (IAT) and Fazio's evalua-
tive priming measure are popular in this regard. Often, implicit
and explicit measures yield the same outcome, but sometimes
there are discrepancies which might suggest that people are
unwilling to report their attitudes due to social desirability con-
cerns or might be unable to report them due to limited access.
Furthermore, in addition to assessing attitudes on a positive
to negative continuum, scholars have also found it useful to
assess attitudes on dimensions other than their valence such as
their accessibility (how quickly the attitude comes to mind),
ambivalence (how consistent the basis of the attitude is), and
certainty (how confident people are in the validity of their
attitudes). These indicators of attitude 'strength' are useful in
determining which attitudes are consequential and which are
not. Strong attitudes are those that persist over time, are resis-
tant to change, and predict other judgments and actions. At any
given moment, one's expressed evaluation can be influenced
by a variety of contextual factors, but a common assumption is
that one's core 'attitude' is the underlying evaluation that is
capable of guiding behavior (one's actions), cognition (one's
thoughts and memories), and affect (emotional reactions).
Attitude change occurs when one's core evaluation shifts
from one meaningful value to another, and is typically inferred
from a change in either an implicit (automatic) or explicit
(deliberative) measure of evaluation. Often, an attitude change
induction produces change on both kinds of measures but
sometimes change is more likely on one type of measure
than another. Most studies of attitude change involve exposing
individuals to a persuasive communication of some sort, but as
explained shortly, some attitude change techniques do not
involve exposure to any message. The earliest work on attitude
change attempted to examine which variables and procedures
increased and which decreased the likelihood of change.
Two Routes to Attitude Change
After numerous studies, the accumulated evidence suggested
that even the simplest variables (e.g., being in a positive vs.
negative mood; using a high vs. low credible source) sometimes
increased, sometimes decreased, and sometimes had no impact
on the likelihood that a person's attitude would change.
Numerous theories and psychological processes were proposed
to account for these divergent results. Even though the many
theories of attitude change that were developed have different
names, postulates, and particular effects and variables that they
explain, the many different theories of attitude change can be
thought of as emphasizing just two relatively distinct routes to
persuasion : the central route and the peripheral route.
Central Route Approaches to Attitude Change
The first attitude change technique, persuasion via the central
route, focuses on the information that a person has about the
cenual merits of the object under consideration. Some of the
theories and techniques that utilize the central route assume
that comprehending and learning the information presented is
224
Attitude Change 225
critical for attitude change, whereas others focus more on the
evaluation, elaboration, and integration of this information.
Message Learning Approach
One of the most influential programs of research on attitude
change was that undertaken by Carl Hovland and his collea-
gues at Yale University in the years following World War II.
The Yale group studied how source, message, recipient, and
channel factors affected the comprehension, acceptance, and
retention of the arguments in a persuasive communication.
Although no formal theory tied together the many experiments
conducted by this group, they often attempted to explain the
results obtained in terms of general learning principles, such as
the more message content you learned, the more your attitudes
should change. Contemporary research shows that people can
be persuaded without learning or remembering any of the
message content. That is, people are sometimes persuaded
solely by the 'cues' associated with the message (e.g., the source
is expert). Or, the message might elicit a favorable thought that
persists in the absence of memory for the information that
provoked it. Message learning appears to be most important
when people are not engaged in an online evaluation of the
information presented to them such as when they do not think
they have to form an opinion at the time of information expo-
sure. In such cases, subsequent attitudes may be dependent on
the valence of information they have learned and can recall.
Self-Persuasion Approach
Self-persuasion theories hold that people's attitudes can
change in the absence of any new external information. This
is because people can self-generate reasons to favor or disfavor
any position. The powerful and persisting effects of completely
self-generated messages were shown in early research on 'role-
playing' where people were asked to generate messages on
certain topics (e.g., the dangers of smoking). The subsequent
attitudes of these people were compared to those who had either
passively listened to the communication or who had received
no message. A consistent result was that active generation of a
message was a successful strategy for producing attitude change,
and these changes persisted longer than changes based on pas-
sive exposure to a communication. Finally, merely asking
someone to think about an issue, object, or person can lead to
attitude change as a result of the evaluative thoughts generated.
Cognitive response theorists hold that just as one's one thoughts
can produce change in the absence of a message, so too are
one's own thoughts responsible for attitude change even when
a persuasive message is presented. That is, to the extent that a
person's thoughts in response to the message are favorable,
persuasion should result, but to the extent that they are unfa-
vorable (e.g., counterarguments), resistance or even boomer-
ang is more likely. These theorists hold that persistence of
persuasion depends upon the decay function for cognitive
responses rather than message arguments per se.
Expectancy-Value Approach
The message learning and self-persuasion approaches focus on
the information (either externally or internally generated) that
is responsible for persuasion. Neither approach has much to
say about the particular features of the information that are
critical for influencing attitudes. In contrast, expectancy-value
theorists analyze attitudes by focusing on the extent to which
people expect the attitude issue to be related to important
values or produce positive and negative consequences. In one
influential expectancy-value model, Fishbein and Ajzen's theory
of reasoned action holds that the attributes (or consequences)
associated with an attitude object are evaluated along two
dimensions - the likelihood that an attribute or consequence
is associated with the object and the desirability of that attribute
or consequence. If a persuasive message says that raising taxes
will lead to reduced crime, the effectiveness of this argument
should depend on how likely people think it is that crime will
be reduced if taxes are increased (likelihood), and how favor-
ably they view the outcome of reducing crime (desirability).
Although some questions have been raised about the necessity
of one or the other of these components, a large body of
research supports the idea that attitudes are more favorable
the more that likely-desirable consequences (or attributes)
and unlikely-undesirable consequences are associated with
them. The major implication of this theory for persuasion is
that a message will produce attitude change to the extent that
it introduces new attributes of an object, or produces a change
in the likelihood and/or the desirability components of an
already accepted attribute. Another proposition of this theory
is that the items of information constituting an attitude are
combined in an additive fashion. Other theorists, however,
have contended that an averaging mechanism is more
appropriate.
Self-Validation Approach
Not all of the thoughts that people have with respect to an
attitude object will necessarily be used to form an overall evalu-
ative judgment. According to the self-validation approach, peo-
ple are more likely to rely on thoughts in which they have
confidence (i.e., see as valid) than thoughts about which they
have doubt. For example, if two people have equal numbers of
favorable thoughts about a proposal and each assesses the like-
lihoods and desirabilities of the consequences as equivalent, the
person who has high confidence in the thoughts generated will
show more attitude change than the person who has doubts
about these thoughts. Numerous studies have identified a wide
variety of variables that can affect thought confidence. One well-
studied factor is the ease with which thoughts come to mind.
For example, it is easier to generate two reasons to stop smoking
rather than 12. Because things that come to mind easily are seen
as more valid, people who generate two reasons could show
more attitude change than those who generate 12, reversing the
typical effect of number of thoughts on persuasion. This is
because when people generate only two reasons, they have
confidence in these reasons, but when they generate 12, there
is less confidence in their validity. Other variables have been
shown to affect thought confidence and thus reliance on one's
thoughts. These variables include whether people are nodding
(up and down) rather than shaking (side to side) their heads
during the persuasion attempt, and whether they are made to
feel powerful or happy right after thought generation. If people
have generated mostly favorable thoughts during the message,
226 Attitude Change
increasing thought confidence will lead to more persuasion, but
if people have generated mostly unfavorable thoughts during
the message, increasing thought confidence will result in
reduced persuasion. Finally, it is important to note that people
assess their thought confidence primarily when they are moti-
vated and able to think carefully about the message, such as
when they are high in their need for cognition or the personal
relevance of the topic is high. When motivation and ability are
low, meta-cognitive activity (i.e., thinking about one's thoughts)
is also low.
Functional Approach
In their expectancy-value theory, Fishbein and Ajzen speculate
that five to seven attributes or consequences are critical in
determining a person's overall attitude. It is not clear, however,
which particular attributes will be the most important (i.e.,
how the attributes are weighted). Functional theories of per-
suasion focus on the specific needs or functions that attitudes
serve for a person and are therefore relevant for understanding
the underlying dimensions of the attitude that are most impor-
tant to influence. For example, some attitudes are postulated to
protect people from threatening truths about themselves or to
enhance their own self-image ('ego-defensive function'), others
give expression to important values ('value-expressive func-
tion'), or help people to understand the world around them
('knowledge function') or facilitate achieving rewards and
avoiding punishments ('utilitarian function'). According to
these theories, change depends on challenging the underlying
functional basis of the attitude. Thus, if a person dislikes low-
ering taxes because of concern about social inequality (value
expressive function), an argument about the amount of money
the taxpayer will save (utilitarian function) will be ineffective.
Consistency Approach
Just as functional theories hold that attitudes serve important
needs for individuals, dissonance and related theories hold
that attitudes are often in the service of maintaining a need
for consistency among the elements in a cognitive system; in
Leon Festinger's original formulation of dissonance theory, two
elements in a cognitive system (e.g., a belief and an attitude; an
attitude and a behavior) were said to be consonant if one
followed from the other, and dissonant if one implied the
opposite of the other. Two elements could also be irrelevant
to each other. One of the more interesting dissonance situa-
tions occurs when a person's behavior is in conflict with his or
her attitudes or beliefs because behavior is usually difficult to
undo. According to the theory, dissonance, experienced as an
aversive tension, can be reduced by changing beliefs and atti-
tudes to bring them into line with the behavior. Thus, if you
were opposed to the election of Candidate Smith, it would
be inconsistent to sign a petition in favor of this candidate.
According to dissonance theory, signing such a petition would
produce discomfort that could result in a more favorable evalu-
ation of the candidate in an effort to restore consistency.
Although early dissonance research was generally support-
ive of the theory, several competing formulations were pro-
posed. Although it is now clear that many of the behaviors
described by Festinger induce in people an 'unpleasant
tension,' just as the theory predicts, current research has
begun to focus more on understanding the precise cause of
that tension. For example, some have questioned Festinger's
view that inconsistency per se produces tension in many peo-
ple. Rather, some argue that people must believe that by their
behavior they have freely chosen to bring about some foresee-
able negative consequence, or that the inconsistency involves a
critical aspect of oneself or a threat to one's positive self-
concept.
Peripheral Route Approaches
Each of the central route approaches described above assumes
that attitude change results from people actively considering
the merits of some position either in a fairly objective manner
or in a biased way (such as when seeking to restore consis-
tency). The next group of theories does not share this assump-
tion. Instead, these theories suggest that people often prefer to
conserve their limited cognitive resources and form or change
attitudes with relatively little cognitive effort. The peripheral
route focuses on eliciting attitude change without much think-
ing about information central to the merits of the attitude
issue. Thus, the peripheral approaches deal with changes
resulting from rewards, punishments, and affective experiences
that are associated directly with the attitude object, or simple
inferences that people draw about the appropriate attitude to
adopt based on their own behavior or other simple cues in the
persuasion environment.
Inference Approaches
Rather than effortfully examining and thinking about all of the
issue-relevant information available, people can make an eval-
uative inference based on some meaningful subset of informa-
tion. One popular inference approach is based on 'attribution
theory' and holds that people come to infer underlying char-
acteristics about themselves and others from the behaviors that
they observe and the situational constraints imposed on these
behaviors. Bern suggested that people sometimes have no spe-
cial knowledge of their own internal states and simply infer
their attitudes in a manner similar to that by which they infer
the attitudes of others. In his self-perception theory, Bern rea-
soned that just as people assume that the behavior of others
and the context in which it occurs provides information about
the presumed attitudes of these people, so too would a person's
own behavior provide information about the person's own
attitude. Thus, a person might reason, 'since I signed Candidate
Smith's petition, I must be in favor of her election.'
The attribution approach has also been useful in under-
standing the persuasion consequences of making inferences
about relatively simple cues. For example, when external incen-
tives (e.g., money) provide a salient explanation for a speaker's
advocacy ('he was paid to say it'), the message is less effective
than when a discounting external attribution is not possible.
Research indicates that these simple attribution processes are
most likely to influence attitudes when people are relatively
unmotivated or unable to think carefully about the issue, such
as when they have relatively little knowledge on the topic and
the issue has few anticipated personal consequences.
Attitude Change 227
Like the attributional framework, the heuristic-systematic
model of persuasion postulates that, when people are not moti-
vated or able to process all of the relevant information avail-
able, attitude change can result from the use of certain
heuristics or rules of thumb that people have learned on the
basis of past experience and observation. To the extent that
various persuasion heuristics are available in memory, they
may be invoked to evaluate persuasive communications.
For example, either because of prior personal experience or
explicit training, people can evaluate a message with many
arguments by invoking the heuristic 'the more arguments, the
more valid it is.' If so, no effortful learning or evaluation of the
actual arguments presented is necessary for influence to occur.
Approaches Emphasizing Affect
The attribution and heuristic models focus on simple cognitive
inferences that can modify attitudes. Other peripheral route
theories emphasize the role of affective processes in attitude
change. One of the most direct means of associating 'affect'
with objects, issues, or people is through classical conditioning.
In brief, conditioning occurs when an initially neutral stimulus
(the conditioned stimulus, CS) is associated with another
stimulus (the unconditioned stimulus, UCS) that is connected
directly or through prior conditioning to some response (the
unconditioned response, UCR). By pairing the LICS with the
CS, the CS becomes able to elicit a conditioned response (CR)
that is similar to the UCR. So, in Pavlov's initial conditioning
studies, when food was paired over and over again with a bell,
eventually the bell elicited salivation in the absence of food.
Considerable research has shown that attitudes can be influ-
enced by pairing initially neutral objects with stimuli about
which people already feel positively or negatively. For example,
peoples' evaluations of words, other people, political slogans,
consumer products, and persuasive communications have been
modified by pairing them with such affect-producing stimuli as
unpleasant odors and temperatures, the onset and offset of
electric shock, harsh sounds, and elating and depressing films.
People are especially susceptible to the simple transfer of affect
from one stimulus to another when the likelihood of object-
relevant thinking is rather low. Some recent research by Jones,
Fazio, and Olson suggests that conditioning in the domain of
attitudes ( evaluative conditioning ) occurs by means of a simple
misattribution process. That is, although people are actually
feeling unpleasant because of the UCS (e.g., an pleasant odor),
they misattribute (confuse) this feeling as originating from the
CS (e.g., the persuasive message).
Another procedure for modifying attitudes through simple
affective means was identified by Robert Zajonc in his work on
mere exposure. In this research, Zajonc and his colleagues have
shown consistently that when objects are presented to an
individual on repeated occasions, the mere exposure is capable
of making the individuals' attitudes toward these objects more
positive. Recent work on this phenomenon indicates that sim-
ple repetition of objects can lead to more positive evaluations
even when people do not recognize that the objects are famil-
iar. Mere exposure effects have been shown in a number of
studies using a variety of stimuli such as polygons, tones,
nonsense syllables, Chinese ideograms, photographs of faces,
and foreign words. Interestingly, what these stimuli have in
common is that they tend to be meaningless and are relatively
unlikely to elicit spontaneous thought. In fact, the simple
affective process induced by mere exposure appears to be
more successful in influencing attitudes when processing of
the repeated stimuli is minimal. When more meaningful sti-
muli have been repeated such as words or sentences, mere
exposure effects have been less common. Instead, when pro-
cessing occurs with repetition, the increased exposures enhance
the dominant cognitive response to the stimulus. Thus, repeat-
ing strong arguments tends to lead to more persuasion (at least
to the point of tedium), and repeating weak arguments tends
to lead to less persuasion.
A Dual Process Approach to Understanding
Attitude Change
Although the theories just described continue to be useful in
accounting for a variety of persuasion phenomena, much of
the contemporary literature on attitude change is guided by
one of the available 'dual process' models of judgment. For
example, one of the earliest approaches of this type, the Elabo-
ration Likelihood Model (ELM), was introduced by Petty and
Cacioppo and represented an attempt to integrate the many
seemingly conflicting findings in the persuasion literature
under one conceptual umbrella by specifying a finite number
of ways in which source, message, recipient, and contextual
variables have an impact on attitude change (for reviews of the
ELM, the related heuristic/systematic model, and other dual
process approaches). The ELM is based on the notion that
people want to form correct attitudes (i.e., those that will
prove useful in functioning in the environment) as a result of
exposure to a persuasive communication, but there are a vari-
ety of ways in which a reasonable position can be adopted.
The most effortful procedure for evaluating an advocacy
involves drawing upon prior experience and knowledge to
carefully scrutinize and think about all of the issue-relevant
information available in the current environment along the
dimensions that are perceived central to the merits of the
attitude object. According to the ELM, attitudes formed or
changed by this central route are postulated to be relatively
persistent, predictive of behavior, and resistant to change
until they are challenged by cogent contrary information
along the dimension or dimensions perceived central to the
merits of the object. However, it is neither adaptive nor possi-
ble for people to exert considerable mental effort in processing
all of the persuasive information to which they are exposed.
This does not mean that people never form attitudes when
motivation and/or ability to think are low; in contrast, they
may change via a peripheral route, wherein relatively simple
associations, online inferences, and well-learned heuristics are
utilized. For example, a person might be more persuaded by a
message containing nine rather than three arguments because
each of the arguments is evaluated and determined to be
compelling (central route), or because the person simply
counts the arguments and reasons and assumes 'the more the
better' (peripheral route). Attitudes formed or changed by
the central route processes are postulated to be relatively
more persistent, resistant, and predictive of long-term behavior
than those based on peripheral route processes. Thus, the ELM
228 Attitude Change
holds that both central and peripheral processes are important
for understanding attitude change, but their influence varies
depending on the likelihood of thinking.
The ELM holds that there are many variables capable of
affecting elaboration and influencing the route to persuasion.
Some variables affect a person's motivation to process issue-
relevant information (e.g., the personal relevance of the issue;
personal accountability for a decision; whether the person is
relatively high or low in need for cognition), whereas others
affect their ability or opportunity to think about a message
(e.g., the presence of distraction; the number of times the
information is repeated; the amount of issue-relevant knowl-
edge available). Some variables affect processing in a relatively
objective manner (e.g., distraction disrupts whatever thoughts
a person is having whether favorable or unfavorable to the
proposal), whereas others influence elaboration in a biased
fashion (e.g., a positive mood makes positive thoughts more
likely than negative thoughts when people are motivated and
able to think). Biases can stem from ability factors (e.g., a
biased knowledge store) and motivational factors (e.g., when
a desire to maintain one's current attitude is more salient than
one's desire to objectively consider new information).
Research on the ELM has shown that when the elaboration
likelihood is high (e.g., high personal relevance, high knowl-
edge of topic, simple message in print, no distractions, etc.),
people typically know that they want and are able to evaluate
the merits of the information presented, and they do so. When
thinking is high, the number, valence, and confidence people
have in their thoughts determine the extent of influence. The
extensive thinking that people do can be relatively objective or
biased. On the other hand, when the elaboration likelihood is
low, people know that they do not want and/or are not able
to carefully evaluate the merits of the information presented
(or they do not even consider exerting effort). Thus, if any
evaluation is formed, it is likely to be the result of relatively
simple associations or inferences (e.g., agreement with an
expert source; counting the number of arguments presented).
When the elaboration likelihood is moderate (e.g., uncertain
personal relevance, moderate knowledge, moderate complex-
ity, etc.), however, people may be unsure as to whether the
message warrants or needs scrutiny, and whether or not they
are capable of providing this analysis. In these situations,
they may examine the persuasion context for indications
(e.g., is the source credible?) as to whether or not they should
attempt to process the message.
There are at least two important implications of the ELM.
First, the model holds that any one variable can produce per-
suasion by different processes in different situations. For exam-
ple, putting people in a positive mood can influence attitudes
because of a simple inference process when the likelihood of
thinking is low (e.g., 'I feel good so I must like if), bias
thinking when the likelihood of thinking is high (i.e., making
positive interpretations more likely than negative ones), and
influence the extent of thinking when it is not already con-
strained to be high or low (e.g., thinking about an unpleasant
message less when happy than when sad). In addition to
considering the extent of thinking, it is also important to
consider the timing of key persuasion variables such as
whether they are salient before or after message processing.
For example, if a person is motivated and able to think about
a message, inducing happiness prior to the message will likely
lead to biased thinking about the message. However, when
happiness is induced right after message exposure, it is likely
to affect the confidence people have in the thoughts that they
have already generated. Thus, by considering the likelihood of
thinking and the timing of persuasion variables, the mecha-
nism by which the variable affects attitudes can be understood.
Second, as explained next, the ELM holds that the mechanism
of persuasion has implications for the strength of the changed
attitude. The ELM holds that not all attitude changes of the
same magnitude are equal. Specifically, thoughtful attitude
changes (central route) tend to be more consequential than
nonthoughtful changes (peripheral route).
Consequences of Attitude Changes Produced by
Different Processes
It is now clear that there are a variety of processes by which
attitudes can be changed, and that the different processes
dominate in different situations. That is, some attitude change
processes dominate when motivation and ability to think are
high, but other change processes dominate when motivation
and ability to think are low. Research suggests that attitudes
formed by different processes often have different characteris-
tics. For example, persistence of persuasion refers to the extent
to which attitude changes endure over time. When attitude
change is based on extensive issue-relevant thinking, it tends
to persist longer than when it is not. However, multiple expo-
sures to positive cues can also produce relatively persistent
attitudes.
Resistance refers to the extent to which attitude change is
capable of surviving an attack from contrary information. Atti-
tudes are more resistant the stronger the attack they can with-
stand. Although attitude persistence and resistance tend to
co-occur, their potential independence is shown conclusively
in William McGuire's work on cultural truisms. Truisms such
as 'you should brush your teeth after every meal' tend to be
highly persistent in a vacuum, but very susceptible to influence
when challenged. People have very little practice in defending
truisms because they have never been attacked. These beliefs
were likely formed with little issue-relevant thinking at a time
during childhood when extensive thinking was relatively
unlikely. Instead, the truisms were probably presented repeat-
edly by powerful, likable, and expert sources. As noted above,
the continual pairing of an attitude with positive cues may
produce a relatively persistent opinion, but it may not prove
resistant when attacked. The resistance of cultural truisms and
other attitudes can be improved by motivating and enabling
people to defend their positions in advance of a challenging
communication. One such inoculation treatment involves ex-
posing people to a few pieces of counterattitudinal informa-
tion prior to the threatening communication and showing
them how to refute it. The inoculation procedure does not
change the valence of a person's initial attitude, but it makes
it stronger. Other persuasion treatments that seem ineffective
in changing the valence of attitudes might nonetheless be
effective in modifying the strength of the attitude, making it
more or less enduring, resistant, or predictive of behavior than
it was initially.
Attitude Change 229
Perhaps the most important consequence of attitudes
changed by high rather than low thought processes is that
these attitudes are more likely to predict and guide behavior.
There are several reasons for this. First, as noted above,
high thought attitudes are more likely to be stable over time
so that they will be available to guide behavior when the
opportunity arises. Second, however, attitudes based on high
amounts of thought are also held with greater confidence,
making people more willing to act on these attitudes. Interest-
ingly, recent research by Barden and Petty shows that if attitude
confidence is increased even in the absence of enhanced think-
ing, people are more willing to act on these attitudes, providing
a relatively low thought way to increase attitude strength.
Summary
In sum, contemporary persuasion theories hold that changes in
attitudes can come about through a variety of processes which
imbue them with a multiplicity of characteristics and render
them capable of inducing a diversity of consequences. Accord-
ing to the popular dual process logic, the processes emphasized
by the central route theories should be largely responsible for
attitude change when a person's motivation and ability to
scrutinize issue-relevant information is high. In contrast, the
peripheral route processes should become more dominant as
either motivation or ability to think is attenuated. This frame-
work allows understanding and prediction of what variables
affect attitudes and in what general situations. It also permits
understanding and prediction of the consequences of attitude
change. It is now accepted that all attitudes can be based on
cognitive, affective, and behavioral information, and that any
one variable can have an impact on persuasion by invoking
different processes in different situations. Finally, attitudes that
appear identical when measured can be quite different in their
underlying basis or structure and thus can be quite different in
their temporal persistence, resistance, or in their ability to pre-
dict behavior. Work on attitude change to the present has
focused on the intrapsychic processes responsible for change
in adult populations mostly in Western cultures. Future
research is needed on the interpersonal processes responsible
for attitude change, and the potentially different mechanisms
that produce change in different population groups (e.g., chil-
dren vs. elderly individuals, those in individualistic vs. collec-
tivist cultures). In addition, as the field matures, current
theories are ripe for exportation to important applied domains
such as health promotion (e.g., AIDS education), political par-
ticipation (e.g., determinants of voter choice), and others.
See also: Persuasion.
Further Reading
Bern DJ (1965) Self-perception theory. In: Berkowitz L (ed.) Advances in Experimental
Social Psychology, vol. 6. New York: Academic Press.
Brinol P and Petty RE (2009) Persuasion: Insights from the self-validation hypothesis.
In: Zanna MP (ed.) Advances in Experimental Social Psychology, vol. 41 ,
pp. 69-118. New York: Elsevier.
Chaiken S and Trope Y (eds.) (1999) Dual Process Models in Social Psychology.
New York: Guilford.
Eagly AH and Chaiken S (1993) The Psychology of Attitudes. New York: Harcourt Brace.
Fishbein M and Ajzen I (1975) Belief, Attitude, Intention, and Behavior: An Introduction
to Theory and Research. Reading, MA: Addison-Wesley.
Harmon-Jones E and Mills J (eds.) (1 999) Cognitive Dissonance: Progress on a Pivotal
Theory in Social Psychology. Washington, DC: American Psychological
Association.
McGuire WJ (1985) Attitudes and attitude change. In: Lindzey G and Aronson E (eds.)
Handbook of Social Psychology, 3rd edn. New York: Random House.
Petty RE and Cacioppo JT (1 981 ) Attitudes and Persuasion: Classic and Contemporary
Approaches. Dubuque, IA: Wm. C. Brown.
Petty RE and Cacioppo JT (1986) The elaboration likelihood model of persuasion.
In: Berkowitz L (ed.) Advances in Experimental Social Psychology, vol. 19,
pp. 123-205. New York: Academic Press.
Petty RE, Fazio RH, and Brinol P (eds.) (2009) Attitudes: Insight from the New
Implicit Measures. New York: Psychology Press.
Petty RE and Krosnick JA (eds.) (1 995) Attitude Strength: Antecedents and
Consequences. Mahwah, NJ: Erlbaum.
Petty RE, Ostrom TM, and Brock TC (eds.) (1 981 ) Cognitive Responses in Persuasion.
Hillsdale, NJ: Erlbaum.
Relevant Websites
Changingminds.org - Changing Minds.
http://en.wikipedia.org/wiki/Attitude_change - Wikipedia.
Attitude Formation
M A Olson and R V Kendrick, University of Tennessee, Knoxville, TN, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Attitude Positive and negative evaluations of things, people,
and concepts.
Implicit cognition Cognitive processes that operate
efficiently and involuntarily, and often without the
awareness of the perceiver.
Priming Happens when exposure to a stimulus influences
responses to another stimulus, often in the absence of
awareness.
Tripartite model A method of categorizing different attitude
formation processes into affect (emotions), cognitions
(beliefs), and information about attitude-related behavior.
Introduction and Relevant Distinctions
Attitudes are positive and negative evaluations of things,
people, and concepts. A person might, for example, feel posi-
tively toward his or her mother, value his or her car for
its reliability, be suspicious of foreigners, and constantly crave
chocolate. All of these are attitudes in that they imply positive/
approach tendencies or negative/avoidance tendencies and
involve summaries of the value of the object. Such precom-
puted summary judgments of attitude objects are functional in
that they help prepare the individual for action so that one
does not need to deliberately make important decisions from
scratch every time a behavioral opportunity arises. In other
words, because people know what they like and dislike, they
can spend less time pondering what to buy and consume, for
whom to vote, how to behave, and with whom to affdiate.
They need only consult their attitudes toward the relevant
object and act in accordance with it. Indeed, attitudes have
been considered 'ready aids' for sizing up the world and how
to live in it, and it is difficult to imagine how people could
function without having a grasp of the things that benefit
and sustain them and things that could potentially hinder
or harm them.
The attitudes people hold toward the objects in their world
have immense personal and social consequences. Attitudes
have been implicated in health behavior (e.g., attitudes toward
regular physicals or safe sex practices), politics (e.g., attitudes
toward general political ideologies or more specific political
figures), consumer behavior (e.g., attitudes toward various
brands and products), and social conflict (e.g., attitudes toward
various religious groups and ethnicities), among many other
domains. Because attitudes are so relevant to such an array of
life domains, and because they are thought to be important
determinants of behaviors and judgments (at least under
certain conditions), the attitude concept is one of the most
well-studied concepts in the field of psychology. Hence, an
understanding of the basic origins of attitudes is critical to
understanding human psychology in a more general sense.
For most people, attitudes' origins are often understood
on a per-case basis, whereby one simply attempts to identify
the source of each attitude individually (e.g., 'I learned to like
blue cheese from my mother,' or 'I like Volvo cars because of
their safety record'), without considering broader themes.
Social scientists consider such broader themes and, in so
doing, have provided some ways to organize attitude forma-
tion processes. The first and most prevalent is referred to as
the tripartite approach, whereby attitudes' origins are consid-
ered in terms of affect (based on affect or emotion), cognition
(based on belief or cognition), behavior (based on informa-
tion derived from one's own behavior), or some combination
of the three. More recently, researchers have distinguished
between explicit origins, where attitudes are derived con-
sciously through some sort of thoughtful and effortful process,
and implicit origins, where attitudes form without conscious
intent or awareness on the part of the person. In this article,
attitude formation processes are discussed in terms of their
tripartite and implicit versus explicit origins, with each being
discussed in turn.
A final distinction is whether a given attitude is learned
(either socially or through direct experience) or biologically
derived (either genetically or through evolutionary processes).
Although most attitudes of interest to social scientists are
typically assumed to be learned, there is evidence that some
attitudes are meaningfully influenced by biological processes
relating to genetics and evolution. This article describes the
main sources of attitudes as they have been studied by social
scientists and considers some additional issues when ponder-
ing about the origins of attitudes.
Origins
Cognitive Origins
One way an individual may come to evaluate a particular
object is through acquiring positive or negative beliefs about
the object. One may read information outlining the significant,
harmful health effects of smoking and consequently develop
beliefs that cigarettes lead to negative outcomes such as lung
cancer and emphysema. Attitudes can develop through this
often thoughtful, 'rational' route as individuals acquire cogni-
tions that inform them whether an attitude object leads to good
or bad outcomes or possesses wanted or unwanted qualities.
The most prominent framework within which to under-
stand the cognitive origins of attitudes is the expectancy-value
model. The model claims that an attitude is a function of
the expectancy that the object has certain attributes and the
value attached to each of the perceived attributes of the atti-
tude object. The expectancy is one's belief about whether the
230
Attitude Formation 231
attribute will occur, and the value represents one's evaluation of
the attribute. For example, in forming an attitude toward
smoking, one might think of various attributes of smoking
cigarettes (that they cause cancer, make one appear cool),
determine how (un)favorable they find each attribute, and
then assess the likelihood that each will occur. I might believe
that smoking cigarettes will make me look cool and that this
outcome is quite good (high positive value) but is only mar-
ginally likely to occur (low expectancy). Additionally, I am
keenly aware of adverts stating that smoking cigarettes leads
to an increased risk of lung cancer, and I conclude that this is
clearly a negative outcome (high negative value) that according
to published statistics is likely to occur (high expectancy).
These values and expectancies are then combined, yielding an
overall attitude toward the object. In the aforementioned
example, I really want to look cool but I do not want to get
lung cancer. Since looking cool is quite positive yet not likely,
and getting cancer is quite negative as well as likely, the net
sum of my attitude should be negative. As one might imagine,
this model represents a conscious, explicit process, requiring a
certain degree of attention and deliberate effort.
Similar to the expectancy-value model is information inte-
gration theory. This model states that as new object-relevant
information is learned, it gets added to existing thoughts and
beliefs about the object, and the result of this integration
produces an attitude. At first, each piece of new information
is added in a way similar to the expectancy-value model. How-
ever, in this process, the potential (un)favorability of each new
attribute is evaluated, and this evaluation is combined with a
judgment about the importance of the attribute in forming the
attitude. Once we determine how we think about the attributes
of an attitude object and how important each attribute is in
determining the attitude, they can then be integrated with
current attitudes and beliefs to form new attitudes. For exam-
ple, I may learn that smoking leads to a risk of developing
emphysema, an unfavorable outcome, which I deem as unim-
portant (perhaps because this consequence occurs in the distant
future). However, ifl already possess thoughts that smoking will
make me feel relaxed - a likeable outcome that I place more
value on because it is so immediate - then it will comprise
a larger portion of my complete attitude toward smoking.
This last point touches on an important distinction between
the value-expectancy model and the information integration
model. Information integration theory acknowledges that
when forming attitudes toward objects, not all pieces of infor-
mation are entered into the equation equally. Rather, some
beliefs are learned earlier, or may be seen as more important,
and therefore have more influence on information learned later.
Among other explicit, belief-based approaches to attitude
formation are the reception-yielding model and the cognitive-
response model. Both of these models are less concerned with
calculations in their descriptions of how attitudes form and
focus rather on the explicit reactions to people's thoughts
about what they learn about attitude objects. The reception-
yielding model describes a two-stage process where one initi-
ally receives object-relevant information and then chooses to
either yield to it (i.e., change one's evaluation of the object in
light of the new information) or reject it (i.e., maintain one's
evaluation of the object despite the new information). Consid-
ering the previous smoking example, when deciding how I feel
about smoking, I might read the surgeon general's warning
about birth defects caused by smoking while pregnant (recep-
tion) and then adjust my attitude toward smoking due to this
potential health risk (yield). Although this model has been
primarily applied to attitude change, it can be used to describe
how people perceive information and form attitudes from
scratch based on their own thoughts about what they learn.
Thus, this approach helps us understand the origins of atti-
tudes based not simply on what people learn about an object
but how they respond to what they learn.
The cognitive-response model presents another process by
which attitudes form in an active, deliberate, and explicit man-
ner. When we encounter an attitude object, we tend to actively
relate qualities and information about the object to existing
thoughts. The process of relating new bits of information to
those already held results in cognitive responses concerning
the attitude object, and the extent to which these responses are
(un)favorable determines our overall evaluation of the object.
When I leam that cigarettes are expensive, for example, I relate
this new, negatively evaluated thought to my preexisting
thought that cigarettes also make one smell bad. Taken together,
I conclude that there are more negative aspects of smoking
(cost and interpersonal consequences) than there are positive
qualities (it feels pleasurable), and I form an overall negative
attitude toward smoking.
A more recent development in the literature on the cogni-
tive origins of attitudes is the metacognitive model (MCM).
According to the MCM, one considers the various qualities of
the object, but also considers the truth value of each of his or
her beliefs about the object. One might, for example, consider
some beliefs about the object to be more certain than others.
Therefore, it is possible to have contradictory cognitions about
an object, with some cognitions labeled as valid (presumably
allowing these attributes to exert greater influence) and others
labeled invalid (presumably not influencing one's overall atti-
tude). Interestingly, the application of these so-called validity
tags requires some amount of effort, and in the absence of
such effort (as when one has little time or energy to think),
all cognitions are available to influence one's attitude, even
those that one might not believe to be true. I may possess the
negative cognition that cigarettes cause cancer, but I supple-
ment this belief with an invalidity tag (i.e., 'this is not applica-
ble because it only applies to long-term smokers and I have
only been smoking recently and plan to quit'), so when asked,
I might claim to have a positive attitude toward smoking,
even though I still harbor some negative beliefs about it.
The aforementioned descriptions of cognitive routes to
attitude formation all share a relative reliance on thoughtful
conscious processes. It is worth noting that cognitive origins
are called so because they have at their core cognition or beliefs
about the attitude object. This fact, while implying that some
minimum of explicit, deliberate effort be present during the
formation process, still leaves open the possibility that attitudes
can form through implicit, less-conscious forms of learning.
Unconscious thought theory, for example, argues that one can
process large amounts of information outside of conscious
awareness, compared with the relatively small amount of infor-
mation that can be consciously processed at one time. Uncon-
scious thought has been shown to give individuals the ability
to process large amounts of information, perhaps allowing for
232 Attitude Formation
attitudes to form by consolidating cognitions even in the
absence of any deliberate or conscious intentions. When bom-
barded with complex information about an object, it often
becomes difficult to weigh the pros and cons using the more
explicit manners discussed previously. Unconscious thought
provides the possibility of taking large amounts of object-
relevant information and, in essence, letting the powerful,
unconscious parts of the brain do the work, yielding an overall
evaluation of the attitude object that is presented to conscious-
ness later as a gut feeling.
Lastly, the value-account model outlines another cognitive
yet implicit process by which attitudes have been shown to
arise. Overall evaluation of objects attitudes can be character-
ized as value accounts comprising the positive or negative
values associated with the various attributes of the object.
When new information is obtained, it gets incorporated into
the account, and the value, or evaluation, of the account linked
to the object changes the overall value or attitude toward the
object. This process has been shown to function even outside
of deliberate control, such that individuals who are engaged in
another task designed to deprive them of cognitive resources
are still able to detect valence information concerning an atti-
tude object, and this information has been shown to influence
overall evaluations of the attitude objects. Applied practically,
if I have an overall positive attitude toward cigarette smoking
(a positive value) and am exposed repeatedly to antismoking
billboards describing the number of deaths caused every
year by smoking, my attitude should become more negative
due to the incorporation of negative-valued information, even
though the bulk of my attention is fixed on the task of driving
and not on deliberately processing public health bulletins.
Affective Origins
In addition to thoughts, attitudes can also form from the
emotional responses that we experience when we encounter
an attitude object. For example, I may experience a pleasurable
feeling while smoking a cigarette, and this affect could lead me
to have a positive attitude toward smoking. Or, I might think
about the negative health effects of long-term smoking such
as lung cancer or death, and I experience a distinctly negative
attitude instead. These examples highlight several ways that
affect can be used to form attitudes toward objects. It is
worth noting that the aforementioned first example derives
solely from an emotional response, but the second example
is based on an emotional experience felt in response to a
cognition concerning the attitude object. This highlights the
fact that affective origins can be directly experienced or derived
secondarily from an object quality. In this section, we focus on
affect, independent of whether it is in response to a thought
about the object. Later, we address how affective responses
lead to attitude formation and then we consider three pro-
cesses that lead to affectively derived attitudes: operant condi-
tioning, evaluative (a form of classical) conditioning, and mere
exposure. Operant conditioning, a process by which the fre-
quency of a response is increased following a positive outcome
and decreased following a negative outcome, can provide one
mechanism for the formation of affectively based attitudes.
If interaction with the attitude object causes a positive emo-
tional response, then the evaluation of the attitude object will
become more positive and this connection will strengthen,
resulting in a positive general attitude. Research has shown
that individuals who were reinforced when recalling favorable
responses toward an object (via hearing 'good' in response
to these recollections) later demonstrated more positive atti-
tudes toward the object, compared to unreinforced partici-
pants. If every time I smoke a cigarette and I am reinforced
with a positive feeling from nicotine, the act of smoking will be
reinforced, resulting in a stronger positive attitude toward
smoking. Conversely, if every time I light a cigarette everyone
leaves the room, the likelihood of my smoking will decrease
and I will develop a negative attitude toward it. Thus, attitudes
can be formed by experiencing positive or negative outcomes
based on our attitudinal responses, and the more often one's
responses are reinforced, the more likely that attitude will
manifest itself in the future.
Operant conditioning tends to occur in an explicit fashion
with one consciously perceiving the reinforcing outcomes of
relevant behavior. Classical conditioning, while similar to
operant conditioning, does not require that the person respond
to an attitude object. Instead, the connection between an object
and an affective evaluation develops by simply observing the
pairings between an attitude object and a positively or nega-
tively evaluated stimulus. Observing these pairings creates an
association between the previously neutral object and the
valence of the object with which it is paired. For example,
researchers created positive and negative attitudes toward
two names ('Ed' and 'George') by visually presenting these
names to participants on index cards while having them
repeat aloud positive or negative adjectives that were paired
consistently with either name. Importantly, this occurred even
though there was no meaningful relationship between the
names and the adjectives. When considering our smoking
example, repeatedly noticing advertisements containing attrac-
tive models engaged in smoking behavior can lead to a change
in one's attitude toward smoking due to the transfer of the
positive affect associated with the attractive model to the ciga-
rette brand. This can occur in an explicit fashion where one is
consciously aware of the pairings of the attitude object (cigar-
ettes) and the valences stimuli (attractive model).
Research on a variant of classical conditioning - evaluative
conditioning - has shown that attitudes can form via affective
connections between an object and another, valenced object,
without conscious awareness of this relationship. Like classical
conditioning, evaluative conditioning happens when a previ-
ously neutral object takes on the valence of stimuli that co-
occur with the previously neutral object. Studies have shown
that participants who view a novel object repeatedly paired
with positive stimuli will later evaluate the novel object more
positively, whereas novel objects repeatedly paired with nega-
tive stimuli will come to be viewed more negatively. There has
been some debate concerning whether this process of attitude
formation occurs in an explicit fashion where awareness of the
connection between attitude objects and valenced associates is
necessary, or whether it can occur implicitly, in the absence of
such awareness. Recent work seems to indicate that it depends.
Explicit awareness of the repeated pairings of neutral objects
with strongly positive or negative stimuli leads an individual
to form consciously held beliefs concerning the relationship
between the two. However, if one remains unaware of the
Attitude Formation 233
repeated contingencies between a neutral object and a valenced
stimuli, then the affect from the valenced stimuli can be implic-
itly misattributed to the previously neutral object, resulting in
an attitude, the origins of which remain obscured to the atti-
tude holder. If every time I sit next to an ashtray, my sense of
smell is insulted by the foul stench of old cigarette smoke, I am
likely to associate smoking with the negative feelings elicited
by the smell of smoke, and whether I become conscious of the
association between the negative affect and smoking will deter-
mine whether this attitude stems from a propositional, explicit
process or an implicit, misattributional process. Research on
the 'mere exposure' effect has also shown that simply increas-
ing the familiarity of an attitude object can create more positive
attitudes toward the object. In a study that presented Turkish
words or Chinese symbols to participants in increasing fre-
quency, those who observed more repetitions of the words or
symbols showed a more favorable attitude toward them com-
pared to others who saw the stimuli infrequently. This is due to
a phenomenon known as 'perceptual fluency,' that is, greater
ease at perceiving an object upon subsequent encounters,
which is interpreted as liking for the object. Thus, the more
familiar one is with the attitude object, the more likely one is to
have a positive evaluation of it. So, if I choose to smoke, in the
absence of other considerations, I am likely to prefer a brand of
cigarettes that I have been exposed to in the past more than a
brand I have never before encountered, even if I am unable to
consciously recall which brand I have seen more often in
adverts. Similar to the conditioning research described previ-
ously, mere exposure research outlines a way attitudes can
form due to affect alone, without reliance on cognitions
regarding an object's attributes, and often in the absence of
explicit awareness of the underlying process.
Behavioral Origins
In situations where we lack either a cognitive or affective basis
for an attitude, we can infer an attitude by observing our past
behavior toward the object in a process of self-perception.
When the internal cues that might otherwise normally lead to
an attitude are weak, we instead make inferences regarding our
behavior much like an outside observer would make. Consider
our earlier example. I may not have any solid beliefs about the
benefits or hazards of smoking, nor does the idea of smoking
elicit any salient emotion. However, I may recall that I have
smoked in the past, and I therefore conclude that I must have a
positive attitude toward smoking - why else would I have
behaved so? This scenario exemplifies how attitudes can have
a behavioral genesis whereby our attitudes fall in line with our
self-perceived actions.
Research on cognitive dissonance provides a framework to
understand how attitudes can stem from past behaviors toward
an attitude object when one has preexisting evaluative knowl-
edge of the object. Cognitive dissonance occurs when an indi-
vidual experiences an inconsistency between thoughts and
behaviors concerning a target object. This state of dissonance
is characterized by a negative state of arousal, which the indi-
vidual seeks to correct by restoring consonance between the
attitude and the behavior. If my behaviors include smoking
cigarettes and I possess cognitions that cigarettes can lead to
cancer, I would be susceptible to the negative affect created by
the conflict between my thoughts and behavior. To restore
consistency, I can either quit smoking, thus realigning my
thoughts and behaviors, or change my cognitions concerning
the health risks of smoking (e.g., arguing that only heavy
smokers will develop cancer) so that the cognitive component
of my attitude toward smoking is no longer in conflict with my
actions related to smoking. Thus, the behavior informs the
attitude.
Similar research has shown that we are often noncon-
sciously tuned into subtle bodily gestures that inform us as to
how we feel about an object. In a seminal study on the behav-
ioral antecedents of attitude formation, participants held pen-
cils in their mouths while viewing cartoons and were asked to
rate how funny they felt the cartoons were. Some participants
held the pencil with their lips, simulating a frown, while others
held it with their teeth, simulating a smile. Participants who
exhibited a smile while viewing the cartoons rated them as
funnier than those who frowned. In a related research, partici-
pants induced to nod their heads while perceiving social sti-
muli rated those stimuli more positively than stimuli perceived
while they were made to shake their head from side to side.
Thus, in the absence of an obvious external reason for behav-
ior, we infer an internal one, and this attribution can result in a
new attitude that is consistent with our actions.
Self-perception, and its effects on attitude formation from
our overt behavior, can have the opposite effect if it is applied
when one already possesses an attitude. Researchers led pre-
schoolers to engage in an enjoyable activity (drawing with
magic markers) and promised some children a reward for parti-
cipating in the activity while providing no reward for other
children. When later given the chance to engage in the activ-
ity, children given the reward spent only half as much time
coloring with the markers as did children who had not been
given a reward. The explanation rests on an overjustification
effect, where children who had played while anticipating
a reward must have perceived that they did so because of the
reward, whereas the others perceived that they engaged in
the behavior simply because they liked it.
Biological Origins
Many social attitudes (e.g., racial, political, and interpersonal
attitudes) probably are learned, as shown by the wide and
easily observed varieties of attitudes in such areas. How-
ever, universal attitudes that show little cultural variation
(e.g., xenophobia or a preference for sweet foods) can also be
observed, suggesting biological origins. Moreover, cultural var-
iation in a given attitude does not itself rule out a role for
biological origins, as biological processes must interact with
experience to produce a given attitude or behavior. Interest in
the biological origins of attitudes has waxed and waned in the
social sciences, and those advocating an evolutionary perspec-
tive have been attacked by critics for a variety of reasons, not
the least of which is the notion that evidence for biological
origins of attitudes might be taken as evidence of 'fixedness'
(the impossibility of change). Others have voiced concern
that evidence of biological origins might be misconstrued as
justification for the existence of a given attitude or trait
(the 'what is natural must be good' argument, also known as
the adaptationist fallacy).
234 Attitude Formation
Evolutionary supporters tend to focus on mate preferences
as examples of evolved attitudes. For example, owing to their
greater costs to produce offspring, women are thought to be
'choosier' in selecting a mate and prefer a mate who shows
evidence of both willingness and ability to invest in the health
of the woman and her offspring. Men, however, are more likely
to prefer a greater number of mates and be less careful about
their mate choices, because their investment in offspring pro-
duction is less. Moreover, men prefer potential mates with
outward indicators of fertility, such as a waist-hip ratio of
about 0 . 80 , which has been shown to lead to more successful
offspring production. Cross-cultural data on attitudes toward
potential marriage partners consistently support this reasoning.
Flexibility in mate preferences is in line with an evolution-
ary perspective as well, as women will tend to show an increased
preference for mates of high genetic value (e.g., more attractive
men), as well as more aggressive and competitive mates, during
the most fertile points of the menstrual cycle. Preferences for a
mate who is willing and able to provide resources do not vary
as a function of the menstrual cycle.
Little progress has been made in addressing more social
scientific-oriented attitudes (e.g., racial and political attitudes)
from an evolutionary perspective. There are some exceptions,
however, including some research suggesting that racial preju-
dice may be a natural extension of an evolved preference for
people who appear more similar to oneself. Another simple
extension of evolutionary approaches to attitude research
involves food preferences. Mammals, which acquire immedi-
ate energy from the ingestion of sugars and starches and long-
term storable energy from fats, have likely evolved a taste for
both. Hence, positive evaluations of foods high in sugars and
fats are likely to have evolved components.
Other biological approaches to the study of attitude forma-
tion come from behavioral genetics research. Here, studies of
identical twins who were either reared in similar or different
environments have come to show that some attitudes are
linked to genetics. For example, attitudes toward jazz, the
death penalty, and some political ideologies have been shown
to have genetic components, whereas attitudes toward coedu-
cation have not. Attitudes may have heritable components
because of their links to other traits with clear evidence for
heritability (e.g., intelligence, openness). For example, because
intelligence is somewhat genetically determined, attitudes
related to intelligence may be at least partially determined by
genes. Because a preference for jazz music is associated with
intelligence, and intelligence has genetic components, a prefer-
ence for jazz is indirectly linked to genetics. Similar arguments
can be made about other biologically determined traits such as
physique and preferences for various physical activities. Indeed,
recent twin research suggests that one of the more studied social
attitudes - prejudice - may have genetic components.
Regarding their implicit versus explicit origins, a preexisting
biologically based tendency toward a given set of attitudes may
suggest a more implicit basis. Indeed, people are seldom able
to explain why they prefer sweet and fatty foods, or members of
their own group, without referring back to their attitude in a
circular fashion. However, people are certainly aware of their
food preferences, and most social scientists would agree that
they are also aware of their prejudices, whether they wish to
admit them publicly or not. Thus, we must consider what
we mean by 'implicit.' Here, researchers typically refer to an
attitude as implicit if one is unaware of its origins, its influ-
ence on judgments and behavior, or both. Attitudes partially
determined by biology probably meet these criteria, as people
have a difficult time generating the underlying reason - the
foundation - for their food preferences, preferences for their
own groups, etc. Similarly, it was mentioned that men tend to
prefer a certain waist-hip ratio in women, but most would be
unable to explain the evolutionary origins of their preference
for more shapely feminine forms. Hence, biologically deter-
mined attitudes are liable to have implicit origins, although
many of them are certainly further molded by culture and
personal experience.
Additional Considerations
The question of how attitudes form has been thoroughly inves-
tigated for several decades by social scientists, but it is difficult
to isolate this origins question from several other relevant
issues. Thus, the final section discusses additional concepts
related to attitude formation in the hopes of providing a
broader context.
First, we have covered the different attitude formation pro-
cesses individually, but we must also point out that such
processes are likely to combine and interact in real-world
attitude formation situations, and some research has investi-
gated these more complex situations. It is also likely that the
different formation processes are mutually reinforcing in natu-
ral environments, such as when one uses one's affective reac-
tions to an object as cause to seek positive cognitively oriented
information about it. Similarly, one's biological predisposi-
tions to prefer sweet foods are often reinforced by cultural
information that encourages the consumption of sugary
foods. However, interesting situations arise when two or more
formation processes conflict. We have already mentioned the
case of cognitive dissonance, where knowledge of one's behav-
ior contradicts either other beliefs or other affective reactions
regarding an object. Conflict also arises when affective and cog-
nitive reactions disagree, such as when one feels warmly toward
an object he or she believes has negative effects (e.g., smoking
or an unhealthy relationship partner). Such situations are
usually unstable and prompt discomfort in the perceiver that
he or she is motivated to reduce.
Attitudes are also related to other social constructs studied
by social scientists, including morals, values, impression for-
mation, and social judgment. Morals and values have been
considered classes of attitudes that relate to interpersonal con-
duct. Impression formation is a bit broader than attitude in
that it involves not only the question of whether we come to
like someone, but also whether we think of the individual
as having specific traits (e.g., introversion, extroversion, etc.).
Social judgments are made at a specific point in time, whereas
attitudes, potentially among other information, are brought to
bear on a specific case, such as when someone judges a partic-
ular politician as qualified to hold public office. Judgments are
single cognitive acts, whereas attitudes are thought to have
some longevity in the person's mind. Thus, answers to ques-
tions about the origins of attitudes can also be found in these
literatures.
Attitude Formation 235
Although this article is about attitude formation, another
article in this volume is directly relevant. Although the con-
cepts differ in that the former addresses how attitudes come
about without the existence of a prior attitude and the latter
addresses how one's attitudinal position can change over time,
in reality, the research paradigms and theories used to describe
these processes differ only trivially. Indeed, most persuasion
research involves cases where no attitude existed prior to expo-
sure to a persuasive message. Hence, most persuasion research
directly informs attitude formation research.
Attitudes also vary in important ways, and attitude forma-
tion processes have implications for these other attitudinal
qualities. For example, attitudes vary in strength, that is, their
persistence over time, their ability to predict behavior, and
their resistance to counter-persuasion attempts. Stronger atti-
tudes are more likely to form from direct experience with the
attitude object as opposed to hearsay or indirect experience.
Affectively based attitudes also tend to be better predictors of
behavior in many domains, particularly regarding political
attitudes and voting behavior. However, evidence from persua-
sion research indicates that attitudes formed through more
thoughtful processes are often stronger. There is also some
evidence that attitudes of a biological origin are less malleable.
An attitude's ability to predict behavior is part of its appeal
to psychologists, but one should not assume that attitudes
should always predict behavior. Specific theories describe
when attitudes are likely to predict behavior, and some argue
that an attitude's origin must be taken into account when
considering whether it might predict behavior. For example,
because people are less able to articulate the bases of implicitly
formed attitudes, they are less likely to notice their influence.
Implicit attitudes are then likely to leak out into one's behav-
ior without one's awareness. For example, racial prejudice,
if formed gradually through socialization processes that one
rarely acknowledges, is likely to appear in one's less-controllable
behavior. For example, in one study, Black participants were
able to detect racial prejudice in Whites' nonverbal behavior.
Controllable behavior is less likely to be influenced by implic-
itly formed attitudes and is instead more influenced by explicitly
held attitudes.
In conclusion, fundamental goals of the science of psychol-
ogy include the ability to explain and predict human behavior.
Because attitudes are so central to achieving these goals, atti-
tudes hold a central place in psychology. Further, an under-
standing of an attitude's origins is useful in understanding
its later influences on behavior. This article has addressed
several formation processes - affective, cognitive, behavioral,
and biological, as well as implicit and explicit. Although the
power of each of these formation processes is appreciated by
social scientists generally, there is still considerable debate
about the role of each (e.g., whether implicit or explicit forma-
tion processes are more pervasive). The further reading section
provides examples of such debates, as well as more thorough
treatments of attitude formation processes in general.
See also: Impression Formation; Persuasion; Prejudice,
Discrimination, and Stereotypes (Racial Bias).
Further Reading
Eagly AH and Chaiken S (1 993) The Psychology ol Attitudes. Fort Worth, TX: Harcourt
Brace Jovanovich.
Fishbein M and Middlestadt S (1995) Noncognitive effects on attitude formation and
change: Fact or artifact. Journal of Consumer Psychology 4: 181-202.
Gawronski B and Bodenhausen GV (2006) Associative and propositional processes in
evaluation: An integrative review of implicit and explicit attitude change.
Psychological Bulletin 132: 692-731.
Lewis GJ and Bates TC (2010) Genetic evidence for multiple biological mechanisms
underlying in-group favoritism. Psychological Science 21(11): 1623-1628.
Olson MA and Fazio RH (2001) Implicit attitude formation through classical
conditioning. Psychological Science 12: 413-417.
Olson MA and Kendrick RV (2008) Origins of attitudes. In: Crano W and Prislin R (eds.)
Attitudes and Persuasion. New York: Psychology Press.
Olson JM, Vernon PA, Harris JA, and Jang KL (2001) The heritability of attitudes:
A study of twins. Journal ol Personality and Social Psychology 80: 845-860.
Petty RE and Brinol P (2010) Attitude change. In: Baumeister RF and Finkel EJ (eds.)
Advanced Social Psychology: The State of the Science , pp. 217-259. Oxford:
Oxford University Press.
Rudman LA (2004) Sources of implicit attitude. Current Directions in Psychological
Science 13: 80-83.
Towles-Schwen T and Fazio RH (2001) On the origins of racial attitudes: Correlates of
childhood experiences. Personality and Social Psychology Bulletin 27: 162-175.
Wilson TD, Lindsey S, and Schooler TY (2000) A model of dual attitudes. Psychological
Review W: 101-126.
Zajonc RB (1980) Feeling and thinking: Preferences need no inferences. The American
Psychologist 35: 151-175.
Attribution
S J Tobin, University of Queensland, St Lucia, QLD, Australia
© 2012 Elsevier Inc. All rights reserved.
Glossary
Actor-observer effect Attributing other people's behavior to
dispositions, but one's own similar behavior to the situation.
Attributional complexity Highly developed attributional
knowledge structures.
Causal attribution An inference about why an event or
behavior occurred.
Causal uncertainty Doubt about one's abilities in the area
of attribution.
Correspondence bias Attributing an event that can be
explained by the situation to dispositions.
Dispositional attribution Something about the person
caused the behavior.
Implicit theories Beliefs about the fixed or malleable nature
of attributes.
Self-handicapping Creating an obstacle that makes it
difficult to attribute failure to low ability.
Self-serving attributions Attributing one's
successes to dispositions and failures to situational
factors.
Situational attribution Something about the person's
environment caused the behavior.
Introduction
Causal attribution involves identifying one or more factors as
being responsible for bringing about a particular behavior or
outcome. In short, it is the answer to the question 'why?'
Research has found that this type of thinking can occur sponta-
neously when we observe another person's behavior. When we
overhear a person shouting at a coworker, we may wonder why
the person is shouting: Is it because he or she is just a rude
person or was the person provoked somehow? Attributions are
especially common when we are confronted with an unexpected
or negative event. If a spouse suddenly announced that he or she
wanted a divorce, the partner would want to know why. If a
student failed a test, he or she would want to know why.
Note that there is no single, overarching theory of attribu-
tion. Rather, the literature about attribution consists of com-
plementary theories that approach different aspects of the
attribution process. The classical theories identify different
types of attributions and specify the conditions under which
each type should be drawn. Researchers then discovered that
people often exhibit various attributional biases. They began to
devote their attention to explaining why these biases occur.
Contemporary theories focus on the cognitive processes that
underlie the attribution process, differentiating the steps that
occur automatically from those that take more effort. Increas-
ing attention to individual and cultural differences in attribu-
tion has shed light on how basic beliefs about people and
various motivations can influence the attributions people
make. And lastly, a final set of theories illustrates the implica-
tions that different types of attributions have for motivation,
well-being, academic performance, and marital satisfaction.
Classical Theories
Heider (1958)
Heider argued that we can learn a lot by examining lay people's
understanding of the causes of action. It is their understanding,
right or wrong, that will influence their thoughts, feelings, and
behavior. He also argued that causal understanding plays an
important role in life. When people attribute transient events
to stable underlying factors, they feel that they can predict and
control their outcomes.
Heider distinguished between two sets of causes: person
and environment. One type of person factor is power, which
includes ability, self-confidence, mood, and fatigue. Another
person factor is trying, which includes intentions and exertion.
Environmental factors include task difficulty, luck, and opportu-
nity. Of primary importance, according to Heider, is understand-
ing the causes of intentional behavior. Intentional behavior
occurs when a person consciously or unconsciously tries to
bring about a particular consequence. When we can attribute
such intentional behavior to stable underlying factors, such as
ability or task difficulty, we gain a sense of prediction and control
over future outcomes.
How do we know when it is appropriate to attribute an
outcome to ability? The level of exertion can be used as a cue.
Greater exertion is necessary when power is low or difficulty is
high. To keep up with a professional marathon runner, for
example, the average person would have to exert a great deal
of effort. Thus, when perceivers observe a person successfully
perform a difficult task with great ease, they might infer that
the person has a high level of ability.
Heider's analysis of attribution sparked a great deal of
interest. The task for subsequent researchers was to translate
Heider's ideas into hypotheses that could be tested in the
laboratory.
Jones and Davis (1965)
Jones and Davis focused on how perceivers infer intentions,
and corresponding dispositions, from a person's choices.
They argued that perceivers consider the effects of an action
and try to figure out which effects were intended (rather than
simply incidental or accidental). Perceivers also consider
whether the actor was capable of bringing about the effects.
236
Attribution 237
If they conclude that one or more effect was intentionally
produced, they then try to figure out what it tells them
about the person. They compare the chosen action with
other possible unchosen actions. Perceivers attend to non-
common effects, or the unique attributes of a chosen alterna-
tive. For example, imagine a person choosing from a set of
computers or job candidates. If all computers in the set are of
the same brand, or if all of the job candidates have expertise in
the same area, these factors cannot help a perceiver under-
stand why a person chose one over another. But if the chosen
computer is the only one of a particular brand (e.g., the only
Mac), or if the chosen candidate is the only one with a
particular type of expertise (e.g., the only health psycholo-
gist), this factor can potentially tell perceivers something
about the person who made the selection.
The social desirability of the effects, or how desirable a
particular feature is, also matters. If a choice has several unique
attributes, people typically assume that the factor attracting the
person to that option was the more desirable one. For example,
if the computer chosen was the only Mac and the most expen-
sive, people would assume that the person chose the computer
because it was a Mac, not because it was the most expensive. In
terms of what one learns about the target, however, more
unusual choices and behaviors are more diagnostic. If a person
does what anyone else would have done, perceivers do not
learn much. But if a target's behavior diverges from that of the
average person, perceivers are more likely to draw an inference
about the correspondent's disposition.
Kelley (1967)
Kelley's model describes how perceivers analyze behavioral
information; typically about a person interacting with a social
or nonsocial stimulus. Perceivers consider whether the person's
reaction is similar when compared to different people (consen-
sus), stimuli (distinctiveness), and occasions (consistency).
Depending upon the information they gather or infer, percei-
vers may then attribute the reaction to something stable about
the person or the entity.
As an example, perceivers may learn that Jennifer likes her
English professor. They would then analyze whether the effect
(liking) covaries with the person (Jennifer), the stimulus
(the English professor), or the particular occasion. Do other
people like the English professor? Does Jennifer like her
other professors? Has Jennifer always liked this professor?
When the cause is present, the effect should be present; when
the cause is absent, the effect should be absent. If we find
that other people do not like the English professor (low
consensus), Jennifer likes her other professors (low distinc-
tiveness), and Jennifer has always liked the English professor
(high consistency), we would draw a dispositional inference:
something about Jennifer is causing her to like the English
professor. However, if we find that other people like the
English professor (high consensus), Jennifer does not like her
other professors (high distinctiveness), and Jennifer has always
liked the English professor (high consistency), we would
draw an external attribution: something about the English
professor is causing Jennifer to like him or her. Other potential
outcomes would lead to attributions of some combination of
the person, the stimulus, and/or the particular circumstances.
Abnormal conditions focus model
An alternative way of understanding the role of consensus,
distinctiveness, and consistency information in attribution
comes from the abnormal conditions focus model. It posits
that perceivers contrast a target event with other events, iden-
tify any abnormal feature(s) of the target event, and use the
abnormal feature(s) to explain the target event's occurrence.
When few other people react the same way to a stimulus (low
consensus), the target person seems abnormal. For instance, if
someone loves a relatively unpopular activity such as filling out
a tax return or doing laundry, we may conclude it is due to
something peculiar about the person. When a target person
responds differently to other stimuli (high distinctiveness), the
target stimulus seems abnormal. If a person likes everyone
except for one particular coworker, for example, we would
wonder about the coworker. When the target person responds
differently to the target stimulus over time (low consistency),
the present circumstances seem abnormal. If a person does not
usually enjoy opera but enjoys it on a specific occasion, we
would look to the occasion for an explanation. Perceivers also
bring real world knowledge to the table. If an event is an
ordinary occurrence (a scripted event that is high consensus,
low distinctiveness, high consistency), nothing seems abnor-
mal, and thus, the event does not need to be explained. If we
learned that someone ordered food when they were at a res-
taurant, for example, we would not experience a pressing need
to figure out why.
Summary
The classical attribution theories reviewed above established an
interest in determining when perceivers attribute intentional
behavior to something stable about the person (e.g., abilities or
dispositions) or something stable about the situation (e.g., task
difficulty or the nature of the stimulus). Each theorist focused
on different potential determinants: Heider on the relation-
ships between ability, task difficulty, and exertion, Jones and
Davis on the outcomes of various choice options, and Kelley
on other potential reactions as a function of differing actors,
stimuli, and occasions. Common to all theories, however, is
the assumption that it is not always appropriate to conclude
that a person's behavior is due to some stable, inner disposi-
tion. However, as the next section reveals, people sometimes
draw such conclusions even when a person's behavior can be
completely explained by the situation.
Biases
Perceivers do not always follow logical rules or use all of the
available information when making attributions. This section
will review three widely studied biases in the attribution pro-
cess: the correspondence bias, the actor-observer effect, and
self-serving attributions. All of these biases involve a tendency
to favor either dispositional or situational attributions for
certain types of events.
Correspondence Bias
Logically speaking, people should not attribute a behavior to
a corresponding disposition when the situation can explain
238 Attribution
the behavior. For example, if a person writes arguments in
favor of Fidel Castro after being instructed to do so by his
or her debate advisor, a perceiver should not conclude
that the author truly likes Fidel Castro. If a person is nervous
while discussing anxiety-inducing topics, a perceiver should
not conclude that the person is dispositionally anxious.
However, studies have found that perceivers continue to
draw dispositional inferences in such cases. This tendency is
referred to as the correspondence bias, or the fundamental
attribution error.
One reason for the bias is that perceivers are sometimes
unaware of situational factors or the target's perception of
them. If they are unaware that the topics are anxiety-inducing
or that the author felt compelled to comply with the debate
advisor's request, they cannot be expected to take these factors
into account. Additionally, even if perceivers are aware of the
situation, they may underestimate its power. They may mistak-
enly think, for instance, that most people would refuse to write
arguments that conflicted with their beliefs. However, even if
they were to recognize the full strength of the situation, that
knowledge could bias their perceptions of behavior. Knowing
that the person is discussing anxiety-inducing topics could
make the person's behavior seem more anxious than it would
otherwise seem. This type of influence is more likely to occur
when the behavior is somewhat ambiguous. Lastly, as I will
explain later, it can require greater effort on the part of the
perceiver to factor in situational information. Thus, if percei-
vers are either unmotivated or unable to think carefully, they
would not be expected to adjust their attributions in light of
situational information.
Another factor that can encourage people to draw disposi-
tional inferences is the motivation to maintain a belief in a just
world. When another person experiences a negative outcome
(e.g., homelessness, assault), blaming the person allows the
perceiver to keep believing that the world is just and fair. It also
can create a false sense of security; perceivers may believe that
as long as they take the right precautions, they will be able to
prevent negative outcomes.
Actor-Observer Effect
Although perceivers are likely to underweight situational
information when explaining someone else's behavior, they
are better at taking it into account when explaining their own
behavior. For example, one study found that when interacting
with another person, people attributed their own behavior to
the situation, but their partner's behavior to his or her per-
sonal characteristics. This is known as the actor-observer
effect. Researchers have discussed several reasons for this
effect. First, actors are more aware of their feelings, intentions,
prior situational factors, and past behavior than are observers.
Second, situational factors are more salient to actors whereas
behavior is more salient to observers. When engaged in an
interpersonal interaction, we may further be cognitively occu-
pied, unable to estimate the effect that we are having on the
other person. We may also focus on how the other person
differs from us since the environment is shared. Lastly, our
tendency to see others primarily in dispositional terms may be
sustained when we have limited opportunities to observe
them across a variety of situations.
Self-Serving Attributions
Some researchers have argued that the motivation to feel good
about oneself can affect the extent to which actors attribute
their behavior and outcomes to situational factors. This moti-
vation can lead people to make internal attributions for their
positive outcomes and behaviors, and external attributions for
their negative outcomes and behaviors. For example, one study
found that participants who lost a game attributed their nega-
tive outcome less to skill and more to luck than did their
opponents. Conversely, participants who won the game attrib-
uted their positive outcome more to skill and less to luck,
compared to their opponents. Other researchers have argued
that results like this can be explained without drawing on
motivation. For instance, actors may perceive that their beha-
viors are connected to a greater extent with their positive out-
comes than with their negative outcomes. Additionally, given
that people typically intend to succeed and expect success,
negative outcomes may be unexpected, and thus attributed to
situational factors such as bad luck.
Automatic and Controlled Components
The classical attribution models reviewed in the first part of the
article were primarily descriptive. Contemporary models delve
into the underlying cognitive processes. Specifically, they break
the attribution process into different stages or systems. They
distinguish between relatively automatic processes and more
controlled processes. Automatic processes are those that are
efficient, unintentional, involuntary, or occur outside of aware-
ness. Controlled processes are just the opposite: effortful,
intentional, voluntary, and aware. These dual process models
shed additional light on attributional biases, especially the
correspondence bias. I will first describe three prominent
dual process models, and then discuss two lines of research
that illustrate specific issues relevant to the automatic and
controlled ends of the processing spectrum.
Gilbert’s Model
Gilbert's model of person perception focuses on the efficiency
of different stages of the attribution process. According to his
model, people first categorize behavior (e.g., she is being rude)
and then characterize the person is dispositional terms (e.g.,
she is a rude person). These two stages are relatively automatic
(efficient). The third stage involves factoring in the situation
(e.g., she was just insulted), and it requires greater cognitive
effort. Thus, if perceivers are unmotivated or unable to think
carefully, they will not make it to the third stage. This model
helps us predict when the correspondence bias is especially
likely to occur: whenever perceivers lack the motivation or
ability to engage in the final stage of processing.
A common methodology used to test Gilbert's model places
some perceivers under cognitive load by having them rehearse
a series of digits while learning about and rating a target
person. Their judgments will be the product of only automatic
stages (categorization and characterization), as they do not
have the cognitive resources available to engage in the con-
trolled stage (factoring in the situation). Researchers then
Attribution 239
compare their judgments to judgments made by undistracted
perceivers. Such studies typically find that cognitive load pre-
vents perceivers from factoring in situational information.
However, other researchers have noted that situational
information is not inherently more difficult to consider.
If it is particularly salient, even distracted perceivers can use
it to explain behavior. Additionally, if a perceiver's initial goal is
to figure out the situation, he or she can automatically catego-
rize the behavior (e.g., she is being rude) and characterize
the situation (e.g., this is an aggravating situation). In the latter
case, factoring in dispositional factors requires greater effort.
Trope’s Model
Trope's model of person perception can easily accommodate
the finding that people sometimes make relatively automatic
situational attributions. In his model, perceivers automatically
identify behavior and then engage in either diagnostic or pseu-
dodiagnostic attributional reasoning. Diagnostic reasoning
involves thinking about multiple potential causes (disposi-
tions and situations), and it requires motivation and ability.
Pseudodiagnostic reasoning is easier, but it focuses on a single
cause (disposition or situation). So when situational informa-
tion is made especially salient, it can be the focus of pseudo-
diagnostic reasoning, which can take place even under
cognitive load.
Trope's model also goes into more detail about how salient
information about the person and the situation can bias the
early, relatively automatic stage of behavior identification. For
instance, studies have found that perceivers will judge an
ambiguous expression as sadder if they think a person was at
a funeral when the picture was taken. This is an example of an
assimilation effect, where perceivers see the behavior in line
with the context (sad situation, sad expression). However,
contrast effects can also occur, where judgments move away
from the context. Several studies have found that exposure to a
highly successful performance can make a subsequent perfor-
mance seem less successful. This effect happens when percei-
vers are under cognitive load, indicating that it is automatic.
Indeed, when perceivers are not under cognitive load, they
correct for the influence. The latter finding also indicates that
perceivers may correct for a number of factors (e.g., exposure to
other targets) and not just situational factors when they are
motivated and able.
Social Neuroscience Approach
Building on their interest in the automatic and controlled
components of attribution, researchers have recently turned
their attention to the neural substrates of attribution. They
have proposed an X-system that carries out automatic, reflexive
operations, and a C-system that carries out reflective, con-
trolled operations. According to this model, the lateral tempo-
ral cortex, amygdala, and basal ganglia are involved in the
X-system. Current experience is processed in a stream of con-
sciousness fashion, interpreted through the lens of relevant
past experience and current goals. The C-system comes online
when the anterior cingulate sounds an alarm, indicating that
further processing is needed. The prefrontal cortex and hippo-
campus then help perceivers use symbolic logic to make sense
of existing events. The social neuroscience approach opens up a
new realm of possibilities for understanding and studying
attribution.
Spontaneous Trait Inferences
Research on spontaneous trait inferences bears on the automatic
side of the processing continuum, but focuses more on inten-
tionality and awareness rather than efficiency. This research
examines whether people spontaneously infer corresponding
Uaits as causes of behavior. For example, if I read that Bob
gave Sarah half of his sandwich, I may spontaneously infer the
trait 'generosity.' Participants in these studies are typically asked
to read sentences that describe behaviors, and then are
prompted to recall the sentences with one of a number of
prompts: the corresponding trait term (generous), the gist
of the behavior (shared), or words that are semantically related
to parts of the sentence (bread). Numerous studies have found
that the Uait terms help participants recall the sentences, sug-
gesting that participants spontaneously inferred the Uaits when
reading the behavior. This occurred even though participants
had not been instructed to form impressions and even when the
sentences were presented as a disuacter task, indicating that
the inferences were unintentional. Furthermore, participants
tended to be unaware that they were drawing trait inferences.
Thus, this research supports the idea that parts of the atuibution
process can occur automatically.
Trait-Behavior Relations
Research on trait-behavior relations allows us to better under-
stand the corrections that may occur at the conttolled end of
the processing continuum. That is, to fully understand correc-
tion, we need to think about different kinds of traits and how
they map onto behavior. When dealing with capacities or
abilities, like intelligence or athletic ability, highly skilled peo-
ple may perform well or poorly (e.g., if they are not trying), but
unskilled people will typically perform poorly (assuming the
task is at least moderately difficult). A poor performance, then,
could be due to lack of capacity, or to situational factors such as
direct instruction from others to perform poorly or the moti-
vation to save an opponent from embarrassment. However,
because an unskilled person is not capable of executing a
highly skilled performance, a highly successful performance is
diagnostic of high ability. Thus, even if perceivers are moti-
vated to correct their dispositional inferences for any relevant
situational factors, a consultation of their trait-behavior the-
ories would tell them it is logical to assume that a highly
successful performance is due to high ability, even if situational
factors also encouraged success (e.g., trying to impress an
audience or wanting to win a cash prize).
A similar asymmetry occurs with morality-related traits
such as honesty. In this case, however, it is the immoral person
who exhibits a greater range of behaviors. This type of person
may behave morally some of the time and immorally at other
times. However, a moral person should behave morally at all
times. We learn more about a person, then, when we observe
an immoral behavior. Even if a situational factor encouraged
immoral behavior (e.g., another person suggested that the
target person lie or steal), perceivers may not adjust their
240 Attribution
dispositional inferences. Instead, they may maintain that a
person who commits immoral acts is immoral. Moral behavior
is more ambiguous, though, so perceivers would be more likely
to correct their dispositional inferences in light of relevant
situational pressures.
Perceiver Differences
Thus far, perceivers have been treated as interchangeable. Given
certain information and certain processing constraints, percei-
vers will follow similar processes and arrive at similar conclu-
sions. Indeed, that is the social psychological approach - to
look for commonalities in people's responses to social stimuli.
However, we gain a deeper understanding of the attribution
process by examining how perceivers differ from one another in
experience and motivation. This section will examine individ-
ual and cultural differences in attribution.
Individual Differences
Researchers have identified a number of individual differ-
ences that are directly relevant to attribution. I will focus here
on three: attributional complexity, causal uncertainty, and
implicit theories about attributes. These constructs have been
found to moderate classic attribution effects, such as the corre-
spondence bias.
Attributional complexity
Attributional complexity includes a high level of interest in
causal explanations, a preference for more complex, abstract,
and temporally distal explanations, greater thought about the
attribution process, and greater awareness of the influence
that people have on each other. Psychology majors are higher
in attributional complexity than natural science majors.
There is also some evidence that women are higher in attribu-
tional complexity than men. People who are high in attri-
butional complexity generate more causes when describing a
friend's personality, and when making attributions for hypo-
thetical events, they prefer explanations that address more than
one potential cause. Additionally, one study found that people
who are high in attributional complexity are less susceptible
to the correspondence bias. Their more intricate causal sche-
mas make them more sensitive to situational constraints when
judging others.
Causal uncertainty
Causal uncertainty refers to doubts about one's ability to
understand the causes of events in the social world. Such
doubts can be triggered by a loss of perceived control over
events in one's life, and once in place, causal uncertainty can
further undermine a person's sense of control. Men and
women do not differ in their levels of causal uncertainty, but
causal uncertainty is associated with depression, anxiety, and
low self-esteem. Because it threatens control and causes one to
feel bad, people are highly motivated to resolve causal uncer-
tainty. As long as they think there is some chance that their
efforts will pay off, causally uncertain perceivers should
attempt to improve their understanding of events. They may
start with relatively automatic strategies for understanding and
progress to more effort-intensive strategies if needed. Auto-
matic strategies could include accessing stored schemas and
heuristics. Controlled strategies could include gathering and
carefully processing causally relevant information. Consistent
with the idea that they use controlled processing to improve
their understanding, studies have revealed that individuals
who are high in causal uncertainty are less susceptible to the
correspondence bias. As long as their cognitive load does not
prevent them from doing so, they carefully consider situational
constraints when judging others.
Implicit theories
Implicit theories refer to people's beliefs about the nature of
attributes such as personality, intelligence, and morality. Entity
theorists think that such attributes are fixed, whereas incremen-
tal theorists think that they are malleable. These beliefs have
implications for the attribution process. Entity theorists place a
high value on dispositional inferences and readily draw them
based on small samples of behavior. Furthermore, they tend to
generalize to broader attributes and assume those attributes
will remain relatively stable over time. To entity theorists,
dispositional inferences provide an explanation for why a par-
ticular behavior or outcome occurred. Incremental theorists
draw dispositional inferences as well, but they are more
responsive to new contradictory information, they generalize
less, and assume less stability. They focus more on the situa-
tional context and a person's strategy, effort, goals, and inten-
tions. To incremental theorists, dispositional inferences simply
summarize a person's behavior in a particular situation.
Researchers have examined how these differing beliefs
affect judgments about the self and others. Implicit theories
predict responses to setbacks. When people receive news that
they performed poorly on a test, entity theorists draw negative,
global inferences about their abilities, while incremental the-
orists focus more on strategy and effort. When judging another
person's negative behaviors, entity theorists draw more global
and stable negative dispositional inferences and recommend
greater punishment. More troubling, they more readily draw
inferences about morality from a person's appearance, and
base their verdicts upon these inferences rather than the
strength of evidence presented in a criminal case. Thus, implicit
theories about the malleability of attributes, which differ across
individuals and domains, help us predict the extent to which
perceivers will draw correspondent dispositional inferences
from behavioral information.
Cultural Differences
Just as individuals differ from one another in attributionally
relevant beliefs, so do cultures. Different thinking styles prevail
in western (European American) and eastern (East Asian) cul-
tures. People in western cultures tend to think analytically,
focusing on objects and individuals. People in eastern cultures
tend to think holistically, focusing on the context in which
objects and individuals are located. These styles affect the
ways in which people in the two types of cultures view them-
selves and others. Westerners use more abstract traits to
describe themselves and others, whereas Easterners refer more
to specific contexts and social identities. Furthermore, a num-
ber of studies have revealed that Easterners are less susceptible
Attribution 241
to the correspondence bias than are Westerners when situa-
tional constraints are made salient (e.g., they just experienced
the same situational constraint themselves). However, when
situational constraint information is available but not salient,
both groups show the correspondence bias.
In some respects, Easterners and Westerners share similar
views of dispositions. They have similar conceptualizations of
personality and predict that people engaging in a certain type
of behavior once will engage in similar types of behavior in the
future. Furthermore, when dispositional and contextual expla-
nations are assessed separately, Easterners and Westerners typ-
ically differ in the amount of contextual explanation, but not
always in the amount of dispositional explanation. However,
there are subtle differences in their views of dispositions. East-
erners are more like incremental theorists, while Westerners are
like entity theorists. Thus, Easterners see dispositions as mal-
leable, and they also acknowledge the role of the context in
shaping dispositions. Lastly, Easterners are more likely than
Westerners to attribute causality to a group, rather than to an
individual.
In sum, while basic social psychological research provides a
good starting point for common biases in the attribution pro-
cess, research on individual and cultural differences sheds light
on moderating factors. Individuals who are high in attribu-
tional complexity or causal uncertainty, or who subscribe to
an incremental theory or Eastern culture are more likely
to consider situational constraints on behavior.
Implications
Motivation and Psychological Well-Being
Up until now, I have focused primarily on the degree to which
perceivers see the cause of a behavior or outcome as internal or
external. However, there are meaningful differences within
these two general categories. Some factors (ability, task diffi-
culty) are more stable than others (effort, luck). In addition,
some factors (effort) can be controlled more easily than others
(luck). Lastly, some factors (social skills) affect a broader range
of events than others (ability to remember names).
These dimensions affect a person's motivation level as well
as their overall psychological well-being. The attributions peo-
ple make for an unsuccessful performance (e.g., a low test score
or negative job evaluation) can greatly affect their motivation.
If people attribute the failure to insufficient effort (an internal,
unstable, and controllable factor), improvement will seem
possible and motivation can be maintained. However, if they
attribute the failure to a lack of intelligence (an internal, stable,
uncontrollable factor), improvement will seem unlikely and
motivation may decline. Research has revealed that people
who are depressed tend to make the latter type of attribution,
and furthermore think the cause will affect many other out-
comes. This stands in contrast to the self-serving attributions
mentioned earlier, where people attribute failures to external
factors.
Self-Handicapping
People sometimes try to manipulate plausible attributions in
order to prevent people from attributing their failure to a lack
of ability. Specifically, when individuals doubt their ability to
succeed, they sometimes withdraw effort or engage in an activ-
ity that interferes with their performance. Because any one of
these factors can explain a failure, it becomes more difficult to
attribute the failure to a lack of ability. This type of behavior is
known as self-handicapping. As one might imagine, it can have
negative implications for performance in the classroom and
beyond.
Marital Satisfaction
Attributions can also affect how we feel about the important
people in our lives. Research has revealed that the type of
attributions married people make about their spouses' beha-
viors are associated with their levels of marital satisfaction,
emotions, and behavior. Relationship researchers have distin-
guished between causal attributions, or judgments about who
or what brought about an event, and attributions of responsibil-
ity and blame, or judgments about who is accountable and at
fault for an event, respectively. These attributions tend to build
on each other, with internal attributions potentially leading
to responsibility attributions, and responsibility attributions
potentially leading to blame attributions. However, in mar-
riage, responsibility and blame tend to co-occur. To measure
causal attributions, researchers have presented individuals with
hypothetical negative partner behaviors and asked them to rate
the extent to which something about the partner caused the
behavior (locus), as well as whether the reason will change
(stability) and affect other parts of their marriage (globality).
Unsatisfied individuals tend to attribute negative partner beha-
viors to causes that are internal (something about the partner),
stable (not likely to change), and global (will affect many other
areas). Responsibility/blame attributions are assessed by hav-
ing individuals rate whether the negative partner behaviors
were intentional, selfishly motivated, and blameworthy.
Greater responsibility/blame attributions are associated with
less marital satisfaction, and during an interaction, greater
anger (wives only) and whining (husbands and wives).
Findings from longitudinal studies have supported the idea
that a negative attributional style predicts decreases in marital
satisfaction over time. There is also some evidence for the
opposite path as well: low satisfaction sometimes predicts
the development of a more negative attributional style over
time. Importantly, these relationships held up when related
variables such as depression were controlled.
Conclusion
Over 50 years of attribution research has shed much light on
the processes people use to understand the causes of events, the
biases they exhibit along the way, the beliefs that impact
people's attributions, and the implications of particular attri-
butions for important domains. We see that people can spon-
taneously and efficiently draw dispositional inferences about
others. Sometimes they stop here, and if they have neglected
relevant situational information, they would be said to have
shown the correspondence bias. Perceivers are more likely to
factor in situational constraints if they are motivated and able
(due to accuracy motivation, attributional complexity, causal
242 Attribution
uncertainty, lack of cognitive load), if the situational informa-
tion is salient, and if their beliefs encourage such a consider-
ation (incremental theories. Eastern cultures, trait-behavior
relations). Furthermore, we see that certain types of attributions
can enhance functioning. Specifically, attributing a negative
performance to a lack of effort, rather than ability leaves room
for improvement. Attributing negative partner behavior to
external, unstable, and specific factors helps people maintain
positive views of their relationships. Finally, we see the difficul-
ties that people face when they lack confidence in their ability to
understand the causes of events. This relatively new research on
causal uncertainty supports assertions that Heider and others
have made that understanding why events occur is critical for
a sense of prediction and control over one's environment.
See also: Cognitive Bias; Marital Dysfunction; Perceptual
Development.
Further Reading
Bradbury TN and Fincham FD (1990) Attributions in marriage: Review and critique.
Psychological Bulletin 107: 3-33.
Choi I, Nisbett RE, and Norenzayan A (1999) Causal attribution across cultures:
Variation and universality. Psychological Bulletin 125: 47-63.
Dweck CS, Hong Y, and Chiu C (1993) Implicit theories: Individual differences in the
likelihood and meaning of dispositional inferences. Personality and Social
Psychology Bulletin 19: 644-656.
Fletcher GJO, Danilovics P, Fernandez G, Peterson D, and Reeder GD (1986)
Attributional complexity: An individual differences measure. Journal of Personality
and Social Psychology 51 : 875-884.
Gilbert DT and Malone PS (1995) The correspondence bias. Psychological Bulletin
117: 21-38.
Heider F (1958) The Psychology of Interpersonal Relations, pp. 79-124. New York:
Wiley.
Hilton DJ and Slugoski BR (1986) Knowledge-based causal attribution: The abnormal
conditions focus model. Psychological Review 93: 75-88.
Jones EE and Davis KE (1965) From acts to dispositions: The attribution process in
person perception. In: Berkowitz L (ed.) Advances in Experimental Social
Psychology, vol. 2, pp. 219-266. New York: Academic Press.
Jones EE and Nisbett RE (1972) The actor and the observer: Divergent perceptions
of the causes of behavior. In: Jones EE, Kanouse DE, Kelley HH, Nisbett RE,
Valins S, and Weiner B (eds.) Attribution: Perceiving the Causes of Behavior,
pp. 79-94. Morristown, NJ: General Learning Press.
Kelley HH (1967) Attribution theory in social psychology. Nebraska Symposium on
Motivation 15: 192-238.
Lieberman MD, Gaunt R, Gilbert DT, and Trope Y (2002) Reflexion and reflection:
A social cognitive neuroscience approach to attributional inference. Advances in
Experimental Social Psychology 34: 199-249.
Reeder GD (1993) Trait-behavior relations and dispositional inference. Personality and
Social Psychology Bulletin 19: 586-593.
Trope Y and Gaunt R (1999) A dual-process model of overconfident attributional
inferences. In: Chaiken S and Trope Y (eds.) Dual-Process Theories in Social
Psychology, pp. 161-178. New York, NY: Guilford Press.
Uleman JS, Newman LS, and Moskowitz GB (1996) People as flexible interpreters:
Evidence and issues from spontaneous trait inference. In: Zanna MP (ed.)
Advances in Experimental Social Psychology, vol. 28, pp. 211-279. San Diego:
Academic Press.
Weary G, Tobin SJ, and Edwards JA (2010) The causal uncertainty model revisited.
In: Arkin RM, Oleson KC, and Carroll PJ (eds.) Handbook of the Uncertain Self,
pp. 78-100. New York: Psychology Press.
Autism and Pervasive Developmental Disorders
R G Romanczyk and E H Callahan, Institute for Child Development, SUNY at Binghamton, Binghamton, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Diagnostic criteria The established guidelines for a specific
diagnosis that include the specific behaviors, the course of
symptom development, and characteristics an individual
must exhibit.
fMRI Functional magnetic resonance imaging, used to
observe brain functioning.
Prevalence The number of people in the population who
have a particular characteristic or diagnosis. Often expressed
as the observed number per 10 000 people.
Stereotyped A rigid repetition of a specific behavior or
activity that is often nonfunctional.
Stress A negative physical or emotional reaction to events or
series of events in the environment that may or not produce
similar effects in other individuals.
Stressor An event or series of events or interactions
that can precipitate a negative emotional or physical
reaction.
Perspective
Leo Kanner first described the syndrome of autism in the early
1940s. There are long historical roots to the conceptualization
of severe early childhood disturbance, primarily referring to
dementia and intellectual limitations. A few detailed accounts
of individual cases, such as Itard's classic description of the
wild boy of Aveyron, provide clear examples of the confusion
that existed as to how to best conceptualize severe pathology in
children. Descriptions of educational programs in the mid-
1800s have fascinating detail of treatment programs and indi-
viduals who today would most likely be diagnosed as having
autism. The term autistic was first used by Eugene Bleuler in the
early 1900s to describe a specific thought process characterized
by poor logic and a detachment from reality, but was not in
reference to a specific disorder.
These historical accounts were at times very systematic
descriptions and provide interesting continuity to the present.
What was lacking was effective formal differentiation or diag-
nostic criteria. Kanner is credited for identifying the syndrome
and asserting its distinguishing characteristics that set it apart
from other severe disorder categories of the time. He provided
clear description of an initial group of eight boys and three
girls, drawn from the very large pool of his clinical experience
as a psychiatrist at John Hopkins. He chose the term 'autism' to
describe his observations, a term also associated with the
description of social withdrawal in schizophrenia. However,
Kanner's perspective was quite different. Unfortunately, there
still persists a popular but simplistic image of individuals with
autism solely as being withdrawn and retreating from social
interaction. Kanner used the description 'aloof to describe the
type of social interactions, and lack thereof, that he observed.
'Aloof better describes the feeling of detachment often
observed in social interactions of individuals with autism,
connoting more of a detachment than withdrawal in the emo-
tional sense.
Kanner identified almost a dozen specific characteristics that
he associated with the syndrome he termed early infantile autism.
Some of these were: aloofness, excellent rote memory, echolalia,
pronoun reversal, an intense desire to maintain sameness, alert
expression, poor eye contact, appearance of being deaf, no antici-
patory reaching out from infancy, lack of social and communica-
tion initiation, normal intelligence (as assessed by the Seguin
Form Board), and normal motor coordination.
In his description of the 1 1 children that he presented, he
noted considerable variation in the number and severity of
behaviors observed, but he conceptualized the syndrome as
primarily an 'autistic disturbance of affective contact' and later
he utilized the term 'early infantile autism.' Importantly, how-
ever, he saw the primary feature as the inability to relate to
other people and situations from infancy. He specifically con-
trasted his syndrome with others such as Heller's dementia
infantilis and DeSanctis' dementia praecocissima in that these
are characterized by a period of normal development prior to
symptom onset. In contrast, Kanner stressed the presence of
symptoms from infancy: hence the use of the term 'early infan-
tile autism.' This emphasis on the presence of autism from
birth was maintained by Kanner when he stated almost
30 years later that the primary characteristics for autism were
'(a) the children's inability from the beginning of life to relate
themselves to people and situations in the ordinary way and
(b) an anxiously obsessive desire for the preservation of same-
ness.' However, as various researchers have pointed out, by
focusing on just the social isolation and preservation of same-
ness, this reduced the number of symptoms thought to be part
of the syndrome, and may have inadvertently led to overly
broad diagnostic criteria.
Diagnosis
Terminology
As listed in the Diagnostic and Statistical Manual of Mental
Disorders, Fourth Edition, Text Revision (DSM-IV-TR) of
the American Psychiatric Association, the current standard
for diagnosis, the pervasive development disorders (PDDs)
include: autistic disorder, Asperger's disorder, Pervasive devel-
opmental disorder-not otherwise specified (PDD-NOS), Rett's
disorder, and childhood disintegrative disorder. With a very
rapid rise in the diagnosis of autistic disorder over the past two
243
244 Autism and Pervasive Developmental Disorders
decades, the attention directed at the PDDs has increased.
However, the terms used to describe the disorders within the
PDDs have grown imprecise. Terms such as 'autism spectrum
disorders' and 'ASD,' which do not appear in the DSM-IV-TR,
are often used as if they were a diagnosis, are frequently used
interchangeably with PDD and autistic disorder. The lack of
clarity when using such terms creates confusion, particularly
among consumers. The inconsistent use of such terms is not
limited to just the popular media, but is also observed in
professional writings. Websites and publications often use the
term ASD to describe individuals without specifying which
DSM diagnostic categories individuals actually fall into, and
in what proportion, for the research article or description pre-
sented. In addition, national organizations such as the National
Institute of Mental Health (NIMH) and the American Academy
of Pediatrics differ in their use of the term ASD. The NIMH
includes all of the disorders within the PDDs as part of
their definition of ASD whereas the American Academy of Pedi-
atrics includes only autistic disorder, Asperger's disorder, and
PDD-NOS under their use of the term ASD. Figure 1 illustrates
the DSM-IV-TR PDD categories and the partial overlap of the
term ASD by the two organizations.
A related problem is the use of descriptive labels for pre-
sumed subcategories of autistic disorder that also have yet to be
adequately defined. For example, the terms high-functioning
autism and low-functioning autism are widely used, often in
the context of presumed severity and prognosis; yet they have no
reference in the DSM-IV TR. The term high-functioning autism
typically is used to refer to individuals with a diagnosis of
autistic disorder who have adequate verbal language abilities
and an average or above-average level of cognitive ability. Indi-
viduals described as having high-functioning autism present
with symptoms that are clinically similar in a number of ways
to that of individuals with Asperger's disorder. However, there
is strong debate in the field as to whether there are meaningful
clinical and prognostic differences between individuals diag-
nosed with Asperger's disorder and those described as having
high-functioning autism. This is an important area of research
that has begun to receive more attention, as it might have
implications for the study of etiology, response to treatment,
type of treatment recommended, and prognosis. Currently the
term high-functioning as a qualifier for autism is not consis-
tently defined and creates potential diagnostic overlap. This can
lead to problems when interpreting research and clinical
descriptions.
Related to this concern, proposals for subtyping within
autistic disorder, which could be essential for both clinical
and etiological research, have come primarily from clinical
observation. Perhaps the most cited proposed subtyping
comes from Lorna Wing and her colleagues who describe
three subtypes: aloof, passive, and active-but-odd, to explain
observations as to the variation in types of social deficits.
Recent research questions the validity of these specific cate-
gories, but does support the presence of subtypes, with
replication of research findings required for more precise
specification.
Symptomatology
The three most prevalent PDDs are autistic disorder, Asperger's
disorder, and PDD-NOS, which from here forward will be
referred to as ASD. While they each have separate diagnostic
criteria, the DSM-IV-TR outlines three core characteristics of
interest that are associated with these three disorders. These
include:
• Impairment in social interaction
• Impairment in communication (specifically as used in
social situations)
• Stereotypic, repetitive behaviors, and restricted interests
Diagnostically, the difference between the disorders lies in
the severity and nature of each of these core characteristics and
their pattern. Table 1 provides a comparison of the three dis-
orders across the three core characteristics. A description of the
specific behaviors associated with each of the core character-
istics of interest follows.
Impairment in social interaction
Kanner's initial description of autism highlighted the deficits
in social interaction as central to the disorder. Currently,
these deficits are still considered to be the most important
of the core deficits. Some have argued that these social inter-
action deficits, because of their specific manner of expression
and pervasive nature, are what differentiate ASD from other
clinical populations who also show social interaction
problems.
As listed in the DSM-IV-TR, impairments in social
interaction include: (1) deficits in nonverbal behaviors such
as eye-to-eye gaze, facial expression, and gestures to regulate
social interaction; (2) failure to develop peer relationships
DSM-IV-TR
Pervasive developmental disorders
Figure 1 A comparison of definitions of autism spectrum disorders.
Autism and Pervasive Developmental Disorders 245
Table 1 A comparison of core diagnostic characteristics for autistic disorder, Asperger’s disorder, and PDD-NOS
Autistic disorder Asperger 1 s disorder PDD-NOS
Age considerations Delay in at least one area is present prior
to the age of 3.
Marked impairment in At least two of the following must be
social skills present:
- Marked impairment in the use of
multiple nonverbal behaviors
(eye-to-eye gaze, facial expression,
body postures, and gestures) to
regulate social interaction
- Failure to develop peer relationships
appropriate to developmental level
- Lack of spontaneous seeking to
share enjoyment, interests, or
achievements with others
- Lack of social or emotional
reciprocity
Marked impairment in At least one of the following must be
communication present:
- Delay in, or total lack of, the
development of spoken language
(without attempts compensate
through alternative means)
- Marked impairment in the ability to
initiate or sustain conversation with
others
- Stereotyped and repetitive use of
language or idiosyncratic language
- Lack of varied, spontaneous make-
believe play or social imitative play
appropriate to developmental level
Restricted, repetitive, At least two of the following must be
and stereotyped present:
patterns of behavior, - Encompassing preoccupation with
interests, and one or more stereotyped and
activities restricted patterns of interest that is
abnormal either in intensity or focus
- Apparently inflexible adherence to
specific, nonfunctional routines or
rituals
- Stereotyped and repetitive motor
mannerisms (e.g., hand or finger
flapping or twisting, or complex
whole-body movements)
- Persistent preoccupation with parts
of objects
Other considerations Behavior not better accounted for by
Rett’s disorder or childhood
disintegrative disorder
After 36 months
Same criteria as autistic disorder
None specified
Severe and pervasive impairment in
development of social interaction
No clinically significant delay in language Impairment in either verbal and/or
development present nonverbal communication skills
and/or presence of stereotyped
behavior, interests, and activities
Same criteria as autistic disorder Presence of stereotyped behavior,
interests, and activities and/or
impairment in either verbal and/or
nonverbal communication skills
- Behavior causes clinically significant
impairment in social, occupational, or
other important areas of functioning
- No clinically significant delay in
cognitive development or in the
development of age-appropriate
self-help skills, adaptive behavior, and
curiosity about the environment in
childhood
- Criteria not met for another specific
PDD or schizophrenia
Criteria not met for a specific PDD,
schizophrenia, schizotypal
personality disorder or avoidant
personality disorder
appropriate to developmental level; (3) lack of spontaneous
seeking to share enjoyment and interests (e.g., by a lack of
showing, bringing, or pointing out objects of interest); and
(4) lack of social or emotional reciprocity.
Social deficits observed in children with ASD include diffi-
culties understanding the facial expressions of others, initiating
social interactions, responding to the social bids of others, and
responding to the emotions of others. Research has also
246 Autism and Pervasive Developmental Disorders
identified specific social behaviors that differentiate children
with ASD from typically developing children at 2 years of age.
These behaviors include:
• decreased attention to the voices of others (e.g., failing to
respond to their name when called);
• the use of another's body as a tool (e.g., taking another's
hand and putting it on a toy in an attempt to have the
person turn it on);
• little to no interest in other children (e.g., preferring to
engage in isolated play when with other children, failure
to initiate social interactions with peers, indifference to the
activities of other children);
• limited use of pointing to communicate with others;
• deficits with understanding gestures;
• lack of seeking to share enjoyment with others (e.g., not
showing items of interest to an adult or peer or directing the
attention of others to events in their environment).
Joint attention behaviors have been an area of particular
interest for researchers and clinicians studying social deficits in
ASD as it has been proposed that this specific deficit of joint
attention is specific to ASD and not observed typically in other
clinical populations. Joint attention has been described as
using gestures and mutual eye gaze to share interest, enjoy-
ment, and attention. Examples of these joint attention beha-
viors include: a child shifting his or her gaze between an
interesting object and a familiar person, combining gaze shift
and pointing to direct another individual's attention to an
object of interest in the environment, or showing a toy to
someone as part of initiating a social interaction. Whereas
typically developing children begin to exhibit joint attention
behaviors around 15 months of age, joint attention deficits
tend to be a characteristic of individuals with ASD throughout
their lives. Other behaviors observed early in the social devel-
opment of typical children that are problematic in individuals
with ASD include, but are not limited to: imitation, social play,
friendship-seeking behavior, comfort seeking, responding to
the facial expressions of others, and cooperative play.
Striking in some, the social deficits in others with ASD are
more subtle. For example, individuals with Asperger's disorder
display many of the social difficulties described above, but in
particular have a difficulty identifying and interpreting the social
cues of others. Some research has shown that some individuals
with Asperger's can learn skills, such as how to correctly identify
emotions and other social cues. However, they often have diffi-
culty using those skills in their everyday environment, making
the development of friendships and maintenance of social rela-
tionships difficult. Complicating the picture further, individuals
with Asperger's possess typical cognitive and verbal abilities.
They are often aware of the fact that they are socially different
from others, and it is not uncommon for them to express
frustration with these issues, placing them at risk for a number
of other issues including lower academic achievement, depres-
sion, and anxiety. Further, the changing nature and often emo-
tional intensity of adult social and sexual relationships present
great difficulty for individuals with Asperger's disorder.
Impairment in communication
The communication deficits observed in individuals with ASD
can vary. A delay in the development in expressive language is
observed in some, others may never develop meaningful
speech, and some may have no delay in the development of
speech but have difficulty with the more pragmatic aspects of
conversation. It is often estimated that up to 50% of indivi-
duals with autistic disorder do not develop spoken language,
whereas to receive a diagnosis of Asperger's disorder, indivi-
duals must not have had any delays in the development of
spoken language.
It is estimated that a majority of individuals with ASD fail to
develop spoken language. Whereas children with speech delays
typically compensate for their difficulties through the use of
nonverbal communication (e.g., physical gestures, pointing,
eye contact), individuals with ASD do not. For individuals
with autistic disorder, these areas of nonverbal communication
are either absent or noticeably impaired throughout their
development. A large body of research has been devoted to
assessing and developing methodologies to teach individuals
with ASD, who do not have spoken language, alternative meth-
ods (e.g., sign language, pictures) of communication. A num-
ber of different strategies have been successful for teaching
individuals with ASD to communicate their wants and needs.
However, even when these individuals acquire an alternative
means of communication, the use of communication to initi-
ate social interactions and conversations tends to remain
deficient.
The language of individuals who develop expressive speech
is often rote, repetitive and can be lacking in communicative
meaning. Echolalia is a characteristic of speech patterns in the
majority of individuals with autistic disorder. Echolalia is the
repetition of the speech of another person or oneself and it can
be immediate (i.e., repeating a word, phrase, or sentence
immediately after it is spoken by someone else) or delayed
(i.e., repeating speech heard in the past). Although echolalia
often appears to be unrelated to an individual's environment,
there is some question in the field as to whether there is a
communicative function of echolalia for individuals with ASD.
When functional language does develop, pragmatic deficits
often persist. Such deficits include: difficulties with 'turn-
taking' in conversation, and preoccupation with one conversa-
tional topic. For example, individuals with Asperger's disorder
have the verbal abilities to carry on a conversation but typically
choose to keep the conversation focused on a specific topic of
interest, be indifferent to requests to change topics, and have
difficulty taking turns in the conversation.
The prosody and intonation of speech are also impaired in
individuals with ASD. Prosody, the rhythm of spoken language,
helps a speaker add specific meaning to a phrase or sentence.
For example, when asking a question, most individuals raise
their voice slightly at the end of the sentence, helping a conver-
sational partner understand the intent of a sentence or phrase.
Individuals with ASD typically have abnormal intonation and
use of rhythm and stress in speech production. Other charac-
teristic speech abnormalities include pronoun reversals, articu-
lation difficulties, and idiosyncratic use of speech, including the
development and use of neologisms.
Stereotyped and repetitive behavior
The broad category of restricted and repetitive behaviors
in individuals with ASD includes self-stimulatory behavior
(e.g., repetitive finger movements, gazing at lights, rocking,
Autism and Pervasive Developmental Disorders 247
toe walking), stereotypical motor behavior (e.g., posturing),
preoccupation with objects and activities, insistency with
sameness or inflexibility in routines, and restrictive interest in
objects or activities. Parents of young children with ASD may
notice that their children play with toys in an odd manner.
Rather than using cars or blocks for pretend play, children with
ASD may prefer to line up toys according to attributes such as
shape, size, or color. Children with ASD may also become
preoccupied with one part of a toy and their toy play may
become restricted (e.g., spinning the wheel of a toy truck over
and over). In addition, individuals may develop interests in
specific types of toys or collections, which are greater in inten-
sity than seen with typically developing individuals.
When compared, the toy play of individuals with ASD
shows clear differences from the play of typically developing
children. Children with autism exhibit limited functional and
symbolic play, atypical play preferences, lack of creativity and
flexibility in their play, and limited capacity for imitation.
Children with ASD also use less appropriate, less varied, and
more repetitive play. Changes in routines can also be difficult
for individuals with ASD. Children with ASD often exhibit a
desire for sameness in their environment. Disruptions in daily
routines, or changes in their environment (e.g., rearranging
furniture in a common room) can precipitate behavior pro-
blems (e.g., crying, tantrum throwing) in children with ASD.
Repetitive behavior in individuals with ASD may also man-
ifest itself as a persistent, focused, and intense preoccupation.
For example, an individual may become fixated on learning
everything there is to know about train schedules. For others, it
may be clocks, maps, radios, or a particular sports team. These
focused interests differ from hobbies or more typical interests
of others, by their exclusive focus. For example, many typically
developing individuals are passionate about a particular sports
team. They enjoy watching their team play, talking to others
about the team, wearing the team colors, and reading about
them. An individual with ASD who is fixated on that same team
may know all of the numbers for every player who has been on
that team and their home towns, but does not necessarily enjoy
watching the team play, and cannot carry on a conversation
about that team beyond what numbers the players wore or
from where they came. Individuals with ASD might talk at
length about a specific area of interest, even after the conversa-
tional partner has displayed disinterest in the topic or
attempted to change the topic of conversation. Further, the
typical fan, no matter how devoted, is able to carry on social
conversations on other topics, whereas the individual with ASD
often has great difficulty in exhibiting this type of flexibility.
Prevalence
For decades after Kanner's publication, autism was thought to
be extremely rare. In the 1960s, it was estimated to affect less
than 1% of the population. Research conducted in the 1980s
and 1990s presented a different picture with prevalence rates
almost 10 times what were reported previously. Most recent
figures released from the Center for Disease Control report the
prevalence rates of ASD between 45/10 000 and 67/10 000.
While the prevalence statistics have changed over time, the
ratio of diagnosis between boys and girls has remained fairly
consistent. It is estimated that the rate of ASD is 3-4 times
higher in boys than in girls. The increase in prevalence statistics
has led some to conclude that we are experiencing an epidemic
while others make arguments that the explosive increase in the
prevalence of ASD does not reflect a single cause epidemic in
the typical sense, but rather may be the result of changing
diagnostic criteria, diagnostic substitution, early identification
programs, increased services specific for individuals with ASD,
and increased awareness.
Changes in Diagnostic Criteria
In the 1950s Kanner and Eisenberg published a list of criteria
for early infantile autism. Their criteria included aloofness and
indifference to others, and an intense resistance to change in
the child's own repetitive routines. They also specified that
these behaviors needed to be present before 24 months of age.
Michael Rutter, a researcher working primarily in Great
Britain, suggested in 1978 modifications for the definition of
childhood autism. He presented four criteria: (1) Impaired
social development, which has a number of special character-
istics out of keeping with the child's intellectual level; (2)
Delayed and deviant language development that also has
certain defined features and is out of keeping with the child's
intellectual level; (3) Insistence on sameness as shown by
stereotyped play patterns, abnormal preoccupations, or resis-
tance to change; and (4) Onset before 30 months. Rutter fur-
ther gave examples of the behaviors based on his own research
and clinical experience, as well as on Kanner's descriptions.
Autism was formally recognized in the DSM, in 1980, with
the publication of the Diagnostic and Statistical Manual Third
Edition (DSM-III). Referred to as Infantile Autism, it was
included under a new class of disorders - the PDDs. To receive
a diagnosis, individuals had to meet six mandatory diagnostic
criteria. The criteria included onset before 30 months of age;
a pervasive lack of responsiveness to other people; gross
deficits in language development; when speech is present,
peculiar speech patterns such as immediate and delayed echo-
lalia, metaphorical language, and pronoun reversal; bizarre
responses to various aspects of the environment (e.g., resis-
tance to change, peculiar interest in or attachments to objects);
and the absence of delusions, hallucinations, loosening of
associations, and incoherence as observed in individuals
diagnosed with schizophrenia.
Seven years later, the diagnostic criteria were modified
with the publication of the revised edition of the DSM-III
(DSM-III-R). Now referred to as autistic disorder, the diagnos-
tic criteria were divided into three main categories (impairment
in reciprocal social interaction, impairment in verbal and non-
verbal communication, and resttictive repertoire of activities
and interests) with specific behaviors (16 total) listed under
each one. To receive a diagnosis of autistic disorder, indivi-
duals had to exhibit at least 8 of the 16 behaviors listed, with
two from the reciprocal social interaction category and at least
one from each of the other categories. Onset also needed to
occur in infancy or early childhood. Described earlier, the
criteria were further modified with both the publication of
the DSM-IV and the DSM-IV-TR.
Studies comparing DSM-III to DSM-IV have found that the
DSM-IV broadened the diagnostic criteria of autistic disorder.
248 Autism and Pervasive Developmental Disorders
To examine the effect that changes in the diagnostic criteria
could have on prevalence, research has been conducted in
which the DSM-III and DSM-IV criteria are applied to a sample
population. Research comparing the diagnostic criteria of the
two versions has revealed that the application of the DSM-IV
criteria resulted in significant increases in diagnosed cases of
autistic disorder over the DSM-III criteria.
If one examines the prevalence rates for autism over the last
two decades, one can see a direct relationship between the date
of publication of DSM-IV, in 1994, and DSM-IV-TR, in 2000,
and a corresponding significant increase in prevalence rates in
the United States. In addition, studies have found that when
case definitions and identification methods are held constant,
no significant increase in diagnosis is observed. While it
appears a relationship may exist between changes in diagnostic
criteria and prevalence of autism, it is not likely the sole cause
of the observed increase.
Diagnostic Substitution
In addition to changes in diagnostic criteria, researchers have
examined the possible influence of diagnostic substitution and
the increase in the prevalence of autism. The premise of 'diag-
nostic substitution' is that an individual who would typically
receive a specific diagnosis would now receive a different diag-
nosis because of changes in social acceptability, consumer
preference and selection of diagnostician, differential availabil-
ity of services based upon diagnostic classification, and lessen-
ing concern about the need for specific differential diagnosis.
If in fact diagnostic substitution is influencing the preva-
lence rate of autism, the observed increase in the prevalence
of autism would coincide with a similar decrease in the preva-
lence of some other disability (or disabilities). This, in fact,
has been reported by numerous researchers, indicating that
substitution may be occurring for the diagnosis of mental
retardation.
Public Awareness of Autism and Early Identification
Awareness and publicity of autism has increased dramatically
during the last two decades. A review of the archives of Time
Magazine revealed a considerable increase in the number of
stories published about autism. Between 1980 and 1989, three
articles were published about autism. Between 1990 and 1999,
the number of articles increased to 1 1 including 2 cover stories.
Between 2000 and 2009, Time Magazine has published 73
articles about autism including 6 cover stories.
Earlier identification and detection of autism has also been
suggested to be a factor in the observed increase in the preva-
lence of autism. Improved diagnostic assessments and
increased awareness have been suggested to play a role in the
early detection and diagnosis of the disorder. In the last two
decades, the number of assessment measures and procedures
has increased dramatically with regard to both the diagnostic
process and screening. As would be expected, research has
indicated that the mean age of diagnosis of autism and start
of service delivery has declined over this period.
There is no denying that there has been an increase in
the prevalence of ASD over time. Whether the increase is due
to factors such as changing diagnostic criteria or better
identification, or multiple other factors is unclear. Continued
research in this area as well as the etiological roots of ASD will
help develop a better understanding of its prevalence.
Etiology
To date, the specific causes of ASD remain unknown. Hypoth-
eses about the etiology of ASD have ranged from psychologi-
cal, to environmental, to biological. However, research focused
on genetic and neurobiological factors has been most
promising. More than 15 genes have been implicated in the
etiology of ASD. Information obtained from twin studies pro-
vides further support for the role of genetic factors in the
development of ASD. In studies with monozygotic twins, the
median concordance rate for autistic disorder is reported to be
as high as 90%. The prevalence of ASD among nontwin sib-
lings of children with ASD is also significantly higher than that
of the general population, with estimates ranging from 4% to
8%. In addition, phenotypic studies have revealed that char-
acteristics such as rigidity, social anxiety, and communication
difficulties are seen more frequently in relatives of individuals
with ASD compared to controls.
While the current research supports a biological link in the
development of ASD, the data on the specifics of the link is less
clear. Multiple genes have been implicated in the development
of ASD including chromosomes 2, 7, 13, 15, 16, and 17 and
the X chromosome. Neuroanatomical features that have been
hypothesized to be linked to ASD include abnormal function-
ing of the brain stem, cortex, cerebellum, hippocampus, and
amygdala. There is also evidence suggesting that the brain is
enlarged in children with autism. However, the implications of
brain enlargement have yet to be determined. Lastly, some
research has implicated neurotransmitter abnormalities in indi-
viduals with ASD. Specifically, abnormalities in serotonin and
GABA levels have been found in some individuals with ASD.
A major difficulty in drawing conclusions from the current
research evidence from the neurobiology and neuroimaging
fields is that the sample size is still small and is typically biased
towards the individuals with ASD who are considered 'higher
functioning.' This is understandable given that the behavioral
issues that many individuals with ASD exhibit prevent them
from participating in intensive assessments such as functional
magnetic resonance imaging (fMRI). However, the absence of
data from individuals with more challenging behaviors makes
generalizing the results of such studies difficult. Advancement
in the research methods used to collect this type of data is
needed to help researchers collect information from a broader
population and develop a better understanding of the etiologi-
cal roots of ASD. It cannot be underscored enough that the
ASD represent a heterogeneous population and that there are
likely multiple pathways in its development.
Treatment Challenges
Individuals with ASD present with three core characteristics:
(1) impairment in social interactions, (2) impairment in com-
munication, and (3) stereotyped or repetitive behavior. Thus, a
comprehensive treatment program needs to address these core
Autism and Pervasive Developmental Disorders 249
areas and the individual differences in the expression of related
problems. Elements that must be included as components of a
comprehensive treatment program include child-focused
instruction, family focused support, addressing the multiple
skill domains, and intensive educational instruction (e.g., last-
ing throughout a school day, and occurring across multiple
settings), and the program typically continues over a period of
months or years. Often, a distinction is made between treat-
ment and education for individuals with ASD. This is a false
dichotomy. School, work, home, and community, should all
be seen as the various settings in which comprehensive treat-
ment must be applied. The focus of this section is not to
describe all of the many hundreds of interventions promoted
for individuals with ASD but to discuss the challenges parents
and practitioners face in developing comprehensive treatment
programs. In particular, the myriad of treatment options,
parental stress, and accessibility to services are discussed.
Lastly, because of the extreme heterogeneity in ASD, it is
important to state explicitly that not all individuals with ASD
require, nor desire, treatment intervention.
Treatment Options
When a simple Internet search is performed with 'autism spec-
trum disorders' and 'intervention' or 'treatment' as the search
terms, over two million results are found. Search results run the
gamut from intensive behavior therapy to restrictive diets and
horse back riding. The number of different approaches and
options available for treating individuals with ASD can be over-
whelming for parents and practitioners trying to do the best they
can for their children. Researchers are recognizing this concern
and have begun publishing scientific reviews of the benefits of
different treatment options. Between the research that has been
conducted and the methodology-based systematic literature
reviews, there is convergence and agreement that interventions
based on the principles of the extensive research in basic
learning and the applied research in behavioral approaches to
comprehensive intervention (behavior therapy, applied behav-
ior analysis, cognitive-behavior therapy) are the most effective.
Such reviews have also found that the majority of treatment
approaches promoted for ASD lack empirical evidence of effec-
tiveness, and some have been identified as potentially harmful.
Further, while not all treatment approaches identified as 'inef-
fective' are harmful, they can come at a significant cost with
respect to time commitment that then reduces a family's ability
to devote time to other important activities and also typically
have a significant monetary cost.
Parental Stress
Parents of all children experience varying levels of stress. How-
ever, having a child with special needs has been associated with
increased levels of stress among parents. Further, it has been
reported that parents of children with ASD experience signifi-
cantly higher levels of stress than parents of children with other
disorders. In particular, primary caregivers (e.g., mothers) typi-
cally experience more stress than other family members (e.g.,
fathers, grandparents).
Factors that have been associated with parental stress in
children with ASD range from child-specific characteristics
(e.g., communication difficulties, social impairments, disrup-
tive behaviors) to making decisions about treatment options
for their child. Increased attention has been directed at devel-
oping interventions to reduce parental stress as well as address
the source of the stressors. Interventions to address parental
stress reactions (e.g., cognitive and behavioral programs for
coping and relaxation) and interventions to address source of
stressors (e.g., parent education programs in behavioral man-
agement techniques, provision of information about and help
attaining community resources such as respite care) have been
reported to be effective in reducing overall levels of stress in
parents of children with ASD. Reducing parent stress is an
important, but often neglected, element of treatment for chil-
dren with ASD, and research has suggested that helping parents
alleviate stress can improve treatment outcomes through
greater effective parent participation in intervention programs.
Access to Services
Access to appropriate treatment is often problematic for par-
ents as well as those professionals assisting in implementing
comprehensive intervention. A number of factors may interfere
with access to quality treatment including monetary cost, geo-
graphic location, and availability of qualified providers. The
standards of care and practice, particularly with respect to
educational services, vary significantly across states and within
states.
Families trying to obtain services are faced with the problem
of availability of service providers. With the increase in the
number of individuals needing services, families are often
faced with extensive waiting lists to receive appropriate services
or with the choice of accepting services that are inadequate
with respect to intensity and/or the evidence base for the
intervention offered. Research on treatment for individuals
with ASD has indicated that early and intensive treatment is
associated with better long-term outcomes. However, indivi-
duals and families can be left waiting for months and, in some
cases, even years to receive appropriate treatment.
The most effective interventions for individuals with ASD
are often quite time and resource intensive. Interventions
based on the principles of behavioral approaches have received
the most consistent and extensive research support in addres-
sing the needs of individuals with ASD. However, comprehen-
sive intervention involving this approach for individuals with
severe needs can involve between 25 and 40 h a week of direct
services. Research has demonstrated that such comprehensive
behavioral services are associated with significantly better out-
comes and are more cost-effective in the long term. Currently,
there are no universal standards or regulations in place to
determine who is responsible for the costs associated with
ASD (e.g., insurance company vs. school district vs. family).
See also: Asperger’s Syndrome and Nonverbal Learning Disorder.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental
Disorders: DSM-IV-TR, 4th edn., text revision. Washington, DC: American
Psychiatric Association.
250 Autism and Pervasive Developmental Disorders
Carter AS, Davis NO, Klin A, and Volkmar FR (2005) Social development in autism.
In: Volkmar FR, Paul R, Klin A, and Cohen D (eds.) Handbook of Autism and
Pervasive Developmental Disorders, 3rd edn., pp. 312-334. Hoboken, NJ: Wiley.
Glasberg BA, Martins M, and Harris SL (2005) Stress and coping among family
members of individuals with autism. In: Baron MG, Groden J, Groden G, and
Lipsitt L (eds.) Stress and Coping in Autism. New York, NY: Oxford University
Press.
Guralnick MJ and Neville B (1997) Designing early intervention programs to promote
children's social competence. In: Guralnick MJ (ed.) The Effectiveness of Early
Intervention, pp. 579-610. Baltimore, MD: Brookes.
Kanner L (1943) Autistic disturbances of affective contact. Nervous Child 2:
217-250.
Luiselli JK, Russo DC, Christian WP, and Wilczynski SM (2008) Effective Practices for
Children with Autism: Educational and Behavioral Support Interventions that Work.
New York, NY: Oxford University Press.
Myers SM and Johnson CP (2007) Management of children with autism spectrum
disorders. Pediatrics 120: 1162-1182.
National Research Council (2001) Lord C and McGee JP (eds.) (2001) Educating
Children with Autism. Washington, DC: National Academy Press.
New York State Department of Health, Early Intervention Program (1999) Clinical
Practice Guideline: Guideline Technical Report. Autism/Pervasive Developmental
Disorders, Assessment and Intervention for Young Children (ages 0-3 years),
No. 4217. Albany, NY: NYS Department of Health.
Odom SL, Rogers S, McDougle CJ, Hume K, and McGee G (2007) Early intervention for
children with autism spectrum disorder. In: Odom SL, Horner RH, Snell ME, and
Blacher J (eds.) Handbook of Developmental Disabilities. New York: Guilford.
Romanczyk RG, Arnstein L, Soorya L, and Gillis J (2002) The myriad of controversial
treatments for autism: A critical evaluation of efficacy. In: Lilienfeld, Lohr, and
Lynn (eds.) Science and Pseudoscience in Contemporary Clinical Psychology.
New York, NY: Guilford.
Romanczyk RG, Gillis JM, White S, and DiGennaro F (2008) Comprehensive treatment
packages for ASD: Perceived vs. proven effectiveness. In: Matson (ed.) Clinical
Assessment and Intervention for Autism Spectrum Disorders. Burlington, MA:
Elsevier.
Santangelo SL and Tsatsanis K (2005) What is know about autism: Genes, brain and
behavior. American Journal of Pharmacogenomics 5: 71-92.
Spenheim E (1996) Changing Criteria of Autistic Disorders: A Comparison of the ICD-
10 research criteria and DSM-IV with DSM-lll-R, CARS and ABC. Journal of Autism
and Developmental Disorders 26: 513-525.
The National Autism Center (2009) National Standards Report. Randolph, MA: National
Autism Center.
Volkmar FR, Paul R, Klin A, and Cohen DJ (2005) Handbook of autism and Pervasive
Developmental Disorders. Hoboken, NJ: Wiley.
Wazana A, Breshahan M, and Kline J (2007) The autism epidemic: Fact or artifact?
Journal of the Academy of Child and Adolescent Psychiatry 46: 721 -730.
White S and Roberson-Nay R (2009) Anxiety, social deficits, and loneliness in youth
with autism spectrum disorders. Journal of Autism and Developmental Disorders
39: 1006-1013.
Relevant Websites
http://www.aap.org/healthtopics/Autism.cfm - American Academy of Pediatrics.
http://www.cdc.gov/ncbddd/autism/index.html - Centers for Disease Control and
Prevention.
http://www.nimh.nih.gov/health/publications/autism/complete-index.shtml - National
Institutes of Mental Health.
http://www.nationalautismcenter.org - The National Autism Center.
http://www.researchautism.org - Organization for Autism Research.
Autobiographical Remembering and the Self
J J Skowronski, Northern Illinois University, DeKalb, IL, USA
W R Walker, Winston-Salem State University, Winston-Salem, NC, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
A positivity bias in recall The tendency to remember
positive autobiographical events better than negative
autobiographical events, or to generate false memories for
positive autobiographical events; sometimes reversed in the
service of current self-positivity.
Autobiographical memory A memory system consisting of
episodes recollected from an individual's life, based on a
combination of episodic (personal experiences and specific
objects, people and events experienced at a particular time
and place) and semantic (general knowledge and facts about
the world) memory.
Episodic memory The element of the memory system that
includes specifics about when and where an event occurred,
thought to be accompanied by the sense that an event is
relived or reexperienced.
Infantile amnesia A fairly universal observation is that
many individuals fail to recall much from the time of their
early childhood; it used to be believed that such failure
reflected failures in the structural or processing capabilities
of children, but more recent research suggests that the failure
lies in retrieval failure, not in a failure of storage.
Meaning-making People sometimes try to make the world
meaningful; to do so, they often construct a narrative that
places events into a story-like context. Such narratives have
implications for autobiographical memory: negative events
that are part of narratives that have the theme of overcoming
or redemption may be well recalled because the negative
event is critical to the theme of the story.
Procedural memory The memory that reflects the
acquisition of skills, often thought to be represented in
nonsymbolic forms and to reflect implicit knowledge (e.g.,
knowledge that it not cognitively accessible).
Self-enhancement motivation The motivation that the self
has to view itself in a positive light; can cause negativity
biases in recall when the current self benefits from recalling
autobiographical events in an especially negative light.
Semantic memory The knowledge that is remembered in a
fashion such that it is known, but not reexperienced.
Typically, such knowledge refers to general knowledge about
the world, but it can also refer to knowledge about the self.
Symbolic self The self s capacity to consider itself as a
concept or as an object of consideration.
The reminiscence bump The tendency to recall
autobiographical events from the adolescent and young
adult years; one reason may be that these years are crucial to
the development of the self.
The self A collection of mental structures that comprise an
individual's identity. These structures can be quite diverse
and multidimensional, so that different aspects of the self
can become activated in different contexts or by different
goals. Current views of the self in psychology position the
self as playing an integral part in human motivation,
cognition, and affect.
An issue that has often been the focus of interest about the self
(an ability that reflects humans' possession of a symbolic self )
concerns the nature of the relation between one's personal
identity and memory of one's own personal past. This interest
is evident in the speculations offered by many philosophers.
For example, John Locke directly linked the self to memories of
the personal past ( Essay Concerning Human Understanding;
1975/1689, Book II, Chapter 27). Locke argued that selfhood
existed entirely in such memories: That a person found identity
in the extension of consciousness backward in time, and that
the person's identity reflected those parts of the past that could
be remembered. An extension of this idea is that a person who
remembers nothing of his or her past has no identity.
Interest in the extent to which selfhood is linked to auto-
biographical memory has extended into modern-day psycho-
logical theory and research, and it has led to the development
of a huge and diverse literature. One exciting element of this
literature is that it is truly a reflection of cross-disciplinary
efforts in psychology, with major contributions coming from
researchers in many of psychology's major subfields. This liter-
ature obviously cannot be adequately reviewed in this modestly
sized article. Instead, our goal in this article is to selectively
highlight findings and ideas that illustrate some of the themes
that have emerged from theory and research into the relation
between autobiographical memory and the self.
One of these modern themes pursues the Locke thesis: Is
autobiographical memory necessary for the presence of a sense
of self? However, this is not the only line of research that
pursues links between the self and autobiographical memory.
Considerable work has addressed the extent to which the cur-
rent self directs and affects autobiographical memories to serve
the purposes of the current self, causing them to be remem-
bered, forgotten, distorted, or to even be entirely fabricated.
Yet, other scholarship has focused on the mechanisms under-
lying such effects. For example, one line of work focuses how
an individual's attempts at meaning-making might affect recall
of events from the personal past and how the self-concept
might direct those attempts at meaning-making. Another line
considers the role that social processes might play in the inter-
play between the self and autobiographical memory. The
remainder of this article attempts to provide an entree into
the area of autobiographical memory and the self by describing
just some of the research that has been conducted in each of
these subareas.
251
252 Autobiographical Remembering and the Self
Is Autobiographical Memory Necessary for Individuals
to Experience a Sense of Personal Identity?
We begin by discussing the issue that was raised by Locke: Is
the sense of self, or the sense of personal identity, dependent
on the ability to recall one's personal past? Modern psycholog-
ical scholarship (which is reviewed in more detail by others)
suggests that the answer to this question is 'no.'
Understanding this answer requires a brief overview of how
psychologists currently think about memory. One widely held
view is that long-term memory reflects the action of multiple
systems. For example, one system is thought to contain proce-
dural memories and another is thought to contain declarative
memories. Procedural memory involves learning and retention
in the body's action systems, reflecting the acquisition and
retention of motor, perceptual, and cognitive skills (e.g., know-
ing how to swim). Such knowledge is often represented in
a nonconscious, nonsemantic form. Certainly, while one can
listen to and acquire a description of how to engage in a given
activity, acquisition of such descriptions clearly are not suffi-
cient to learn how to engage in the actions themselves. To learn
how to perform the activity, one has to engage in, and practice
the activity. That ability to do stuff reflects the learning that has
occurred in the procedural memory system.
In comparison, it is thought that the declarative memory
system is the repository of symbolic and semantic knowledge
about the world. Tulving, among others, argues that the declar-
ative memory system contains two subtypes of information
(perhaps residing in separate subsystems): semantic memories
and episodic memories. Semantic memories reflect relatively
generic knowledge about the world (e.g., DeKalb is a city in
Illinois). Such memories typically are not accompanied by
recall of when or where the memory was acquired: rather,
they are retrieved without such information.
In contrast, episodic memories contain such specifics.
Examples of episodic memories (e.g., from one of the chapter
authors) are a memory of hitting a walk-off home run in a little
league baseball game and a memory of throwing a hanging
curveball in a college baseball game that was hit by an oppos-
ing batter for a walk-off home run.
The key to these memories being classified as episodic is that
they have a 're-living' quality that includes the notion that the
events were experienced by the self at a particular point in space
and time. Certainly, a rememberer can 'know' things about
themselves in a way that reflects semantic memory. However,
that same knowledge can also be recalled in a way that, at least
to some extent, provides a sense of reexperiencing the event.
Such a memory is exemplified by Proust when he writes about
an encounter with a Petit Madeline given to him by his mother.
Proust indicates that when his mother offered him the little
cake, he retrieved 'a feeling of inexplicable happiness' and a
childhood memory of Saturday mornings in Combray where
his Aunt Leone used to give him a Madeleine after first dipping
it in her own cup of tea. Proust describes such a memory by
saying that "It is a complete fragment of the past, with its
original 'perfume,' that is for a moment given back to us." By
definition, episodic memories have this 'reliving' quality.
In modern terminology, Locke's assertion about the neces-
sary link between memory for the personal past and a sense of
identity can be framed in terms of episodic memory: Is the
ability to retrieve such memories crucial to an individual's
sense of identity or a sense of self? The answer appears to be
'no.' A moment's reflection should reveal why this should be
the case. The semantic memory system can contain a lot of self-
relevant information that is important to the sense of self.
While some of this information might reflect the symbolic
system's encoding of self-events (I know that I hit a home
run), other information may reflect higher-order conceptual
knowledge (e.g., I believe that I am lazy; other people think 1
am driven; I am a golfer). Hence, it seems reasonable that
disruption of the ability to 'relive' the specific events contained
in episodic memory should have no effect on those compo-
nents of self-knowledge that reflect semantic memory.
Indeed, this proposition is now supported by considerable
research. Even people who exhibit total amnesia for events from
their personal past can describe their own personal characteris-
tics both reliably and accurately. For instance, Clive Wearing, an
accomplished musician who suffered viral encephalitis-induced
episodic memory loss in 1985, has retained both his musical
ability and his personality, including a profound and often-
professed love for his wife. Such observations would seem to at
least partially refute Locke's notion that one cannot have a sense
of self without access to autobiographical memories (or, at least,
to the episodic variety).
However, the fact that episodic memories are not necessary
for the possession of a sense of self does not mean that such
memories are unimportant to the sense of self. For example, as
illustrated by their observations of the patient D.B. Klein, et al.
note that episodic memory seems to be required for a sense of
continuity in personal identity across time.
Another way in which the autobiographical memory is
important to the self lies in the extent to which the self uses
memory to maintain and promote the current self. For exam-
ple, the Self-Memory System outlined by Conway proposes
that the self-system works to maintain the seifs 'status quo,'
as well as working to maintain the extent to which the self
perceives goal coherence. Conway speculates that control pro-
cesses of the working self may act on the stored corpus of
autobiographical event knowledge in the service of such main-
tenance, doing so by means of editing memory content and by
generating false memories. Such ideas imply that the self plays
an important role in the ability to recall events from one's
personal past. It is to this idea that we turn our attention in
the next section.
Does the Self Direct, Affect, or Alter the Content of the
Autobiographical Memories That Can Be Recalled?
Event Valence and Memory for the Personal Past
One example of this notion of selectivity lies in the potential
relation between event valence and recall, sometimes expressed
in the widespread belief that people remember the past
through 'rose-colored glasses.' In more precise terms, this idea
suggests that there may be a positivity bias in the content of
memory such that when trying to remember the events of their
lives, people's memories tend to retain the positive and elimi-
nate the negative.
The hypothesis that such a positivity bias may emerge in
autobiographical memory can be generated from modern
Autobiographical Remembering and the Self 253
thinking about the self. The argument is that the powerful moti-
vations of self-enhancement and self-protection, as well as both
cognitive tendencies and motivational tendencies toward self-
consistency for those who think positively about themselves,
should generally work to promote recall of positive autobio-
graphical memories and to diminish recall of negative autobio-
graphical memories. Moreover, psychologists have detailed a
large set of mental mechanisms that can work to produce such
positivity biases in recall. Indeed, the results of many studies
have evinced some form of a positivity bias. Such results induced
Walker, Skowronski, and Thompson to confidently declare that
"Life is pleasant - and memory helps keep it that way."
These biases can be manifest in different ways. One mani-
festation of the positivity bias is that people tend to selectively
remember the positive and selectively forget the negative. For
example, D'Argembeau and Van der Linden asked participants
to recall positive and negative events that involved self-
evaluations (i.e., pride and shame) and positive and negative
events that involved evaluations about others (i.e., admira-
tion and contempt). The results showed that for events that
involve self-evaluations, people reported a subjective remem-
brance of positive events with more details than they did
negative events; this outcome did not occur for events that
involved evaluations of others. A study by Skowronski, Betz,
Thompson, and Shannon similarly found that people exhib-
ited positivity biases in self-reported self-memory that were
not evident in memories about others (in fact, the memories
about others evinced a negativity bias).
A second way in which the positivity bias in memories of
the personal past can be manifest relies on distortions in mem-
ory. In one type of distortion, events are misrecalled to be more
positive (or less negative) than warranted when compared to
the objective historical record. One example comes from Bah-
rick, Hall, and Berger, who reported that college students
remembered more As than were present on their high school
transcripts. A second example comes from a study conducted
by Croyle, Loftus, Barger, Sun, Hart, and Gettig, the results of
which suggested that people misrecalled (minimized) both
their cholesterol scores and the cardiovascular risk categories
into which they were placed; those who received the most
undesirable test results were most likely to evince evidence
of this distortion (for results with similar implications, see
Christensen, Wood, & Barrett).
These memory tendencies seem to be present at an early
age. For example, Wilson, Smith, H.S. Ross, and M. Ross inter-
viewed pairs of siblings to assess memories for their sibling
disputes. Although describing the same episodes, both older
(M age = 7.0) and younger (M age = 4.4) siblings ascribed
more serious transgressions to the other than to themselves.
However, a thorough review suggests that the autobio-
graphical memory data are not always entirely congruent
with the presence of a hedonic valence effect in recall. Such
inconsistency emerged in an early review of the literature con-
ducted by Meltzer and again in a later review reported by
Matlin and Stang. Indeed, modern studies that look for evi-
dence of a positivity bias in memory sometimes continue to
fail to find evidence of the effect.
Additional pessimism about consistency of the positivity
effect in memory comes from results provided by researchers
who have studied the relation between trauma and memory.
One of these researchers suggests that the available evidence
shows that traumatic events are essentially unforgettable. How-
ever, results from such studies are not definitive: They cannot
rule out the possibility that, despite their negativity, such events
are well recalled because of their extremity. Indeed, studies con-
sistently show that event extremity is much more strongly pre-
dictive of recall than event valence. Nonetheless, such results do
pose a challenge to the idea that self-related processing of auto-
biographical events will tend to promote positivity in recall.
Skowronski notes that such inconsistencies can be at least
partially resolved by a focus on the mental structures, processes,
and mechanisms that are involved in processing autobiograph-
ical memories. Such an idea is wonderfully exemplified by the
work of Wilson, M. Ross, and their colleagues. The work pro-
duced by these researchers suggests that the self-enhancement
and self-protective motives that often serve to maintain the
positivity of the current self can sometimes produce a negativity
bias in memory for past events. The Wilson and M. Ross team
suggests that this can occur as a consequence of the meaning-
making activities in which people engage as events are retro-
spectively considered and rehearsed.
For example, one way in which one might glorify one's
current ability to play golf is to remember how bad one was
at the activity when one started to play the game. A parable can
be constructed from such events, one that says: "Look at how far
I've come - I never broke 1 20 when I started, but now I break 90
almost every time." This exact effect has been observed in the
research conducted by the Wilson and M. Ross team: People
reconstruct their memories of their early performance levels,
and such reconstructions often reflect performance that is
worse than is objectively suggested by the historical record.
Similar results were reported by Dewhurst and Marlborough,
who reported that students who surpassed their target grade
sometimes exaggerated how anxious they had been prior to the
exam. Echoing the conclusion of the Wilson and M. Ross team,
Dewhurst and Marlborough attributed these findings to the
action of the motives of self-enhancement and self-protection
that bias the recall of preexam anxiety in the direction that
maximizes current self-esteem.
The implication of such research is that the self-enhancing
or self-protective goals of accentuating the positive do not
always eliminate memory for the negative - in the right cir-
cumstances, when it benefits the current self, these goals can
actually maintain, enhance, or distort memory for negative life
events. However, given the way that memory becomes framed
over time, remembering the negative actually has at least one
positive effect - it makes the current self feel good. More
important to the current article, though, is the idea that regard-
less of whether they are positive or negative, these memory
biases occur because of the influence of the self. The consistency,
protection, and enhancement motives that are activated by the
current self in a given situation cause differential cognitive
treatment of memories of an individual's personal past. It is
this differential treatment that is responsible for biased recall of
autobiographical information.
The Reminiscence Bump
Other data link the self to recall by showing that the extent to
which an event is important to the self is related to memory for
254 Autobiographical Remembering and the Self
the event. A fine example of this effect comes from research
examining the reminiscence bump. The reminiscence bump
effect shows that when recalling events from their personal
past, people tend to produce a disproportionately high propor-
tion of memories for events that occur roughly between the
ages of 10 and 30. Presumably, this is a time that is crucial to
the emergence of an individual's self-concept, and the events
that are remembered during such periods may bear special
significance to the self.
Consistent with this view, one study has shown that bump
memories were found to be more novel, more important for
identity development, more distinct, and more likely to
involve developmental transitions than memories from other
age periods. The content of the memories in the bump period
is more likely to be positive than negative, an outcome that
arguably is also consistent with the action of the self on mem-
ory. Other studies have shown that reminiscence bumps are
not always restricted to their typical age range, but can emerge
during any period of substantial life transition that is likely to
cause significant changes to an individual's self-conception.
Thus, as with the valence biases in memory, these studies
document the fact that memory of the personal past is strongly
related to an individual's identity.
Mechanisms that May be Related to the Recall of
Events from the Personal Past
Meaning-Making
What are some of the mechanisms that might be responsible
for the relations between the self and memory of the personal
past? Clearly, there are a host of cognitive mechanisms that
may be relevant. However, one mechanism that has received
considerable recent empirical attention reflects the cognitive
efforts that one undertakes to make sense of one's life. Such
meaning-making seems to be a routine activity, and as such,
has the potential to affect memory for many negative autobio-
graphical events that might become incorporated into the
life story.
Such processing is exemplified by the results of a recent
study by Ritchie, Skowronski, Hartnett, Wells, and Walker.
This paper examined events that changed their affective tenor
over time, moving from negative to positive, or vice versa. The
authors argue that such changes might occur because of the
way in which events become reinterpreted in the light of
subsequent life events. For example, consider the plight of a
person who had been denied her tenure in her first academic
job. Such an event might have been crushing when it hap-
pened, but later, as the person's life progressed, the event
might be viewed in a different light. For example, the person
might have obtained a second job that was a perfect fit for
them. As the person reflects on and tries to understand her own
life, the tenure denial (I was crushed) may come to be viewed
positively (in the end, it was the best thing that could have
happened to me).
In such a case, one would likely not expect memory for the
original event to be diminished. In fact, one might expect that
because of the high relevance of the entire event sequence,
memory for the tenure denial event might actually be quite
good and might strongly persist over time. Indeed, the story
becomes even more important to the self, (e.g., illustrating self-
resilience) if the crushing nature of the tenure denial is
emphasized.
Certainly, then, mental activity such as meaning-making
and event reprocessing might be responsible for the fact that
not all negative events are poorly recalled. When such negative
events are essential to a life story that helps to ascribe meaning
to events, those negative events may be recalled quite well.
These considerations suggest that, more generally, a full under-
standing of how the self affects autobiographical memory
necessitates an understanding of how an event fits into the
personal narratives that people use to make sense of their
lives (an idea anticipated by Singer).
Social Mechanisms
However, it is also the case that these meaning-making activ-
ities may not solely be for private consumption. People also
need to explain themselves to others, and the need to concoct
such public explanations may affect the meaning-making
process.
Consistent with this proposition, data indicate that people
often relate autobiographical event descriptions to one another
in the course of those conversations. Moreover, some data
suggest that social rehearsals occur at least as often as, and
perhaps more often than, other rehearsal types. For example,
in each of the two studies reported by Walker et al., participants
listed autobiographical events that occurred within the last
6 months. After listing the events, participants were asked to
estimate the number of times that they had rehearsed each
event for one of the several different reasons (e.g., to remember
the details of the events, to reexperience the emotions asso-
ciated with the events, to better understand the events, to talk
about the events to others, or the events were thought of
involuntarily). In both studies, the most frequent reason that
people rehearsed events was for the purpose of talking to
others, and its reported frequency was significantly greater
than any other reason.
Consideration of the social underpinnings of autobio-
graphical event rehearsal adds additional complexity to the
study of autobiographical memory, for different motives may
drive the public presentation of different kinds of memories.
Certainly, one might expect the need for others to view the self
positively to motivate the selective self-presentation of positive
events over negative events, a process that might be expected
to push autobiographical memory toward positivity. However,
sometimes in social circumstances, one needs to tell an inter-
esting story, and stories can be especially entertaining when
they involve themes such as overcoming obstacles (a theme
that often appears from studies of meaning-making). More-
over, sometimes one can ultimately gain benefit by presenting
one's foibles to others (e.g., self-deprecating humor might
serve the purpose of ingratiation), which suggests that, at
least under some circumstances, one might be motivated to
relate one's negative life experiences to others. Thus, as in the
M. Ross research, it may be that enhanced recall of negative
personal life events may be a consequence of the need to
enhance the current positivity of the self; that need sometimes
induces rehearsal of negative events for the purpose of relating
those events to others.
Autobiographical Remembering and the Self 255
However, the effects of public event disclosures may go
beyond the moderation of positivity effects in recall. For exam-
ple, Skowronski and Walker suggest that other elements of social
interactions that might affect autobiographical public self-
presentations are the Grice conversational norms to be relevant,
to be informative, and to be brief. These norms should not only
work toward selectivity in the rehearsal of autobiographical
memories, but they might also work toward distorting their
content. This can happen in two ways. The norm to be brief
may cause people to omit many details from their event descrip-
tions; such omissions might cause those details to be perma-
nently inaccessible when the memory is retrieved. Moreover,
people may sometimes add event details, or fabricate events
entirely, in the service of telling a good story. When the story is
told often enough, a storyteller may experience a false memory
because of a failure to accurately engage in reality monitoring:
The storyteller may ultimately lose track of the fact that an event
element was self-fabricated, coming to believe that the fabricated
element was true (though this may not always occur).
The social nature of remembering has also been explored
from a developmental perspective. In fact, the themes of
social interaction and narrative construction have been
remarkably combined in a program of research that has exam-
ined the emergence of the self as it related to the emergence
of the capacity to produce early autobiographical memories.
Until the middle 1980s, it was widely believed that autobio-
graphical memory emerged relatively late in development. One
example of this came from the literature on literature detailing
infantile amnesia. Observations suggested that this amnesia was
widespread. These observations led scholars to believe that such
children did not report such memories simply because they
lacked the cognitive mechanisms to do so. This apple cart was
upset by research that began to emerge in the late 1970s.
Researchers in these studies began to ask 3- and 4-year-old
children to report what happens in the course of everyday events
and routines. The children could do so, and readily provided
rich reports of their experiences. As it has evolved, this research
has now demonstrated conclusively that young children do
have considerable memory for significant life events.
There have been a number of intriguing lines of research that
have emerged from these examinations of early autobiographi-
cal memory. One of these lines of research links frequent social
discourse with parents to both the emergence of autobiograph-
ical memory reporting capabilities and to the emergence of a
personal identity. The scholars in this area argue that through
participating in joint reminiscing with parents, young children
develop a sense of subjective perspective on their personal past.
These scholars argue that, even more importantly, through
participating in adult-guided conversations about their past
that are replete with information about their inner mental life,
children come to develop a subjective sense of self.
Part of this thesis revolves around the idea that reminiscing
is not just about what happened, but what happened to the
actor (the 'me'). This necessitates distinguishing what hap-
pened to the self from what happens in the world at large. It
is thought that a child's interaction with parents about the
child's own life events aids the development of this self-
perspective, especially to the extent that such activity promotes
the inclusion of the child's thoughts, feelings, and evaluations
of the past in their autobiographical memories.
In a rather ironic sense, these scholars have moved back, at
least a bit, toward Locke's idea that autobiographical memory
is essential to the sense of self. While we have argued that
considerable data suggest that autobiographical memory (at
least in its episodic form) does not appear to be necessary for
adults to maintain a sense of identity, those data do not pre-
clude the possibility that the recall of the personal past as is
reflected in child discussions with parents is important to the
development of a personal identity. It should be emphasized
that it seems to be the social nature of these interactions, not
the emergence of language itself, that is the driving force
behind these findings. Indeed, developmental psychology
scholars have provided impressive evidence that links such
interactions to personal identity. For example, considerable
data now show that maternal style early in development has
a substantial effect on children's emerging ability to recall their
own past, and that this effect emerges over a relatively long
developmental period. For example, data provided by Reese,
Haden, and Fivush, showed that the ways in which mothers
reminisce with their young preschool children continued to
affect children's memory skills more than 2 years later.
One important idea outlined by these scholars is the notion
that it is discourse with caregivers that contributes to the devel-
opment of a sense of time in autobiographical memory. That
is, it is thought that talking about experienced events with
parents whose narrative styles induce a child to incorporate
memory fragments into coherent narratives of the past pro-
vides not only a way for the child to organize memory for
future recall, but also a scaffold for understanding the order
and specific locations of personal time, an element that is
essential to autobiographical memory.
This temporal element is also essential to the sense of self.
As noted earlier in this article, one contribution that autobio-
graphical memory does make to the self is that it helps to
provide information about both stability and change in the
self. Thus, the content of the self is closely connected to the
sense of a personal past, and discourse with caregivers can help
children to develop the understanding that it was the same self
that exists in the present that experienced an event in the past.
Moreover, the ability to report autobiographical events is
closely tied to evidence that a child has acquired a sense of
self. For example, Harley and Reese measured children's self-
recognition with the mirror test at age 1 Vi years and used it as
a predictor of children's memory reports with mothers and
experimenters at ages 2 and 2Vi years. Children's memory
reports with mothers and experimenters varied as a function
of their earlier self-recognition skill, with early recognizers
engaging in more interactions with mothers and with experi-
menters over time. Additional analyses showed that the initial
importance of self-recognition was mediated by maternal remi-
niscing style when children were 2 years old: Children who
recognized themselves earlier in the mirror test had mothers
who responded with increased elaborations in past event con-
versations with children at age 2 years, and maternal reminis-
cing style directly predicted children's later memory reports.
In summarizing the 2002 'state of the art,' Reese argued that
the data collected up to that point showed that social factors,
such as parental reminiscing style, interacted with children's
self-understanding to produce autobiographical memory. She
also argued that such effects occurred for at least two reasons.
256 Autobiographical Remembering and the Self
First, parents' talk about personally relevant past events may be
an important source of information for children's self-under-
standing. Second, children may not be able to benefit from
increased self-understanding in their retention of autobio-
graphical memories unless they have the added elaborative
and evaluative support from their parents in understanding
past events. Such ideas clearly support the contention that
one must incorporate social variables when attempting to
understand the relations between the self and autobiographical
memory.
Coda
Scholars who are interested in the relation between the self and
autobiographical memory have a daunting task. This task
begins with gaining an appropriate understanding and appreci-
ation of the philosophical issues relevant to the topic, repre-
sented in this article by the works of Locke and Grice. This task
continues by keeping abreast of developments in the area,
which means monitoring literatures across many of psychol-
ogy's subareas, especially social psychology, developmental
psychology, cognitive psychology, personality psychology,
clinical psychology, and neuroscience. Finally, to contribute
empirical scholarship in this area, one needs to be open to
multiple methodologies and analytic techniques. The informa-
tion presented in this article merely samples the field and is
obviously substantially influenced by the research programs,
interests, and explanatory stories preferred by the article's
authors. While the task is daunting, it is likely that their payoff
is substantial. For one, assimilating the scholarship in this area
will contribute to a broad view of psychology and will perhaps
allow one to better appreciate the methods, areas, and
approaches that might characterize the subareas that differ
from one's own specialty. Moreover, we believe that the inter-
disciplinary nature of the area continues to make it a fertile
ground for scholars who want to pursue new avenues of
research into this relation. We hope that this article may serve
to stimulate interest in such research, and we look forward to
examining the products of such research in the years to come.
See also: Agnosia (including Prosopagnosia and Anosognosia);
Amnesia; Confabulation and Reality Filtering; Episodic and Semantic
Systems of Autobiographical Memory; Episodic Memory; Eyewitness
Identification; Implicit Memory; Memory; Self-Efficacy; Self-Esteem;
Semantic Memory.
Further Reading
Bauer PJ (2007) Remembering the Times of Our Lives: Memory in Infancy and Beyond.
Mahwah, NJ: Lawrence Erlbaum.
Beike DR, Lampinen JM, and Behrend DA (eds.) (2004a) The Self and Memory.
New York: Psychology Press.
Berntsen D (2009) Involuntary Autobiographical Memories: An introduction to the
Unbidden Past. Cambridge: Cambridge University Press.
Courage ML and Cowan N (eds.) (2009) The Development of Memory in Infancy and
Childhood. Hove, UK: Psychology Press.
Fivush R and Haden CA (eds.) (2003) Autobiographical Memory and the
Construction of a Narrative Self: Developmental and Cultural Perspectives.
Mahwah, NJ: Erlbaum.
Harris CB, Sutton J, and Barnier AJ (2010) Autobiographical forgetting, social
forgetting, and situated forgetting: Forgetting in context. In: Della Sala S (ed.)
forgetting, pp. 253-284. New York: Psychology Press.
Kihlstrom JF, Beer JS, and Klein SB (2002) Self and identity as memory. In: Leary MR
and Tangney J (eds.) Handbook of Self and Identity, pp. 68-90. New York:
Guilford Press.
Klein SB, German T, Cosmides L, and Gabriel R (2004) A theory of autobiographical
memory: Necessary components and disorders resulting from their loss. Social
Cognition 22: 460-490.
Libby LK and Eibach RP (2007) How the self affects and reflects the content and
subjective experience of autobiographical memory. In: Sedikides C and Spencer SJ
(eds.) The Self, pp. 75-91. New York: Psychology Press.
McNally RJ (2003) Remembering Trauma. Cambridge, MA: Belknap Press/Harvard
University Press.
Nelson K and Fivush R (2004) The emergence of autobiographical memory: A social
cultural developmental theory. Psychological Review 111: 486-511.
Pasupathi M (2001 ) The social construction of the personal past and its implications for
adult development. Psychological Bulletin 127: 651-672.
Rubin DC (ed.) (1996) Remembering our past: Studies in autobiographical memory.
New York: Cambridge University Press.
Thompson CP, Skowronski JJ, Larsen SF, and Betz A (1996a ) Autobiographical
Memory: Remembering What and Remembering When. New York: Erlbaum.
Walker WR, Skowronski JJ, and Thompson CP (2003) Life is pleasant — And memory
helps to keep it that way. Review of General Psychology 7: 203-210.
Avoidant Personality Disorder
C A Sanislow, E E Bartolini, and E C Zoloth, Wesleyan University, Middletown, CT, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Anhedonia A mental health symptom, commonly seen in
mood disorders and schizophrenia-related disorders,
involving the inability to experience pleasure from enjoyable
activities.
Comorbidity The presence of two or more psychiatric
diagnoses found simultaneously in the same individual.
Comorbidity may reflect the presence of two distinct
pathologies, but may also reflect imperfect diagnostic
categories.
Distress disorders A higher order classification of the
broad range of anxiety and depressive disorders.
DSM The Diagnostic and Statistical Manual of
Mental Disorders (current version: Diagnostic and
Statistical Manual of Mental Disorders, 4th revision,
text revision, or DSM-IV-TR) is published by the
American Psychiatric Association and provides a
multiaxial classification system for mental disorders,
with personality disorders such as APD listed under
Axis II disorders.
FFM The Five Factor Model of personality describes five
broad personality domains: openness, conscientiousness,
extraversion, agreeableness, and neuroticism. Each domain
has six facets descriptive of lower-order personality traits.
Generalized social phobia A subtype of social phobia
marked by a fear of social situations across all aspects of the
social world. GSP is very closely linked to APD.
ICD The International Statistical Classification of Diseases and
Related Health Problems (current version: International
Classification of Diseases, 10th Revision, Clinical
Modification, or ICD-10-CM) is published by the World
Health Organization and provides a classification system for
a broad range of medical symptoms and conditions,
including mental health diagnoses. It is widely used
internationally but is less common within the United States.
SASB Structural Analysis of Social Behavior, a model of
personality based on interpersonal theory that describes
patterns of behavior, suggests origins of maladaptive
interpersonal behavior, and guides psychotherapy process.
Social anxiety disorder The term introduced in the DSM-IV
as an alternative name for social phobia.
Social phobia A disorder marked by an enduring fear of
social situations and the embarrassment that may result
from participation in social interactions.
Introduction
When does anxiety in interpersonal situations become patho-
logical, and what makes it so? Clearly, functional impairment
is an important element. For personality disorders in general,
there is the assumption that there are clinically significant
deficits in both the self and interpersonal relationships. For
avoidant personality disorder (APD) self-definition encom-
passes a desire for affiliation hobbled by a sense of personal
inadequacy, and intense fears of interpersonal rejection stem-
ming from a heightened sensitivity to criticism from others
is another key feature, leading to social detachment that is
perhaps the most obvious clinical feature of APD. In broader
terms, the feelings of the person suffering APD have been
described as anxiety, and the behaviors as shyness, suggesting
overlap with other forms of psychopathology. Historically, one
thing that differentiates APD from other forms of psychopa-
thology, however, is that APD is characterized by isolative
interpersonal behaviors combined with a desire to be close to
others that is stilted by fears of rejection.
Diagnosis
APD effects about 1-2% of the general population, and is
frequently present among patients in outpatient psychiatric
clinics, often ranging between 10% and 20% of the cases.
Personality pathology corresponding to APD is included in
the two major diagnostic manuals for psychiatric diagnosis of
mental disorders: the tenth edition of the International Statistical
Classification of Diseases, or the 'ICD- 10/ which is published
by the World Health Organization, and the fourth edition of
Diagnostic and Statistical Manual of Mental Disorders, or the
'DSM-IV-TR,' which is published by the American Psychiatric
Association. Both diagnostic manuals specify avoidant person-
ality pathology as chronic and enduring. The actual diagnosis
'APD' is listed among DSM-IV-TR personality disorders,
whereas the corresponding diagnosis in ICD- 10 is 'Anxious
(Avoidant) Personality.' The DSM-IV-TR diagnostic criteria are
shown in Table 1, and criteria from the ICD- 10 in Table 2. (The
'TR' in DSM-IV-TR stands for 'Text Revision.' The fourth edition
of the DSM, published in 1994, was updated in 2000 by revi-
sions to the 'text' describing some diagnoses; however, none of
the diagnostic criteria were altered).
DSM Definition and Criteria
In the DSM-IV-TR, APD (code number 301.82) is characterized
by a ". . .pervasive pattern of social inhibition, feelings of
inadequacy, and hypersensitivity to negative evaluation. . ."
(p. 718). The diagnostic title 'Avoidant Personality Disorder'
was first introduced into official diagnostic nomenclature with
the DSM-III, which was published in 1980. Within the DSM
(DSM-III through DSM-IV-TR) Axis II, APD resides in 'Cluster C'
of the Personality Disorders, along with Dependent and
257
258 Avoidant Personality Disorder
Table 1 DSM-IV-TR diagnostic criteria for avoidant personality
disorder
A pervasive pattern of social inhibition, feelings of inadequacy, and
hypersensitivity to negative evaluation, beginning by early adulthood and
present in a variety of contexts, as indicated by four (or more) of the
following:
(1 ) avoids occupational activities that involve significant interpersonal
contact, because of fears of criticism, disapproval, or rejection
(2) is unwilling to get involved with people unless certain of being
liked
(3) shows restraint within intimate relationships because of the fear of
being shamed or ridiculed
(4) is preoccupied with being criticized or rejected in social situations
(5) is inhibited in new interpersonal situations because of feelings of
inadequacy
(6) views self as socially inept, personally unappealing, or inferior to
others
(7) is unusually reluctant to take personal risks or to engage in any
new activities because they may prove embarrassing
Source: American Psychiatric Association (2000) Diagnostic and Statistical Manual
for Mental Disorders, 4th edn., text revision, p. 721. Washington, DC: American
Psychiatric Association.
Table 2 ICD-1 0 diagnostic criteria for anxious (avoidant)
personality disorder
Anxious (avoidant) personality disorder is characterized by at least four
of the following:
(1) Persistent and pervasive feelings of tension and apprehension
(2) Belief that oneself is socially inept, personally unappealing, or
inferior to others
(3) Excessive preoccupation about being criticized or rejected in
social situations
(4) Unwillingness to get involved with people unless certain of being
liked
(5) Restrictions in lifestyle because of need of security
(6) Avoidance of social or occupational activities that involve
significant interpersonal contact, because of fear of criticism,
disapproval, or rejection
Source: World Health Organization (1992) The International Statistical Classification of
Disease (ICD), 10th edn., pp. 155-156. Geneva: World Health Organization.
Obsessive-Compulsive Personality Disorders. Personality dis-
orders in this cluster share features of internalizing distress
including anxiousness and fearfulness. As is the case with all
personality disorders in the DSM-IV-TR, there is for APD the
general requirement of . .an enduring pattern of inner experi-
ence and behavior that deviates markedly from the expectations
of the individual's culture and is manifest in at least two of the
following areas: cognition, affectivity, interpersonal function-
ing, or impulse control. . (p. 686).
Of the seven criteria for the DSM-IV-TR APD shown in
Table 1, at least four are required for a patient to meet the
diagnostic threshold. The first criterion, 'avoids occupational
activities that involve significant interpersonal contact, because
of fears of criticism, disapproval, or rejection,' may apply to
work or school. Often a person suffering from APD will limit
their educational opportunities, and they are likely to choose
an occupation where interpersonal contact is minimal. They
will also avoid working in teams, preferring to do things them-
selves. This is to be distinguished from the perfectionism aspect
in obsessive-compulsive personality disorder, in which people
prefer to work alone because their standards are so high they
believe that others would bring the work down.
The second criterion, 'is unwilling to get involved with
people unless certain of being liked,' is based on a preoccupa-
tion with feelings of inadequacy. This preoccupation essen-
tially creates a cognitive disturbance that interferes with
interpersonal interactions in novel social situations. An avoi-
dant person will deal with this by being reticent and reluctant
to 'make the first move.' This is unlike the case of dependent
personality disorder(DPD) where a person is reticent based on
fears that differentiating their opinions from another will
destroy or otherwise damage the relationship. For avoidants,
the fear is the confirmation of their own perceptions of
inadequacy.
The third criterion, 'shows restraint within intimate rela-
tionships because of the fear of being shamed or ridiculed,' is
also driven by the fear of being criticized. Being avoidant
means always being petrified about what is to them, 'deep,
dark, secrets' that, if revealed, would almost certainly invite
ridicule. Even a slight teasing might be perceived as humil-
iating. For these reasons, APD is often associated with a great
deal of secretive behavior that is, for the most part, irrational,
except for the fact that it brings a needed security even in
the 'closest' relationships. In contrast, such restraint observed
at the clinical surface with DSM-IV-TR 'Cluster A' (odd, eccen-
tric) personality disorders such as schizoid, schizotypal, or
paranoid, is better attributed to an indifference to intimacy
(schizoid) or paranoid fears underlying secretive behavior
(schizotypal and paranoid).
The fourth criterion, 'is preoccupied with being criticized
or rejected in social situations,' at the first blush may appear
similar to the DSM-IV-TR Axis I clinical syndrome of social
phobia. Clearly, anxiety in a social situation is the hallmark
of this criterion. For avoidants, any aspect of their being is
fair game, from their hairstyle and clothing to their ideas and
intellect. In contrast to social phobia, anxiety associated with
avoidant personality disorder is not situation specific, but
wide-ranging across all areas of the social world. It is not,
for example, limited to situations where they might be called
on to speak in front of others. A variant of social phobia,
generalized social phobia, interestingly appears to diagnosti-
cally cooccur with greater frequency than the simple type,
and will be considered in greater detail in a later portion of
this article.
The fifth criterion, 'is inhibited in new interpersonal situa-
tions because of feelings of inadequacy,' represents more
than mere cautiousness. Often it involves social comparisons
that may have no basis in reality, for instance, the feeling that
others whom the avoidant person is with are smarter, more
successful, have better relationships, better family life, a better
job, etc. Again, rooted in this criterion is the idea of being 'less
of a person' across multiple domains. Here again, with APD,
the person will unfortunately hold back information about
himself or herself, information that might actually lead to the
experience of being liked or respected by others if it weren't for
this inhibition or competing cognitive process of evaluation of
self and others.
Avoidant Personality Disorder 259
The sixth criterion, Views self as socially inept, personally
unappealing, or inferior to others,' is a pervasive inferiority
complex that includes the beliefs that one is not at all likeable
and possesses no social skills. The insecurity that accompanies
these feelings of awkwardness is paired with the belief that
others will always judge or otherwise evaluate them negatively.
Here, it is important to distinguish this criterion from cognitive
features of depressive disorders. For depressive disorders,
whether they are shorter in duration but intense, such as the
case of major depression, or longer in duration but less intense,
such as dysthymic disorder, the key distinction is that in APD,
these feelings of ineptness and being unlikeable are indepen-
dent of depressive mood states. In other words, the depressed
mood may come and go or otherwise vacillate, but with APD,
the inferiority remains constant.
The seventh criterion, 'is unusually reluctant to take per-
sonal risks or to engage in any new activities because they may
prove embarrassing,' translates to 'never be the first to express
deeper feelings.' Thus, the risk is not about 'risk-taking behav-
ior' as might be found in people with antisocial tendencies or
thrill seekers. For APD, the risk is much more pedestrian and
mundane. The person will be loath to reveal feelings that may
expose him or her, whether positive or negative. Even if some-
one has a positive feeling, the typical approach would be to
wait for the other to share the feeling or observation and then
to agree with that observation rather than risk the certain
humiliation that would come with a 'stupid' disclosure that
was 'off the mark.'
ICD Definition and Criteria
The ICD- 10 diagnosis, anxious (avoidant) personality disorder
(code number F60.6) is very similar to that of DSM-IV-TR,
despite the slight variation in the title. As with the DSM-
IV-TR, anxious PD features must meet the accepted general
criteria for personality disorder including characteristic and
enduring disturbances in a person's inner experience and be-
havior that is outside the accepted norm. Disturbances in spe-
cific processes that impede functioning may include cognition,
affect, impulse control, and interpersonal relatedness. In the
ICD- 10, the behaviors must also be seen as maladaptive and
inflexible, evidence personal distress or otherwise have an
adverse impact, and be of long duration and not attributable
to other mental or organic disorders.
Comparing the APD criteria from the ICD- 10 to the DSM-
IV-TR reveals many similarities and a few differences. On the
whole, the criteria appear to be tapping an isomorphic con-
struct. Both require four criteria for a diagnosis, though the
ICD- 10 has total of six possible criteria compared to seven in
the DSM-IV-TR. Comparing Tables 1 and 2, reveals an almost
direct correspondence of 1CD-10 to DSM-IV-TR criteria (ICD-
10 #2 to DSM-IV-TR #5, ICD- 10 #3 to DSM-IV-TR #4, ICD-
10 #4 to DSM-IV-TR #2, and ICD- 10 #6 to DSM-IV-TR #1). Of
the two remaining ICD-10 criteria (#s 1 and 6), criterion #6,
'Restrictions in lifestyle because of need of security,' roughly
approximates DSM-IV-TR criterion #7, 'is unusually reluctant
to take personal risks or engage in new activities because they
may prove embarrassing,' and may also partially capture DSM-
IV-TR criterion #6, 'is inhibited in new interpersonal situations
because of feelings of inadequacy.' A subtle difference is that
the ICD-10 focuses on 'security' which is in contrast to the
DSM-IV-TR focus on 'embarrassment' or 'feelings of inade-
quacy.' Finally, the first ICD-10 criterion #1, 'Persistent and
pervasive feelings of tension and apprehension,' appears to tap
a more general construct of anxiety. In contrast, DSM-IV-TR
attempts to focus more specifically on anxiety related to a
negative concept of self.
In the version of the ICD-10 updated in the year 2007, there
is a general description of anxious [avoidant] personality dis-
order that reads as follows:
Personality disorder characterized by feelings of tension and appre-
hension, insecurity, and inferiority. There is a continuous yearning
to be liked and accepted, a hypersensitivity to rejection and criticism
with restricted personal attachments, and a tendency to avoid cer-
tain activities by habitual exaggeration of the potential dangers or
risks in everyday situations. (Chapter V of Version 2007 published
on the Web)
In Version 2007 of the Web-based ICD-10, there are no
specific criteria lists for any of the personality disorders, hinting
at a shift to more prototypic representations or descriptions of
personality pathology. This is consistent with future plans for
the APD diagnosis in the next edition of the DSM.
ICD-11, DSM-5, and the Future of APD
Both the ICD-1 1 and the DSM-5 are under revision at the time
this article is going to press, and so there are many uncertain-
ties about the future diagnostic conceptualizations of APD.
Though still subject to further revisions, a general model of
personality disorders for the DSM-5 has been developed and
informs the likely diagnostic future of APD in this manual.
While the DSM-IV had conceptualized personality disorders in
terms of essential features, the committee developing DSM-5
is considering a prototypic model in which specific domains or
facets may be characteristics of multiple disorders.
The DSM-IV opens the discussion of APD in terms of its key
feature, namely, a "pervasive pattern of special inhibition,
feelings of inadequacy, and hypersensitivity to negative evalu-
ation." This idea of a disorder having one 'essential feature' is
different from the DSM-5, which opens with the phrase "Indi-
viduals who match this personality disorder type" and also later
uses the phrase "individuals resembling this type," both of
which illustrate prototypic description rather than one derived
from a polythetic collection of an essential number of criteria.
One proposal is to change the diagnostic system from a
categorical system to a hybrid prototype-dimensional system
based on a prototype rating and ratings on six-domain trait
dimensions. The first component is an assessment of where an
individual fits in the five severity levels of personality function-
ing using the Self and Interpersonal Functioning Continuum.
This assessment of level of personality functioning was not a
part of the DSM-IV, as historically the diagnosis of a functional
impairment was seen as sufficient evidence of a personality
disorder without a quantitative rating of disturbance. The sec-
ond component is the prototype rating system. Each personal-
ity disorder has a prototypic description that emphasizes
typical deficits and features of the disorder. Patients are
260 Avoidant Personality Disorder
compared to the prototype in terms of how well they match the
description, and a rating from 1 (no match) to 5 (very good
match) is made. The description of the APD prototype is
shown in Table 3.
The second component of the proposed personality disor-
der diagnosis for DSM-5 has roots in trait theories of personal-
ity, including the Five Factor Model developed by Thomas
Widiger, Paul Costa, and many others. The resulting DSM-5
model is a 'hybrid' system that combines specific personality
disorder 'types' along with ratings for traits and functioning.
Traits relevant to APD in the DSM-5 rating system include the
tendency to feel responsible for bad things that happen, anhe-
donia, having difficulty allowing oneself to acknowledge or
express wishes and emotions, and a desire to be recognized
as unique. As shown in Table 3, the traits are rated 0 (very
little or not at all descriptive) to 3 (extremely descriptive).
The American Psychiatric Association maintains a web site
illustrating the diagnoses and keeping the field abreast of
updates to the proposed revisions.
The future of APD, or anxious personality disorder, in the
ICD-1 1 is less certain at this point in time. Mental disorders in
general are among a handful of disorders that are a focal point
for substantial revision. In contrast to the DSM, which focuses
solely on mental disorders, the ICD covers all disease-related
classification and the 'Mental and Behavioral Disorders' is
but one component of the manual. Empirical studies on the
structure of mental disorders are being presented to the work
group for ICD Mental and Behavioral Disorders led by Dr.
Steven Flyman, and there exists the possibility for substantial
revision, including folding the symptoms of personality dis-
orders into clinical syndromes. These considerations stem
from questions raised by overlapping features of psychopa-
thology and personality disorders in the current form of
diagnoses, which call for new ways to examine and parse
clinical symptoms to better address the problem of diagnostic
comorbidity, a problem for many disorders including APD to
which we now turn.
Comorbidity
Like APD, fear or avoidance of social situations and fear of
possible scrutiny characterize social phobia, an Axis 1 disorder,
in the DSM-IV-TR. However, the social situations tend to be
more circumscribed for social phobia than for APD. Still, there
is considerable diagnostic overlap between APD and social
phobia, especially the generalized subtype of social phobia
where a number of social settings are associated with an anx-
ious state. In light of this overlap, and as the DSM moves more
toward a dimensional approach, a key question reemerges:
Does APD exist as part of a continuum, representing the
more chronic, wide-ranging, and severe point on a spectrum
of anxiety disturbance? In general, empirical results have sup-
ported the idea of a continuum model where APD is a more
severe and pervasive form of 'social phobia' that is so long-
standing that it has become integral to personality and inter-
personal functioning. Perhaps it is an anchor point toward the
more extreme end, with social phobia falling somewhere in
the middle, and anxious behavior colloquially described as
'shyness' falling in the more mild, clinically insignificant end
of the continuum. In considering the continuum model, it is
important to determine if it is merely quantitative differences
that distinguish the two disorders.
Are APD and Social Phobia Separate Disorders?
The question of whether or not social phobia and APD are
separate and distinct forms of psychopathology is a question
that has plagued the field since the diagnosis was introduced in
the 1980 DSM-III. Many studies have found overlap between
APD and social phobia; this overlap may reflect shared or
similar facets of the pathology, or it could be due to similarities
in the descriptive criteria for the diagnoses. In the revisions
from the DSM-III to DSM-IV-TR, there have been several
adjustments to the criteria to reduce the diagnostic overlap
and increase specificity. As previously described, the proposed
revisions for the DSM-5 include reformulating APD as avoi-
dant type; characterized by a 'negative sense of self, associated
with a profound sense of inadequacy, and inhibition in estab-
lishing intimate interpersonal relationships,' specifying that,
for APD, the social anxiety stems from the personal feeling of
inadequacy.
Despite modifications and changes in wording and empha-
sis of criteria in DSM-III through DSM-IV, a trend for the
cooccurrence of these two disorders has persisted. An apparent
lack of qualitative differences leads researchers to hypothesize
that the difference between the two is quantitative. Given
similarities in criteria sets, this finding is somewhat intuitive
and unsurprising as similar relations are also found between
other Axis I and personality disorders. For APD, various
research groups including the collaborative longitudinal per-
sonality study (CLPS) have reported significant associations
with Axis I disorders, including social phobia, that have
anxiety/inhibition as criteria. Several personality disorders
also show this same association with related Axis 1 disorders
with which they share criteria. A potential limitation in many
of these studies is that data is derived from clinical samples that
are comorbid for multiple Axis I and II disorders. However,
other researchers, such as Chambless and colleagues have
assessed impairment and distress among participants with the
generalized type of social phobia, with and without APD, using
self-report and observation and found no significant differ-
ences between the groups when they controlled for social
phobia, thus supporting the continuum hypothesis.
To argue against the continuum model, it has been noted
that higher levels of social avoidance, depressive symptoms,
neuroticism, introversion, and social and occupational im-
pairment were found for participants with generalized social
phobia in combination with APD than for those participants
with generalized social phobia alone. Thus it may be that APD
differs from social phobia because it manifests itself in more
severe cases of depression, introversion, and social and occu-
pational impairment. Rettew has argued for a qualitative dis-
tinction between APD and social anxiety disorders, suggesting
that widening the assessment scope to include nonsocial
domains (introversion, passivity, fear of novelty, for example)
would increase the likelihood of clarifying the distinction
between APD and social phobia. Regardless of the outcome
of this debate, two things seem clear. First, those with APD are
Avoidant Personality Disorder 261
Table 3 Prototype description for avoidant personality disorder DSM-5 presently under consideration by the DSM task force
Prototype description
Individuals who match this personality disorder type have a negative sense of self, associated with a profound sense of inadequacy, and inhibition in
establishing intimate interpersonal relationships. More specifically, they feel anxious, inadequate, inferior, socially inept, and personally unappealing;
are easily ashamed or embarrassed; and are self-critical, often setting unrealistically high standards for themselves. At the same time, they may have a
desire to be recognized by others as special and unique. Avoidant individuals are shy or reserved in social situations, avoid social and occupational
situations because of fear of embarrassment or humiliation, and seek out situations that do not include other people. They are preoccupied with and
very sensitive to being criticized or rejected by others and are reluctant to disclose personal information for fear of disapproval or rejection. They
appear to lack basic interpersonal skills, resulting in few close friendships. Intimate relationships are avoided because of a general fear of attachments
and intimacy, including sexual intimacy.
Individuals resembling this type tend to blame themselves or feel responsible for bad things that happen, and to find little or no pleasure, satisfaction, or
enjoyment in life’s activities. They also tend to be emotionally inhibited or constricted and have difficulty allowing themselves to acknowledge or
express their wishes, emotions - both positive and negative - and impulses. Despite high standards, affected individuals may be passive and
unassertive about pursuing personal goals or achieving successes, sometimes leading to aspirations or achievements below their potential. They are
often risk-averse in new situations.
Description
Having frequent, persistent, and intense feelings of nervousness/tenseness/ being on edge; worry and
nervousness about the negative effects of past unpleasant experiences and future negative possibilities; feeling
fearful and threatened by uncertainty
Having fears of rejection by, and/or separation from, significant others; feeling distress when significant others
are not present or readily available; active avoidance of separation from significant others, even at a cost to
other areas of life
Having a negative outlook on life; focusing on and accentuating the worst aspects of current and past
experiences or circumstances; expecting the worst outcome
Having a poor opinion of one’s self and abilities; believing that one is worthless or useless; disliking or being
dissatisfied with one’s self; believing that one cannot do things or do them well
Having frequent and persistent feelings of guilt/ shame/ blameworthiness, even over minor matters; believing
one deserves punishment for wrongdoing
Disinterest in and avoidance of close relationships, interpersonal attachments, and intimate sexual relationships
Preference for being alone to being with others; reticence in social situations; avoidance of social contacts and
activity; lack of initiation of social contact
Lack of emotional experience and display; emotional reactions, when evident, are shallow and transitory;
unemotional, even in normally emotionally arousing situations
Lack of enjoyment from, engagement in, or energy for life’s experiences; deficit in the capacity to feel pleasure or
take interest in things
Indifference to or disinterest in local and worldly affairs; disinterest in social contacts and activity; interpersonal
distance; having only impersonal relations and being taciturn with others (e.g., solely goal- or task-oriented
interactions)
Complete lack of risk-taking; unwillingness even to consider taking even minimal risks; avoidance of activities
that have even a small potential to cause injury or harm to oneself; strict adherence to behaviors to minimize
health and other risks
Traits
Domain
Facet
Negative
Anxiousness
emotionality
Separation
anxiety
Pessimism
Low self-
esteem
Guilt/Shame
Introversion
Compulsivity
Intimacy
avoidance
Social
withdrawal
Restricted
affectivity
Anhedonia
Social
detachment
Risk aversion
Source: Accessed from http://www.dsm5.org/ on 30 December 2010.
Ratings are made on the prototype (1 = no match to 5 = very good match) and on the traits (0 = very little or not at all descriptive to 3 = extremely descriptive).
more severely impaired; whether or not the distinction is qual-
itative or quantitative is secondary. Second, it may be helpful
to understand APD as a dysfunctional developmental adapta-
tion to chronic social anxiety that, along with low self-worth,
becomes folded into an individual's personhood.
In sum, the evolution of the criteria for APD from DSM-III
to DSM-TV-TR, and beyond to those under consideration for
DSM-5, places greater emphasis on the characteristics or traits
of fear of novelty, passivity, and introversion, along with a
poor sense of self and the expectation to draw the criticism
from others. The proposed changes for APD emphasize intro-
version and avoidance of activities, including social situations,
while criteria for social phobia emphasize the anxiety limited
to what is felt in the midst of social situations.
There are several aspects among the APD criteria that high-
light nonsocial factors, such as low self-esteem and guilt/
shame. There is also an emphasis on behavior in intimate
relationships in the criteria for APD that is not present in the
social phobia. Additionally, criteria for social phobia highlight
the reaction to social situations; fear and anxiety out of pro-
portion to the danger presented by the situation, which is not
discussed in APD definitions. These distinctions may eventu-
ally clarify qualitative differences, or may suggest personality
traits that make individuals more vulnerable to anxiety. It is
worth noting that differences in the criteria sets for APD and
social phobia are not wholly mutually exclusive, and the rates
of cooccurrence still merit investigation and raise questions
about the Axis I and Axis II division in the DSM. This suggests
262 Avoidant Personality Disorder
the possibility that there are core pathological elements that
may be the same if diagnosis were based on a different level
than the clinical reports and observation, for instance, neural,
cognitive, or more basic temperamental features.
APD and Other Personality Disorders
Prior to the introduction of APD in the 1980 DSM-III, 'avoi-
dant' personality was captured by the diagnosis of schizoid
personality and the psychoanalytic construct of 'phobic char-
acter.' In 1969, Millon first argued for a distinct diagnostic
construct for 'avoidant personality disorder' but it was not
until the DSM-III that APD was officially cleaved from the
old DSM I/ll diagnosis of schizoid personality. Thus, among
the personality disorders, APD and schizoid personality disor-
der share a close relationship based on their diagnostic lineage,
though there have been efforts to distinguish the two. Both
personality types tend to engage primarily in solitary as
opposed to social activities, and appear aloof or inhibited in
social situations. Although similar on this clinical symptom,
the distinction between avoidant and schizoid personality dis-
orders resides in motivation or desire for intimacy for APD,
something that is said to be lacking in schizoid. In addition,
there is a capacity, albeit an often unfulfilled capacity, for social
attachment in APD that does not seem to be present for schiz-
oid personality disorder. Unlike individuals with avoidant
personality disorder, who intensely desire relationships and
avoid them because of exaggerated fears of rejection, persons
with schizoid personality disorder have little or no apparent
interest in developing interpersonal relationships. In 1982,
Millon clarified this distinction in motivation for attachment,
and his distinction has been included in the criteria and the
text description of APD in the DSM ever since. For schizoid, the
motivation was passive, whereas for avoidant, it was active.
Millon's distinction between active and passive detachment
was central to the official introduction of avoidant personality
disorder in DSM-III.
Like schizoid, those with schizotypal or paranoid personal-
ity disorder may also react negatively to the idea of interper-
sonal relationships. However, in both of these cases, the
wariness in interpersonal relatedness is more apt to stem
from paranoia rather than the hypervigilance for criticism
common to APD or the indifference of schizoid. Schizoid,
schizotypal, and paranoid personality disorders all reside in
the DSM Cluster A 'odd-eccentric.' Though paranoid features
connote a certain sense of fear, that type of reaction is based
more on disordered thinking rather than anxious fear based
on the anticipation of criticism, as is the case with APD.
Another personality disorder that may share some similar-
ity on the clinical surface with APD is dependent personality
disorder. Like APD, dependent personality disorder rests in the
anxious or fearful 'Cluster C' of DSM Axis II personality dis-
orders. Characteristic of both disorders are low self-esteem,
rejection sensitivity, and an excessive need for reassurance.
Both may be defined in terms of attachment anxiety. However,
a key distinction of APD is that the anxiety is driven by a fear of
rejection, whereas in the case of dependent personality the
anxiety stems from fear of separation and/or rejection once
a relationship has been established. Thus, with dependent
personality disorder, distress is clear in the midst of a close
relationship or after a close relationship ends. Often, those
with dependent personality disorder scurry to find a new rela-
tionship after one has ended. People suffering from APD, on
the other hand, are slow to enter into new relationships.
Stability and Course
A primary distinction between Axis I and Axis II personality
disorders in the DSM system is the chronic and enduring
course of personality disorders relative to clinical syndromes
on Axis I. This long-held assumption is part of the definition of
personality disorders as chronic and enduring, which has been
challenged with recent empirical studies that have examined
the prospective course of personality disorders, such as the
CLPS and the McLean Study for Adult Development. The gen-
eral finding from both of these longitudinal, naturalistic stud-
ies is that personality disorders remit diagnostically much
more often than was originally assumed. This pattern has
been demonstrated specifically with APD in report from the
CLPS led by Grilo (at the symptomatic level) and by Sanislow
(at the constmct level). However, other CLPS reports led by
Skodol have shown that functional impairment, including
social and occupational deficits, persist for APD and other
personality disorders even though the symptom counts may
drop below diagnostic threshold (true for other personality
disorders, too). Thus, clinically significant impairments that
impede the social relationships and occupational potential of
those with APD remain evident over the long run.
DSM-IV-TR personality disorder criteria are a mixture of
traits, symptoms, and behaviors, and it is reasonable to query
if certain APD symptoms tend to persist more than others.
In another CLPS report led by McGlashan, it was evident that
certain APD criteria were more enduring than others. In assess-
ments conducted 2 years later by clinical interviewers blind
to the baseline diagnoses, the criteria 'feels inadequate' and
'socially inept' were the most persistent. 'Preoccupation with
rejection,' 'need to be liked first,' and 'void risks for fear of
embarrassment' continued to a moderate degree, while 'fears
ridicule and shame' and 'avoids jobs with interpersonal con-
tact' were the most likely to remit. Thus, the top two most
persistent criteria include both a deficit in self and an interper-
sonal deficit. At the other end, the least persistent are tied to
behavior in a specific situation (jobs with interpersonal con-
tact) and fear-based anxiety (fear ridicule and shame). The
more persistent are more global, representing general person-
ality constructs, whereas the least persistent are based on beha-
viors (interpersonal jobs) and symptom based (fear). Like
APD's Axis I counterpart, social phobia, these latter two criteria
may be easier to overcome than one's sense of self or more
global deficits in social skills.
Interpersonal Experiences, Development, and
Risk for APD
Like other personality disorders, interpersonal problems are a
critical feature of avoidant personality disorder. In fact, almost
all DSM criteria for APD describe aspects of the interpersonal
Avoidant Personality Disorder 263
problems generated and encountered by an individual with
APD. The most articulate description of the interpersonal
features of the APD symptoms are derived from Lorna Smith
Benjamin's structural analysis of social behavior (SASB). The SASB
codes patterns of interpersonal behavior of both the self and
the other, and also codes for the patterns of interpersonal
behavior that one has repeatedly experienced as they relate
to how a person perceives or thinks about him or herself.
Benjamin has applied this model to the DSM personality dis-
orders, including APD, and has linked characteristic interper-
sonal patterns to APD criteria. From there, it is possible to trace
back characteristic interpersonal experiences of the person
suffering APD and to infer how they 'talk to' or otherwise
treat themselves.
Interpersonal Origins of APD
According to Benjamin's interpersonal theory, the people with
APD live in a state expecting degrading, humiliating attack.
Their self-protective response to this possibility, which for
them looms with certainty, is social withdrawal. This promotes
a perception of others that borders on paranoid and a reliance
on safety at home that can sap the resources of the few indivi-
duals with whom the person with APD does have a relation-
ship. Benjamin describes these patterns in her book, The
Interpersonal Diagnosis of DSM-IV Personality Disorders. The ori-
gins of this pattern of behavior require that the person had
enough love and nurturance to form a good sense of self, and
the difficult interpersonal transactions came during later stages
of psychological development. For instance, "exhortations
combined with degrading mockery" (p. 292) put a premium
on the need to occlude any personal failings, thus leading to
intense self-scrutiny along with vulnerability to criticism. For
Benjamin, the sine qua non for APD is a ". . .defensive with-
drawal out of fear of humiliation, attack, and rejection, and
the wish for acceptance" (p. 298). Clearly, for professionals
treating the persons with APD, it is important provide uncritical
support while not pushing them too far, too fast.
Treatment
Individual therapy has historically been used as a treatment
for APD. This type of therapy can be difficult for an avoidant,
because the maladaptive behaviors and thought processes
characteristic of APD interpersonal relationships also extend
to the relationship with the therapist. Those with APD often
fear being rejected by the therapist, tend to doubt the authen-
ticity of the therapist's concern, and are likely to reject help.
Therapists should take extreme care to avoid behavior that may
be interpreted as judgmental, as the person with APD is gener-
ally sensitive to even mild criticism, and should be prepared for
the patient to perceive the therapist as critical even if there is no
real basis for this perception. Benjamin stresses that therapists
should aim to provide 'accurate empathy' and 'warm support'
in order to build trust. Once trust has been successfully estab-
lished, the therapeutic relationship may serve as a testing
ground for new coping skills. In the context of a 'safe' relation-
ship, dysfunctional beliefs about the therapy relationship can
be frankly examined, serving as a model for the patients to
evaluate other relationships in their life.
Cognitive behavioral therapy (CBT) techniques, which
encourage patients to identify the fears of rejection and criti-
cism that underlie APD and then modify these distorted
thought processes, have proven particularly helpful in the treat-
ment of APD. Cognitive restructuring, for example, involves
examining the patient's thoughts about the feared situations
and challenging them. This technique is predicated on the
theory that social anxiety arises from misconceptions about
dangers that social situations pose and other false and overly
negative cognitions. The process of cognitive restructuring
includes identifying these negative beliefs, evaluating their
accuracy, and recognizing possible alternatives to help address
the overwhelming anxiety characteristic of patients with APD.
Behavioral therapy strategies to treat APD generally fall under
two categories: graduated exposure and skills training. Exposure
therapy rests on the idea that graduated exposure to the fear-
provoking situation (in this case, social interaction) will help
the individual overcome their fear and bring about behavioral
change. For those suffering APD, exposure therapy typically
involves learning relaxation techniques and then applying them
to gradually higher-risk social situations both in and outside of
therapy. This treatment is most effective when the patient is fully
engaged in the threatening situation and is open to becoming
fully immersed in the experience despite the negative emotions
destined to arise. For the person with APD, this is difficult because
it runs counter to instinctive avoidance of engaging in frightening
social situations. The response by someone suffering from APD
is not always obvious avoidance, and a patient may surrepti-
tiously or unwittingly utilize maladaptive cognitive processing
strategies to reduce their anxiety such as not fully attending to
the stimuli meant to induce negative emotion. Thus, in the case
where social anxiety has become bound with self-definition and
esteem, exposure therapy may prove ineffective.
Social skills training is intended to directly address any inter-
personal deficits the avoidant may exhibit. This strategy is pre-
dicated on the assumption that people with APD exhibit poor
social skills, and that teaching them positive social skills will
increase positive reactions from others and thus reduce anxiety.
Much of this work has focused more specifically on shyness and
related social deficits, and clinical researchers who have targeted
the symptomatic aspects have demonstrated good results that are
applicable to treatments aimed at APD. For instance, basic skills
training in social behaviors such as assertiveness, eye contact, self-
disclosure, nonverbal social cues, and listening skills are likely to
invite positive, warm, and uncritical responses from others and
thus help the person suffering from APD build confidence.
Both social skills training and graduated exposure therapy
can be performed in a group or individualized setting, though
there are obvious advantages for those patients who can toler-
ate group treatment as the results may more effectively gen-
eralize to other areas of the patients' lives. As with any
therapeutic treatment, a personalized approach matched to
the patient's needs is desirable, though much more research
is needed on how to effectively identify needs and match those
needs with different interventions.
Treatment Outcome Studies
Alden and colleagues conducted a rare study that specifically
focused on group treatment for APD. They performed a
264 Avoidant Personality Disorder
randomized controlled clinical trial investigating the efficacy of
graduated exposure therapy alone versus graduated exposure
with interpersonal skills training, and found that the APD
patients who received both types of treatment had improved
in reports of social reticence, social anxiety, and a wide variety
of other measures, and that the gains had been maintained
3 months after treatment had ended. However, the treatment
groups did not significantly differ from one another in terms of
the amount of improvement, suggesting that the addition
of social skills training did not increase the benefits of grad-
uated exposure alone. Alden and colleagues later reexamined
the data from their study and concluded that variations in
the symptom patterns of APD may differentiate treatment
response. The reanalysis showed that patients who exhibited
more distrust and anger benefited most from graduated expo-
sure, while patients with problems asserting themselves were
shown to benefit from both treatments, especially an intimacy-
focused social skills training. Unfortunately, despite the gains
made by the APD patients during treatment, the patients still
functioned significantly below normal levels of social behavior
at the conclusion of the 1 0 week treatment. The authors suggest
that a longer period of treatment may be necessary to further
increase progress. Research is also needed to compare the
efficacy of group and individual therapy for APD, as Alden
studied all treatments in a group setting.
Insights from Treatment of Social Anxiety
Scientific research on the clinical outcomes for treatment of
the APD diagnosis per se is very limited. Other, related forms
of anxiety, such as social phobia, have been more frequently
studied and this more general work offers insights about tech-
niques that may be effective for APD. Like APD, treatment for
social anxiety often consists of CBT techniques such as cogni-
tive restructuring, graduated exposure, relaxation techniques,
and social skills training. Heimberg reviewed the efficacy of
treatment for social anxiety looking at the efficacy of types
of CBT as treatment for social anxiety. He reported that all
types of CBT produce better effect sizes than no treatment
control groups, and exposure alone and exposure plus cogni-
tive restructuring produce similar, significant effect sizes, while
social skills training, cognitive restructuring, and relaxation
training all produce slightly less change than exposure-based
interventions. No differences were found between individual
and group treatments, and these gains were maintained at
follow-up assessments. Another group of researchers, Alden
and colleagues, looked at exposure alone versus exposure
plus social skills training versus a waitlist control group for
adults who reported impairment in various areas of social
functioning, including experiencing extreme shyness. The
group given the combination of exposure plus social skills
training reported significantly improved functioning com-
pared to other groups.
Pharmacotherapy
A number of studies have looked at the effectiveness of pharma-
cotherapy for social anxiety disorders, although results
are inconclusive as to whether the benefits of medication
are clinically significant over CBT and other psychotherapy
approaches. Antidepressants, particularly the SSRI's are the
most studied and appear to be the most effective medication
treatment, although antianxiety medications may provide
short-term relief. Beta blockers are also often helpful. In compar-
ative studies, most work demonstrates that medication groups
and CBT groups improve about the same, and curiously, placebo
groups also show significant improvement, though usually to
a lesser degree. A combination of medication plus CBT is com-
mon in practice, but the efficacy of adding medication to CBT,
relative to CBT alone, is not well studied enough to draw firm
conclusions. Most important, studies examining longer-term
outcomes of medication versus psychotherapy are needed.
Limits to Understanding APD Treatment from Other Disorders
Treatment for social anxiety appears to address some, but not
all types of impairment in APD. It addresses the social im-
pairment, and can thus be somewhat helpful in treating APD.
In his review, Heimberg showed that one of the factors pre-
dicting outcomes of treatment is the severity of symptoms at
pretreatment. Patients with both social anxiety disorder and
APD tend to begin treatment more impaired than those with
social anxiety alone, and while they showed improvement,
they were still more impaired than patients with just social
anxiety at posttreatment. Patients with the generalized type of
social anxiety (the diagnosis of social anxiety most similar to
APD) also tended to be more impaired, and CBT was less
effective for this type than for those with the nongeneralized
types of social anxiety. This result is not too surprising as
treatment for social anxiety typically does not address nonso-
cial characteristics that typify APD, such as guilt/shame, avoid-
ance of (nonsocial) activities, fear of novelty, and passivity.
Conclusion
Most diagnosticians and taxonomists of psychopathology and
personality are well aware that these diagnoses are constructs
that help us to organize patterns in myriad traits, symptoms,
and behaviors and to recognize those features that coalesce
and thus suggest some coherence. As such, diagnostic con-
structs such as APD provide a useful explanatory tool to help
understand and communicate among professionals, and to
develop, implement, and guide treatments. The APD diagnosis
provides valuable clinical utility, but diagnostic systems are not
perfect, in part because the diagnostic constructs themselves
are not 'real' or natural kinds. Rather, diagnostic constructs are
approximations of a hypothetical construct; one that itself is
not directly observable or otherwise tangible. In the absence of
an etiological understanding and the identification of a specific
pathology (rather than a syndrome), problems such as diag-
nostic overlap are part of the course. Thus, it is important to
bear in mind that the comorbidity with APD and Axis I dis-
orders such as social phobia may be a matter of intensity, or
that criteria overlap and comorbidity with other Axis II disor-
ders such as schizoid personality disorder may reflect shared
pathological mechanisms or more simply may be an artifact of
our diagnostic architecture.
APD is a useful diagnostic construct that tells us about a
person chronically suffering extreme levels of anxiety that stems
Avoidant Personality Disorder 265
from a negative view of the self, as evident in social avoidance.
While APD has manifest interpersonal 'symptoms,' it is equally a
problem of one's own inner experience. Core to the disorder is
the fear of being perceived as a failure, and this fear is so intense
that it drives the person suffering from APD to isolation. The
relief experienced from this self-imposed isolation can further
reinforce the problem, and thus perpetuate a long-standing per-
sonality pattern from which it is difficult to break free.
The anticipated DSM-5 and ICD-11, along with a host of
issues that the field is presently tackling, present challenges
beyond human prescience when describing a diagnosis that
may change or disappear altogether as knowledge advances.
For instance, although the prototype description helps make
clear the definition of APD, a trait approach to diagnosis may
be favored in the DSM-5. Nonetheless, there are exciting pos-
sibilities of changes ahead, including whether dimensional
approaches offer advantages over categorical approaches, or if
an optimal number of factors or criteria better captures a form
of personality-related pathology than does the DSM nomen-
clature that has persisted for the last three decades. Last, ques-
tions about the boundary between Axis I and II disorders have
also been raised, for APD with social phobia. Future develop-
ments in personality pathology and psychiatric diagnosis are
sure to be exciting, and whatever viewpoint one holds of the
DSM system, it has in the last three decades yielded a wealth of
empirical information to scaffold for future diagnostic efforts.
See also: Anxiety Disorders; Self-Esteem; Separation Anxiety; Social
Anxiety Disorder.
Further Reading
Alden L (1989) Short-term structured treatment for avoidant personality disorder.
Journal of Consulting and Clinical Psychology 57: 756-764.
Alden LE and Capreol MH (1 993) Avoidant personality disorder: Interpersonal problems
as predictors of treatment response. Behavior Therapy 24: 357-376.
Alnaes R and Torgersen S (1988) The relationship between DSM-III symptom disorders
(Axis I) and personality disorders (Axis II) in an outpatient population. Acte
Psychiatrica Scandinavica 78: 485-492.
American Psychiatric Association (1980) Diagnostic and Statistical Manual tor Mental
Disorders , 3rd edn. Washington, DC: American Psychiatric Association.
American Psychiatric Association (2000) Diagnostic and Statistical Manual for Mental
Disorders, 4th edn., text revision. Washington, DC: American Psychiatric
Association.
American Psychiatric Association (2010) DSM-5 Development. (2010) DSM-5: The
Future of Psychiatric Diagnosis. Retrieved from http://www.dsm5.org/Pages/
Default.aspx.
Benjamin LS (1974) Structural analysis of social behavior. Psychological Review
81: 392-425.
Benjamin LS (1 996) Interpersonal Diagnosis and Treatment of Personality Disorders,
2nd edn. New York: Guilford.
Brooks RB, Baltazar PL, and Munjack DJ (1989) Co-occurence of personality disorders
with panic disorder, social phobia, and generalized anxiety disorder: A review of the
literature. Journal of Anxiety Disorders 3: 259-285.
Cappe RF and Alden LE (1986) A comparison of treatment strategies for clients
functionally impaired by extreme shyness and social avoidance. Journal of
Consulting and Clinical Psychology 54: 796-801 .
Chambless DL, Fydrich T, and Rodebaugh TL (2008) Generalized social phobia and
avoidant personality disorder: Meaningful distinction or useless duplication?
Depression and Anxiety 25: 8-19.
Costa PT and Widiger TA (1 994) Personality Disorders and the Five-Factor Model of
Personality. Washington, DC: American Psychological Association.
Cuthbert BN (2005) Dimensional models of psychopathology: Research agenda and
clinical utility. Journal of Abnormal Psychology 14: 565-569.
Grilo CM, Shea MT, Sanislow CA, et al. (2004) Two-year stability and change in
schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders.
Journal of Consulting and Clinical Psychology 72: 767-775.
Heimberg RG (2002) Cognitive-behavioral therapy for social anxiety disorder: Current
status and future directions. Society of Biological Psychiatry 51: 101-108.
Horney K (1945) Our inner conflicts. New York: Norton.
Lenzenweger MF, Lane MC, Loranger AW, and Kessler RC (2007) DSM-IV personality
disorders in the National Comorbidity Survey Replication. Biological Psychiatry
62: 553-564.
McGlashan TH, Grilo CM, Sanislow CA, et al. (2005) Two-year prevalence and
stability of individual criteria for schizotypal, borderline, avoidant, and
obsessive-compulsive personality disorders. The American Journal of Psychiatry
162: 883-889.
Millon T (1969) Modern Psychopathology: A Biosocial Approach to Maladaptive
Learning and Functioning. Philadelphia: W.B. Saunders.
Millon T (1982) Millon Clinical Multiaxial Inventory Manual, 2nd ed. Minneapolis:
National Computer Systems.
Otto M, Pollack M, Gould R, Worthington J, McArdle E, and Rosenbaum JE (2000) A
comparison of the efficacy of clonazepam and cognitive behavioral group therapy for
the treatment of social phobia. Journal of Anxiety Disorders 14: 345-358.
Ralevski E, Sanislow CA, Grilo CM, et al. (2005) Avoidant personality disorder and
social phobia: Distinct enough to be separate disorders? Acta Psychiatrica
Scandinavica 112: 208-214.
Rettew DC (2000) Avoidant personality disorder, generalized social phobia, and
shyness: Putting the personality back into personality disorders. Harvard Review of
Psychiatry 8: 283-297.
Sanderson WC, Wetzler S, Beck AT, and Betz F (1994) Prevalence of personality
disorders among patients with anxiety disorders. Psychiatry Research 51:
167-174.
Sanislow CA, Little TD, Ansell EB, et al. (2009) Ten-year stability and latent structure of
the DSM-IV schizotypal, borderline, avoidant, and obsessive-compulsive
personality disorders. Journal of Abnormal Psychology 118: 507-519.
Sanislow CA and McGlashan TH (1998) Treatment outcome of personality disorders.
Canadian Journal of Psychiatry 43: 237-250.
Sanislow CA, Morey LC, Grilo CM, etal. (2002) Confirmatory factor analysis of DSM-IV
schizotypal, borderline, avoidant, and obsessive-compulsive personality disorders:
Findings from the Collaborative Longitudinal Study of Personality Disorders. Acta
Psychiatrica Scandinavica 105: 28-36.
Shea MT, Stout RL, Yen S, et al. (2004) Associations in the course of personality
disorders and axis I disorders over time. Journal of Abnormal Psychology
113: 499-508.
Skodol AE, Gunderson JG, Shea MT, et al. (2005) The collaborative longitudinal
personality disorders study (CLPS): Overview and implications. Journal of Personal
Disorders 19: 487-504.
Skodol AE, Oldham JM, Bender DS, et al. (2005) Dimensional representations
of DSM-IV personality disorders: Relationships to functional impairment.
The American Journal of Psychiatry 1 62: 1 91 9-1 925.
Skodol AE, Pagano ME, Bender DS, et al. (2005) Stability of functional impairment
in patients with schizotypal, borderline, avoidant, or obsessive-compulsive
personality disorder over two years. Psychological Medicine 35: 443-451.
Stein MB, Liebowitz MD, Lydiard RL, Pitts CD, Bushnell MS, and Gergel I (1998)
Paroxetine treatment of generalized social phobia (social anxiety disorder).
Journal of the American Medical Association 280: 708-713.
Trull TJ, Widiger TA, and Frances A (1987) Covariation of criteria sets for avoidant,
schizoid, and dependent personality disorders. The American Journal of Psychiatry
144: 767-771.
Turner S, Beidel D, and Jacob R (1994) Social phobia: A comparison of behavior
therapy and atenolol. Journal of Consulting and Clinical Psychology 62: 350-358.
van Velzen CJM, Emmelkamp PMG, and Scholing A (2000) Generalized social phobia
versus avoidant personality disorder: Differences in psychopathology, personality
traits, and social and occupational functioning. Journal of Anxiety Disorders
14: 395-411.
Widiger TA (1992) Generalized social phobia versus avoidant personality disorder:
A commentary on three studies. Journal of Abnormal Psychology 1 01 : 340-343.
Widiger TA, Trull TJ, Clarkin JF, Sanderson C, and Costa PT (1994) A description of the
DSM-lll-R and DSM-IV personality disorders with the five-factor model of
personality. In: Costa PT and Widiger TA (eds.) Personality disorders and the
five-factor model of personality, pp. 41-56. Washington, DC: American
Psychological Association.
World Health Organization (1992) The International Statistical Classification of Disease
(ICD), 10th edn. Geneva: World Health Organization.
266 Avoidant Personality Disorder
World Health Organization (2007) The International Statistical Classification of Disease
(iCD), 10th edn. Retrieved from http://apps.who.lnt/classificatlons/apps/icd/
icdlOonline/.
Zanarini MC, Frankenburg FR, Hennen J, Reich DB, and Silk KR (2006) Prediction
of the 10-year course of borderline personality disorder. The American
Journal of Psychiatry 163: 827-832.
Zimmerman M, Rothschild L, and Chelminski I (2005) The prevalence of DSM-IV
personality disorders in psychiatric outpatients. The American Journal of
Psychiatry 162: 1911-1918.
Relevant Websites
http://www.dsm5.org/Pages/Default.aspx - DSM-5 Development Website.
http://apps.who.int/classifications/apps/icd/icd10online/ - ICD-10/11 Website.
http://www.mentalhelp.net/poc/view_doc.php?type=doc&id=29133&cn=8 -
Interview with Professor Lorna Benjamin on SASB and the Structure and Treatment
of Personality Disorders.
Behavior Analysis
H S Roane, SUNY Upstate Medical University, Syracuse, NY, USA
A M Betz, Florida Institute of Technology, Melbourne, FL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Applied behavior analysis The application of principles of
operant conditioning to the behavior of individuals in
socially significant contexts.
Extinction Withholding reinforcement of a previously
reinforced response, which results in a decrease of that
response.
Operant behavior Behavior that is influenced by stimulus
changes that follow its occurrence.
Punishment A process of behavior change by which a
change in a stimulus event or condition that immediately
follows a response decreases the future frequency of that
response.
Reinforcement A process of behavior change by which a
change in a stimulus event or condition that immediately
follows a response increases the future frequency of that
response.
Selection by consequences Behaviors that produce
favorable outcomes are more likely to be repeated and those
that produce aversive outcomes are less likely to be repeated.
Three-term contingency The relation between an
antecedent event, a target behavior, and a consequent event.
Introduction
Behavior analysis is a discipline with three primary branches:
(a) behaviorism, which focuses on the world view or philoso-
phy of behavior analysis; (b) the experimental analysis of
behavior (EAB), which focuses on identifying and analyzing
the basic principles and processes that explain behavior; and
(c) applied behavior analysis (ABA), which focuses on solving
problems of social importance using the principles and proce-
dures of behavior analysis. Behavior analysis can be better
understood by examining the history and philosophy of how
behaviorism came to be.
Behavior analysis began as a school or subfield within the
discipline of psychology. While the majority of psychologists
in the early 1900s studied mental processes such as conscious-
ness, John B. Watson argued that the proper subject matter was
not states of mind or mental processes, but rather observable
behavior. Thus, Watson created a new direction in the field
of psychology by objectively studying behavior as a natural
science consisting of direct observation of the relationship
between environmental stimuli (S) and responses (R) which
came to be known as S-R psychology.
One component of S-R psychology is respondent behavior,
which is defined as behavior that is elicited by antecedent
stimuli. In 1927, Ivan Pavlov, published a collection of experi-
ments demonstrating that new stimuli can acquire the ability
to elicit respondents. At the time of this experimentation,
Pavlov was studying digestion using dogs as experimental sub-
jects. In a well-documented paradigm, a tone signaled the
availability of food. Although unintentional, Pavlov found
that the dogs began to salivate when the tone occurred (i.e.,
before the food was presented) presumably because the dog
had learned to 'expect' the forthcoming food. This line of
research led to the development of a type of learning typi-
cally referred to as respondent (Pavlovian) conditioning.
In Pavlov's basic respondent conditioning arrangement, the
presence of food was termed an unconditioned stimulus
(US), which elicited a reflexive response (in this case,
salivation) which was termed the unconditioned response
(UR). Initially, the tone exerted no differential effect on the
dog's behavior and was therefore referred to as a neutral stim-
ulus (NS). However, after several trials in which the tone was
presented just before the presentation of the food, the dogs
began to salivate at the presence of the tone. Thus, through this
pairing, the previous NS acquired characteristics of the US;
thereby becoming a conditioned stimulus (CS). The salivation
which followed the CS was referred to as the conditioned
response (CR) to differentiate that this response (although
topographically similar) was under the influence of a different
stimulus (a tone as opposed to the presentation of the food).
The work of Pavlov was influential to many scholars.
Among them was a young Harvard graduate student named
B. F. Skinner. Skinner was interested in replicating much of
Pavlov's work and began his doctoral work in the joint areas of
psychology and physiology. After a great deal of work based
upon Pavlov's findings, Skinner came to the conclusion that
although respondent conditioning could explain why several
behaviors occurred, it could not explain the majority of beha-
viors; especially those with no apparent antecedent causes
in the environment. This conclusion drove Skinner to study
other types of behavior (such as a rat's movement down a
runway) and a near-continuous modification of his experi-
mental methods (e.g., modifying devices so they recorded
the rats' behavior). These modifications, in turn, led to the
discovery of patterns of behavior among the rats such as how
quickly a rat ate food after running or what occurred if the
device failed to deliver food to the rat following running.
Skinner observed that these patterns were predictable across
organisms. Moreover, he observed that the organisms' behavior
was not influenced by the prior stimuli that elicited a reflexive
response; instead behavior seemed to be influenced by conse-
quent stimuli (changes in the environment that occurred
following behavior).
In 1938, Skinner published The Behavior of Organisms which
summarized his laboratory research attempting to account for
behavior that did not appear to be respondent. Skinner was
267
268 Behavior Analysis
interested in the scientific account of all behavior, including
those behaviors that appear to follow antecedent stimuli,
not just those that appeared to be involuntary or reflexive.
They are influenced by stimulus changes that have occurred
after a behavior (i.e., a consequent stimulus) and are called
operant behaviors. In other words, Skinner identified behavior
that operated on the organism's environment to produce
differential effects. He surmised that an organism's behavior
was governed by a process of selection by consequences in
which behaviors that produce desirable outcomes are repeated
(or strengthened), whereas those that produce aversive out-
comes are decreased (or weakened). In examining the effects
of consequences on behavior, Skinner interpreted responding
within the context of the operant three-term contingency
(antecedent-behavior-consequence).
Skinner's laboratory findings led to the development of a
new field of study which was referred to as the experimental
analysis of behavior. This field produced its own set of experi-
mental methods and (eventually) therapeutic practice as well
as a fundamentally distinct conceptualization of behavior.
Nevertheless, behavior analysis is often considered a subfield
of psychology. Many would argue, however, that the field of
behavior analysis differs considerably from the broad field
of psychology. To illustrate, the basic tenets that distinguish
behavior analysis from other areas of psychology include its
emphasis on (a) behavior as the basic datum for the field rather
than the psyche, the self, or other internal mental or metaphys-
ical structures or phenomena; (b) continuity between publicly
observable behavior and private events (e.g., thinking, feeling);
(c) prediction and modification of the behavior of indivi-
duals (rather than groups); (d) environmental explanations
of behavior; and (e) the study of behavior as a natural science.
These basic principles of behavior analysis continue to define
the work of behavior analysts and the separate domains of the
EAB and ABA. Before considering these respective domains, a
brief discussion of the underlying principles and concepts of
behavior analysis will be reviewed.
Basic Principles of Behavior
Reinforcement may be considered the most important princi-
ple of behavior and is the key to behavior change. Reinforce-
ment occurs when there is a change in a stimulus event or
condition that immediately follows a response, which
increases the future frequency of that response under similar
conditions. The changes in stimuli that function as reinforcers
can be described as either presenting a new stimulus into the
environment or removing an already present stimulus from
the environment.
Positive reinforcement occurs when a behavior is immedi-
ately followed by the presentation of a stimulus which, as a
result, increases the future frequency of that behavior. For
example, a rat pushes a lever in an operant chamber and
receives a pellet of food. If the frequency of lever pressing
increases, one can conclude that the presentation of the food
pellet positively reinforced the lever-pressing behavior. A simi-
lar effect might be observed for the child who throws a tantrum
in the department store. If the parent 'gives in' to the child's
behavior by purchasing a toy, the future occurrence of
throwing a tantrum is strengthened such that the child may
be more likely to throw a tantrum in the future to produce a
desired outcome. (It should be noted that positive reinforce-
ment is a technical term used to describe the effect of a contin-
gency on the occurrence of a response. This is in contrast to the
term 'reward' which is used to describe an item or activity
presented to someone in an attempt to change behavior. That
is, reinforcement is an effect demonstrated by an increase in
behavior, whereas a reward is simply something presented to a
person that may or may not affect their behavior).
When the frequency of a behavior increases because a
stimulus in the environment has been removed, the process
is referred to as negative reinforcement. Negative reinforce-
ment is characterized by escape or avoidance contingencies in
which the organism emits a response that either removes or
avoids the presentation of an aversive stimulus. The concept of
negative reinforcement can be demonstrated by revisiting the
rat and operant chamber. For example, a low dose of electricity
is sent through the floor of the operant chamber which gives
a mild shock to the rat and when the rat pushes the lever
the shock is terminated. If the frequency of lever presses
in the presence of the shock increases, the lever press response
has been negatively reinforced. In a more applied example,
loosening one's belt following a large meal often results in
the temporary attenuation of discomfort associated with eating
too much. In this example, the response of loosening the belt is
reinforced by the cessation of aversive stimulation (physical
discomfort) which increases the future likelihood of this
response in similar situations.
It is important to note that once a behavior has been
established and strengthened with reinforcement, it is not
necessary to reinforce that behavior on each occurrence.
Many behaviors are maintained by intermittently reinforcing
its occurrence. An intermittent schedule of reinforcement is a
contingency of reinforcement in which some, but not all,
occurrences of the behavior produce reinforcement. However,
if reinforcement for a behavior that was previously reinforced
is withheld for all occurrences, the frequency of that behavior
will decrease to similar levels demonstrated prior to reinforce-
ment or will cease altogether. This behavioral procedure is
referred to as extinction.
Punishment is another principle of behavior that is defined
based on its function. Punishment occurs when a change in a
stimulus, event, or condition immediately follows a behavior
and decreases the future frequency of that behavior. Like the
process of reinforcement, punishment affects behavior through
either the presentation or the removal of a stimulus. If an
aversive stimulus is presented contingent on a particular
behavior which results in a decrease in that behavior, positive
punishment has occurred. For example, when a child runs
into the street unsupervised, his mother reprimands him.
Therefore, the frequency of child running into the street
decreases. In contrast, negative punishment occurs when a
stimulus (or access to forms of stimulation) is removed from
the environment contingent upon a response, and decreases
the future frequency of that response. An example of a negative
punishment contingency can be demonstrated by a situation
in which a child throws a toy. The mother then takes the toy
away so he cannot play with it. In this example a preferred
stimulus, the toy, is removed following inappropriate
Behavior Analysis 269
behavior. Timeout is another example of a negative punish-
ment contingency in that, while in timeout the individual
cannot access sources of reinforcement. If a punishment con-
tingency is removed, the behavior will ultimately reverse
(increase) to levels near those seen prior to punishment. This
process is called recovery from punishment.
Although punishment has been shown to be an effective
procedure to decrease behavior, it can be argued that there are
potential side effects and problems that occur when imple-
menting punishment procedures. First, punishment can pro-
duce emotional and aggressive reactions. This is especially seen
when positive punishment procedures are used when an aver-
sive stimulus is presented as a consequence for a response.
Second, inappropriate escape and avoidance behaviors may
arise when a behavior is being punished. For example, a child
may begin to lie or hide behaviors to avoid contacting the
punishment contingency. Third, punishment may involve
undesirable modeling of the punishing behavior. Finally, the
decrease in the undesirable behavior of the person being pun-
ished may negatively reinforce the behavior of the punisher. In
other words, the person implementing the punishment con-
tingency may continue to implement punishment procedures
more frequently in the future.
Critical variables for the use of both punishment and rein-
forcement are consistency and contiguity. A consequence
should be implemented consistently to obtain the desired
effect on behavior. If reinforcement is applied sporadically,
the response will be more resistant to strengthening and less
resistant to extinction. Likewise, if punishment is applied on a
variable basis, the behavior will not decrease as quickly. Conti-
guity also plays a role in the efficacy of reinforcement and
punishment procedures. Generally speaking, delayed conse-
quences are less effective at changing behavior than are more
immediate consequences.
Some behavior analysts argue that, from a functional per-
spective, reinforcement and punishment are the only principles
needed to explain the basic effects of behavioral consequences.
However, a number of factors may influence a response within
behavioral contingencies. One such factor is the antecedent
stimulus that signals the availability of reinforcement to
occur. Such a stimulus is called a discriminative stimulus (S D ).
For example, a friend asking 'How are you' acts as a S D for you to
say, 'I'm fine,' which is then reinforced by social approval. One
would not say, 'I'm fine,' in the absence of another person
asking how you are feeling. Likewise, stopping at a traffic signal
is an example of behavior under the influence of a S D . There is
not physiological response governing this behavior; rather, our
learning history dictates the relation between stopping or
moving in the presence of red, yellow, or green lights. When a
response comes under the control of some stimuli and not
others, it is said to be under stimulus control (as in the case of
stopping instead of driving in the presence of a red light).
Another factor that may potentially influence a response
within a behavioral contingency is the reinforcing value of
the consequence. A motivating operation (MO) refers to an
environmental variable that alters the reinforcing effective-
ness of a stimulus, event, or condition and alters the current
frequency of all behavior that has been reinforced by that
stimulus event or condition. A common example of an MO
is food deprivation. Food deprivation acts as an MO that
increases the reinforcing effectiveness and value of food.
Therefore, food deprivation evokes behavior that, in the past,
has been reinforced with food. For example, the response of
food ingestion (e.g., eating a sandwich) is more likely to occur
when one has not eaten for several hours. It is unlikely that the
sandwich will be highly reinforcing immediately after eating a
large lunch.
Throughout the past 70 years, literally thousands of studies
have supported the fundamental principles of behavior analy-
sis through extensive empirical research carried out in both
basic laboratory and applied settings. In the following sections,
the focus of research and practice will be outlined for both EAB
and ABA.
The Experimental Analysis of Behavior
As mentioned previously, the EAB began in 1938 with Skinner's,
The Behavior of Organisms. The book summarized Skinner's lab-
oratory research conducted from 1930 to 1937. In the early
studies, Skinner primarily used animals (rats and pigeons) as
subjects to demonstrate reliable functional relations between
behavior and various environmental stimuli. He did this by
systematically manipulating the arrangement and schedules
of stimuli that precede and follow behaviors while recording
the rate at which a single subject emits a given behavior in a
controlled setting.
Because behavior analysts working in EAB are focused on
basic laboratory research, studies are typically conducted in a
well-controlled setting in which the researcher can control the
conditions and environment under which the research occurs.
When working with animals, operant chambers are used that
typically consist of at least one operandum for the subject to
manipulate (typically a lever or response key), and a device for
reinforcement delivery. During a typical session in an operant
chamber, the subject emits a specified response (e.g., lever press)
which results in the delivery of a reinforcer (e.g., a food pellet).
Skinner developed the operant chamber during the early years of
his experimentation to measure the rate of responding and to
study various schedules of reinforcement. Since Skinner began
his work with operant chambers, thousands of studies examin-
ing the effects of changes in environmental stimuli prior to and
preceding behavior have been conducted by behavior analysts.
Such studies have examined differing schedules of reinforce-
ment, novel or variable responding, allocation of responding
or choice (commonly known as the matching law), the effects
of different drugs on responding, and stimulus and discrimina-
tive control of responding, along with several other areas.
Since the development of the operant chamber, some
behavior analysts have moved to conducting basic laboratory
research with human participants as well. In recent years,
research has focused on studying complex human behavior
such as stimulus equivalence, choice, self-control, and the
effects of drugs on the behavior of humans.
Applied Behavior Analysis
The general principles under which ABA was founded were
developed (and continue to be refined) from the results of
laboratory experiments conducted in EAB. Thus, many of the
270 Behavior Analysis
principles that are applied to human behavior are the result
of translational research in which laboratory findings have
informed the development of novel applications. This transla-
tional link serves as the bridge between EAB and ABA.
ABA differs from EAB in that it is generally interpreted as a
clinical discipline in which the general principles of learning
and behavior are used to solve or improve problems of social
relevance. Behavior analysts who work in ABA conduct
research that assist in developing and evaluating practices
directed toward the remedy of problems associated with
socially significant behavior. Applied behavior analysts then
use the results of the applied research to create and implement
effective evidence-based procedures in more natural settings
such as schools, home, and the community. Such work often
focuses on behavioral problems that occur in particular set-
tings and are associated with particular populations (e.g.,
autism or other developmental disabilities) as well as those
that are within a large social context such as organizational
behavior management. ABA also has been used to treat clinical
behavioral problems such as anxiety, phobias, and depression.
Dimensions of ABA
Early in its development, applied behavior analysts worked
primarily in the field of psychology and education. As
described by Baer, Wolf, and Risley, there are seven dimensions
of ABA that must be implemented in order to assure that
effective interventions are developed and implemented. The
seven dimensions are (1) applied, (2) behavioral, (3) analytic,
(4) technological, (5) conceptually systematic, (6) effective,
and (7) generalizable.
Applied behavior analysts select behaviors that are applied,
meaning that they are socially meaningful and are currently of
importance to the individual whose behavior is being modi-
fied. For example, if a child with a diagnosis of autism does
not have verbal language in his repertoire, it would be more
socially acceptable and appropriate to teach the child to request
items (e.g., snacks, drinks, and toys) as compared to teaching
the child to ride a bicycle. At any given time, a behavior analyst
could modify numerous behaviors an individual emits. It is
important to prioritize which behaviors are most important
to modify. In addition to identifying target behaviors for indi-
viduals, behavior analysts must modify behaviors which will
affect how behavior analysis is communicated to the media,
the general public, and legislatures who fund services.
As described previously as one of the tenets of behaviorism,
applied behavior analysts directly observe and manipulate
behavior. They typically do not use indirect measures of behavior
like self-report, interviews, or checklists. In addition, they do not
attribute behavior as characteristics of inner qualities such as
personality traits. Instead, they attempt to identify a function of
the behavior by manipulating environmental variables and
observing interactions between behavior and those variables.
Even when working in an applied setting in which the goal
is to develop an intervention that will effectively increase or
decrease a behavior, it is important to demonstrate experimen-
tal control via a systematic analysis of behavior. Experimental
control is achieved when an experimenter demonstrates a
functional (i.e., causal) relationship between environmental
variables and a behavior. In ABA, experimental control is
demonstrated using various single-case experimental designs
including the reversal, the multielement, and the multiple
baseline design. These basic designs share three commonal-
ities: (a) prediction - the anticipated outcome of future mea-
surement, (b) verification - demonstration that the previous
expectation of behavior would remain unchanged in the
absence of an intervention, and (c) replication - repeating
the previous changes in behavior via the introduction and
removal of the intervention.
In addition to the analysis of an intervention through the use
of a controlled research design, behavior analysts often track one
or more dimensions of behavior to determine if an intervention
is producing significant behavior changes. Measures such as the
frequency (the number of times a behavior occurs), duration
(the amount of time in which a behavior occurs), and latency
(the measure of time from the onset of a stimulus to the start of
a response) are used to track the behavioral progress of an
individual. By repeatedly measuring and evaluating different
dimensions of behavior, the behavior analyst can determine if
changes are needed to the specified intervention.
An example of the analysis component of ABA can be demon-
strated by the single-subject treatment analysis conducted within
a reversal (ABAB) design that is depicted in Figure 1. Each data
point represents a 5-min session and sessions are presented
along the x-axis in the order in which they were conducted. The
y-axis depicts the percentage of each 5-min session in which
inappropriate behavior was observed. An initial baseline was
conducted to measure the occurrence of inappropriate behavior
in the absence of an intervention. High levels of inappropriate
behavior were observed (M = 65.9%). On the basis of these data,
one would predict that similar levels of responding would con-
tinue in the absence of an intervention (which was verified by the
collection of each subsequent baseline data point). Conversely,
one would predict that the introduction of an intervention
would result in a change in behavior, which it did ( M = 6.9%).
To ensure that the observed change in behavior was due to the
intervention presented (as opposed to an extraneous influence),
the treatment was withdrawn and the initial baseline pattern of
responding was replicated (M = 88.0% during the second set of
baseline data). Finally, the treatment effect was replicated when
the intervention was reintroduced (M— 14.6% during the sec-
ond set of treatment data).
In addition to selecting an appropriate design for evaluating
a functional relation, it is important that behavior analysts are
technological, meaning that they thoroughly and accurately
describe the procedures they use when conducting experiments
and implementing behavioral interventions. This information,
which includes written procedures, operational definitions of
target behaviors, and procedural integrity data, must be docu-
mented in a way that allows another individual to replicate the
study after reading these documents.
The assessments and interventions that behavior analysts
implement are applied in nature. However, these interventions,
and the approaches used to develop the interventions, must be
conceptually systematic. This means that they are based on the
basic behavior principles that have been empirically validated
over many years by scientists who conduct basic research on the
behavioral theories of EAB. Examples of conceptually systematic
intervention components are reinforcement, extinction, and
schedules of reinforcement (described above).
Behavior Analysis 271
Baseline Treatment Baseline Treatment
Figure 1 Percentage of session during which inappropriate behavior occurred during the baseline and treatment conditions for Gail. Adapted with
permission from Roane HS and Kelley ME (2008) Decreasing problem behavior associated with a walking program for an individual with developmental
and physical disabilities. Journal of Applied Behavior Analysis 41: 423-428.
Many experiments that utilize group designs use statistics to
determine if there are statistically significant differences between
groups. Applied behavior analysts rarely use statistics to deter-
mine if behavior change is significant (although there are several
statistical procedures for use with single-case research designs).
Instead, behavior analysts determine the effectiveness of their
procedures by evaluating their data, often through visual inspec-
tion, that is, whether the individual whose behavior was changed
and the family, caregivers, and friends of the individual find the
behavior change significant. lust because a behavior change is
statistically significant does not mean that the change is socially
significant. For example, a reduction of head banging from a rate
of 12 per minute to 6 per minute may be statistically significant.
However, the individual is still hitting her head ~360 times an
hour. Consequendy, this is not a socially acceptable level of
reduction of head banging. A more significant reduction needs
to occur in order to classify the intervention as effective. Typi-
cally, behavior analysts examine data visually to assess changes in
the stability (relative degree of variability in the data), trend
(direction of the behavior; either no trend, a therapeutic trend,
or a countertherapeutic trend), and level of behavior across
different experimental manipulations or interventions.
The last principle of ABA is that the findings must be
genemlizable to other settings, caregivers, or behaviors. If a
child is treated at a clinic and the child's aggressive and disrup-
tive behaviors are decreased to near-zero levels at the clinic, but
at school and at home the child is still engaging in the problem
behavior, then the behavior reduction has not generalized.
Generalization is important because it is not beneficial to
decrease a child's behavior in the clinic if he only spends a
few hours of his week in that setting. The behavioral interven-
tion is only beneficial if the intervention decreases the child's
behavior across different settings when different caregivers
implement it. The most effective way of ensuring that generali-
zation occurs is to program it into the intervention.
Researchers and clinicians have been implementing the
seven dimensions of ABA across several settings for years. Most
notably, ABA has been implemented in educational and clinical
settings with individuals with disabilities. However, interven-
tions based on the principles of ABA have been implemented
successfully in educational, clinical, sports, and business settings
to address a wide range of behavioral issues. The following
sections briefly describe the application of ABA in these areas.
Assessment and Treatment of Destructive Behavior
Destructive behavior (e.g., physical aggression, property
destruction, self-injurious behavior, pica) is fairly common
among individuals with developmental disabilities. Although
the principles of ABA have been applied therapeutically to
many types of behavior, one of the most notable utilizations
of these methods has been within the context of the assessment
and treatment of destructive behavior displayed by individuals
with developmental disabilities. The following text describes
the general course of assessment and treatment of these beha-
viors using principles of ABA.
Conducting a functional behavior assessment (FBA) is one
of the most crucial components of behavioral interventions in
educational and clinical settings. It allows a hypothesis to be
made about the relationship between the environmental events
and behavior. More specifically, it obtains information about
the functions a behavior serves (i.e., the causal effects these
behaviors have on altering the individual's environment).
FBAs allow the behavior analyst to hypothesize what conse-
quences an individual is obtaining for engaging in the problem
behavior. There are three common FBA methods: (1) experi-
mental functional analysis, (2) descriptive assessments, and
(3) indirect assessments.
An experimental functional analysis is the most structured
form of FBA. During a functional analysis, the antecedents
272 Behavior Analysis
and consequences similar to those in the person's natural
environment are experimentally arranged so that the effects
on destructive behavior can be separately observed and mea-
sured. An experimental functional analysis is often referred to
as an 'analog assessment' because, although the antecedents
and consequences are similar to those in the natural environ-
ment, the assessment itself is not conducted in the natural
environment. This allows the behavior analyst to control the
environmental variables to better assess the conditions under
which the destructive behavior occurs.
Typically, an experimental functional analysis comprises
four separate conditions: contingent attention, contingent
escape from demand, an alone condition, and a control condi-
tion. During the control condition, it is expected that destruc-
tive behavior will be low because reinforcement is freely
available and no demands are placed on the individual. The
four conditions are presented one at a time in a random
sequence to determine which conditions result in highest levels
of destructive behavior.
Although functional analyses result in clear demonstrations
of variables that influence the occurrence of destructive behav-
ior, there are some limitations as well. First, there is a risk of
temporarily strengthening or increasing the destmctive behav-
ior because the individual is contacting reinforcing stimuli
contingent on that behavior. It is also possible that the indivi-
dual's destructive behavior may acquire new functions during
the assessment. Second, it may not be optimal to assess some
behaviors using a functional analysis (e.g., those behaviors
that, although very serious, occur infrequently or those beha-
viors that are too dangerous to briefly reinforce). Third,
although variables are contrived to simulate the natural setting
during a functional analysis, the contrived setting may not
detect the variables in the natural setting that are maintaining
the destructive behavior.
If conducting an experimental functional analysis is either
not feasible or optimal, a behavior analyst may choose to
conduct either a descriptive FBA or indirect FBA. Similar
to an experimental functional analysis, a descriptive FBA
requires direct observation and measurement of the problem
behavior; however, during a descriptive assessment, the obser-
vation of the behavior is in the natural environment. Flence,
descriptive assessments involve observation of destructive
behavior under naturally occurring variables that are not sys-
tematically arranged. If direct observation is not possible, an
indirect FBA can be conducted. An indirect FBA encompasses
methods that result in information about the behavior without
directly observing that behavior. Such assessments include
behavioral interviews with caregivers or teachers, behavioral
checklists, rating scales, or questionnaires. Although indirect
FBAs are helpful when the behavior cannot be observed, the
information gathered must be taken with caution. Often, those
being interviewed may have a personal relationship with the
individual exhibiting the problem behavior. This may make
it difficult to obtain an unbiased and accurate description of
the problem behavior.
It is often important to develop an intervention that incor-
porates an individual's preferences for different activities,
foods, etc. For many individuals, identifying preferred items
and activities which can be incorporated into behavioral inter-
ventions as reinforcers is as easy as asking the individual or
significant others (i.e., caregivers, teachers, etc.) what kinds
of things they like. However, identifying effective reinforcers
for those individuals with severe disabilities may be more
challenging. Fortunately, there are a variety of procedures sup-
ported by research that can be used to determine the stimuli
an individual prefers as well as the conditions under which
reinforcing values of the preferred stimuli change. Although
a number of methods have been presented in the literature,
most preference assessments involve measuring relative inter-
action with each stimulus and ordering the stimuli hierarchi-
cally in terms of relative level of interaction. In general, stimuli
that are associated with higher levels of interaction (i.e., more
preferred) function as more effective positive reinforcers.
Once an FBA and a stimulus preference assessment are
conducted, a behavioral intervention is typically implemented.
Behavioral interventions typically include extinguishing the
problem behavior by stopping the delivery of the maintaining
consequence identified by the FBA. As noted above, withhold-
ing a reinforcer that has previously been shown to maintain
a response is a process referred to as extinction. In the context
of destructive behavior, extinction would involve no longer
delivering the reinforcer (identified in the FBA) following the
target behavior. For example, if a child's physical aggression
occurred because it resulted in others allowing him out of
school work, extinction would be arranged such that the
child would no longer be given a break from his work follow-
ing instances of aggression (i.e., the response no longer pro-
duces the reinforcer). An FBA is useful because it indicates
what variables to withhold during extinction. However, extinc-
tion as a solitary form of intervention is typically not recom-
mended due to the possibility of negative side effects (i.e., an
extinction 'burst' or an immediate worsening of behavior
before it decreases). Thus, most FBA-based behavioral inter-
ventions involve the combined use of extinction and a process
called differential reinforcement (DR). In a typical DR proce-
dure, the reinforcer that has been found to maintain destruc-
tive behavior is delivered following some alternative response
(or the omission of destmctive behavior) and destmctive
behavior is placed on extinction. A commonly used DR
approach is called functional communication training in
which the individual is taught an alternative, appropriate
form of communication which is reinforced while no longer
providing reinforcement for the target destmctive behavior.
Using the above example, the child might be taught that in
order to receive a brief break from working he must complete a
problem first. Finally, once a treatment is developed to reduce
the occurrence of destmctive behavior, the treatment must
be modified slightly such that it can be implemented in the
individual's natural environment (a process referred to as
treatment generalization). Again using the above example,
the treatment might be gradually modified such that the
child is required to complete one problem, then five problems,
ten problems, etc. before a break is given (all the while destmc-
tive behavior is placed on extinction).
ABA and Autism
In recent years, the term 'applied behavior analysis' has
become synonymous with a therapeutic approach targeting
autism. Autism is a neurological disorder associated with
Behavior Analysis 273
ritualistic and restricted engagement in activities and behaviors
as well as impairments or abnormalities in a child's develop-
ment in the areas of communication and social interaction.
Although a number of procedures have been presented as
potential treatments for autism, procedures based on the prin-
ciples of ABA have received the most empirical support.
When used to refer to the treatment of autism, 'applied
behavior analysis' is most often associated with a specific
method of instruction (called discrete trial instruction, DTI)
that was pioneered by Ivar Lovaas and colleagues at UCLA. The
procedures involved in DTI are firmly grounded in the princi-
ples of ABA; however, it is incorrect to use the terms DTI and
ABA as synonyms. That is, ABA as a field incorporates many
different procedures beyond those utilized in DTI for indivi-
duals with autism. Moreover, there are additional behavioral
procedures, also based upon principles of ABA, which are
effective for teaching autistic individuals. The basic compo-
nents of these procedures are described below.
In general, the strongest empirical support for the treatment
of autism favors the use of early and intensive behavioral
intervention (EIBI). In an EIBI program, the procedures from
the field of ABA are used (a) to reduce the occurrence of
problem behaviors associated with autism and (b) to replace
those behaviors with appropriate communication, social skills,
play skills, academic skills, and independent living skills. The
basic procedures employed in an EIBI program involve rear-
ranging the antecedent and consequent events in the environ-
ment such that appropriate behavior reoccurs and problem
behavior is extinguished (i.e., the 'selection by consequences'
paradigm presented by Skinner). Such treatment is developed
by a practitioner with specialized training in ABA and is typi-
cally implemented at a 1 : 1 therapist to child ratio for ~20-40 h
per week.
A number of empirical studies have found that ABA-
based programs are consistently more effective than eclectic
or play-based therapies and the use of ABA programs has
been endorsed by a number of professional organizations. A
critical review of the features and demonstrated efficacy of
ABA-based treatment approaches for autism is beyond the
scope of the current chapter; however, common procedures
that are based on principles of ABA include: (a) DTI, (b)
incidental teaching, (c) pivotal response training, (d) verbal
behavior training, (e) natural language paradigm, (f) peer-
mediated interventions, (g) precision teaching, and (h) error-
less learning.
ABA in Other Areas of Focus
Although the principles of ABA are typically used in educa-
tional and clinical settings for individuals with developmental
disabilities, these are not the only settings in which ABA has
been implemented. Applied behavior analysts have shown that
the same principles hold tme and are effective in changing
individuals' behaviors in organizational settings such as fac-
tories, restaurants, and retail. In addition, ABA has been shown
to be effective with different populations such as adults with
dementia, children with attention deficit/hyperactivity disor-
der, and athletes.
Conclusion
There are three branches of behavior analysis: behaviorism,
experimental behavior analysis, and ABA. Behaviorism is the
study of the philosophy and theories of behavior. The major
tenants of behaviorism which collectively make it unique are
that (a) behavior analysts view behavior as its own subject
matter; (b) that private events are the function of the same
environmental variables as public events; (c) that individuals,
not groups, behave; (d) behavior is a function ofvariables in the
environment; and (e) the study of behavior is a natural science.
The purpose of the branch of the EAB is to further develop the
basic principles of behavior analysis. The third branch of behav-
ior analysis is ABA and its purpose is to implement the basic
principles of behavior analysis in ways that result in socially
significant changes for individuals in communities. There are
seven dimensions of ABA, including: applied, behavioral, ana-
lytic, technological, conceptually systematic, effective, and gen-
eralizable. ABA is commonly implemented in educational and
clinical settings with individuals with disabilities and this term
is sometimes utilized (albeit inaccurately) as a label for a spe-
cific treatment for autism. However, the principles of ABA
extend beyond individuals with developmental disabilities.
ABA is an effective method of changing the behavior of many
populations across many settings.
See also : Autism and Pervasive Developmental Disorders; Behavior
Measurement in Psychobiological Research; Intellectual Disabilities;
Operant Conditioning.
Further Reading
Austin J and Carr JE (2000) Handbook of Applied Behavior Analysis. Reno, NV:
Context Press.
Baer DM, Wolf MM, and Risley TR (1968) Some current dimensions of applied behavior
analysis. Journal of Applied Behavior Analysis 1 : 91-97.
Cooper JO, Heron TE, and Heward WL (2007) Applied Behavior Analysis, 2nd edn.
Saddle River, NJ: Pearson Education.
Fisher WW, Piazza CC, and Roane HS (2011) Handbook of Applied Behavior Analysis.
New York: Guilford Publications.
Hanley GP, Iwata BA, and McCord BE (2003) Functional analysis of problem behavior:
A review. Journal of Applied Behavior Analysis 36: 147-185.
Howard JS, Sparkman CR, Cohen HG, Green G, and Stanislaw H (2005) A comparison
of intensive behavior analytic and eclectic treatments for young children with autism.
Research in Developmental Disabilities 26: 359-383.
Kazdin AE (1982) Single-Case Research Designs: Methods for Clinical and Applied
Settings. New York: Oxford University Press.
Lovaas 01 (1993) The development of a treatment-research project for developmental ly
disabled and autistic children. Journal of Applied Behavior Analysis 26: 61 7-630.
Maurice C, Green G, and Foxx RM (2001 ) Making a Difference: Behavioral Intervention
for Autism. Austin, TX: Pro-Ed, Inc.
Myers SM and Johnson CP (2007) Council on children with disabilities. Management
of children with autism. Pediatrics 120: 1162-1182.
Parsonson BS and Baer DM (1978) The analysis and presentation of graphic data. In:
Kratochwill TR (ed.) Single-Subject Research: Strategies for Evaluating Change.
New York: Academic Press.
Repp AC and Horner R (1 999) Functional Analysis of Problem Behavior: From Effective
Assessment to Effective Support. Belmont, CA: Wadsworth.
Skinner BF (1938) The Behavior of Organisms: An Experimental Analysis. New York:
Appleton-Century.
Stokes TF and Baer DM (1977) An implicit technology of generalization. Journal of
Applied Behavior Analysis 10: 349-367.
Behavior Genetics of Personality
S C South, Purdue University, West Lafayette, IN, USA
R F Krueger, University of Minnesota, Minneapolis, MN, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behavior genetics A scientific field that uses both
quantitative genetics and molecular genetics to understand
both the genetic and environmental contributions to
individual variation in human behavior.
Gene The basic unit of inheritance, a section of DNA that
codes for a particular product.
Gene-environment correlation When the genetic
influences on a phenotype are correlated with the genetic
influences on an environmental variable.
Gene-environment interaction When the effect of genes
on a phenotype depends on exposure to a certain
environment, and vice versa.
Heritability The proportion of total variance in a behavior
or trait that varies among individuals in the population that
can be attributed to genetic influences.
Nonshared environment Environmental experiences/
influences on a phenotype that differ between family
members.
Phenotype An observed characteristic of an individual,
resulting from a combination of genetic and environmental
influences.
Shared environment Environmental influences on a
phenotype that are shared between family members and
result in making them more similar.
Introduction
The characteristic ways that people think, feel, and behave
constitute personality, and personality varies among people.
Some people are more introverted, some more extroverted;
some enjoy trying new experiences, while others are cautious,
restrained, and stick to old habits. Psychologists have long
been interested in understanding the reasons for this variation.
One obvious place to look is the person's rearing environment,
particularly the family. If an individual is more introverted, we
might reasonably expect them to have, for instance, a mother
who is also more introverted. However, when mother and
child are biologically related, it becomes difficult to determine
whether the child is introverted because of the genetic similar-
ity to the mother, or if the learning processes and environ-
ments that the mother exposed her child to over the course of
the child's younger years are the relevant causal factors.
Methods in behavior genetics for modeling genetic and
environmental influences on human behavior are a fundamen-
tal tool for parsing out these possibilities. The field of behavior
genetics, with its focus on twin and adoption studies and
'biometrical modeling' of these kinds of family data, was vitally
important in demonstrating genetic influences on most behav-
ioral traits long before the genome was sequenced. We can now
say with reasonable certainty that individual differences in
personality are influenced by genetic and environmental fac-
tors. Some may ask whether, in the era of molecular genetics,
these biometric approaches are still necessary. In fact, biomet-
ric modeling of personality still has much to contribute, partic-
ularly as the field of molecular genetics struggles to find
replicable results linking measured genes to personality traits.
Recently, the field of behavior genetics has extended
earlier work that estimated the heritability (the magnitude of
genetic influence) of personality. New focus on forms of gene-
environment interplay - that is, how genetic and environmental
factors interact and correlate - may have important implications
for understanding the etiology of personality. In this article, we
provide a brief overview of biometric approaches, particularly
using twin sample methodology, and how this work informs
our understanding of the etiology of personality. We then pres-
ent extensions of twin modeling that have implications for
the structure of personality, links between personality and psy-
chopathology, and connections between personality and envi-
ronmental risk factors. Our goal is to provide the reader with a
broad overview of conceptual and statistical issues related to
the modeling of personality data in behavior genetics. Space
limitations prevent us from providing a more in-depth treat-
ment of the issues, but we refer interested parties to several
excellent texts on behavior genetics.
Genetic and Environmental Influences on
Personality Traits
Twin Studies
The most common method for examining genetic and envi-
ronmental influences on personality variation involves the use
of the twin study, thus we focus much of our discussion on this
method and the findings from twin studies. Twins are nature's
way of providing a natural experimental design to disentangle
the relative influence of genes and the environment. Twins come
in two varieties: identical (monozygotic, MZ) twins result from
the splitting of one fertilized egg, and thus MZ twins share
the same DNA/ genes. Fraternal (dizygotic, DZ) twins result
from two eggs being fertilized at the same time, and share on
an average, half of their segregating genes. Biometric modeling
takes advantage of the known degrees of genetic and environ-
mental similarity between MZ and DZ twins, using the concor-
dance between twin pairs to decompose the variance in a
phenotype into percentages as explained by genes relative to
environmental influence. Any phenotype in the human popu-
lation that differs among individuals (e.g., a personality trait)
274
Behavior Genetics of Personality 275
can be said to display 'phenotypic variance,' which is attributed
to both genetic and environmental influences.
The basic goal of a univariate ('one variable') twin study is
to estimate the relative influence of genes and environment.
Biometric models separate the variation in a phenotype into
three sources that, taken together, can be summed to account
for the total variance of the phenotype. The first source of
influence is genetic variance, or the amount that a phenotype
differs among people because of genetic differences in the
population. This source of influence is typically indexed with
the heritability statistic, h 2 , or the proportion of total variance
in the phenotype that is due to genetic influences. It is impor-
tant to note that heritability, like the commensurate estimates
for environmental influences, is a population-based statistic;
thus, if we say that the heritability of a personality trait is 50%,
we are saying that 50% of the variance in that trait in the
population from which we sampled is due to genetic differ-
ences among people in that population; not that 50% of any
one individual's personality is due to genes. If genetic influ-
ences completely explained all of the variation in a personality
trait, we would expect that the correlation between MZ pairs
would be 1.0, because MZ twins share all of their genes, while
the correlation between DZ pairs would be 0.5 because DZ
twins share ~50% of their genes.
The second source of influence is the shared or com-
mon environment, abbreviated c 2 . The shared environment
indexes the extent to which growing up in the same household
makes twins more similar to each other. Shared environment
includes things like socioeconomic status, traditions, customs,
habits, personality traits of rearing parents, and neighborhood
characteristics. If the shared environment completely explained
the variance in a personality trait, we would expect that both
MZ and DZ twins reared together would correlate 1.0 because
all twin pairs share the experience of having been reared
together, regardless of their zygosity. The nonshared environ-
ment indexes the extent to which twins are different despite
sharing the same household and genes. It is abbreviated e 2 and
is also referred to as unique environment. Examples of non-
shared environment include having friends, life experiences,
and traumas that are different from one's siblings. The estimate
of nonshared environment also contains any error or impreci-
sion of measurement, and will thus be inflated to the extent that
people are measured with less than perfect fidelity. Expectations
resulting from nonshared environment completely explaining
the variation in a personality trait would be MZ and DZ twin
correlations of 0.0, consistent with the idea that the nonshared
environment represents the uniqueness of a specific person.
Both shared and nonshared environmental effects are also
commonly discussed as proportions of variance, like the heri-
tability statistic. It is important to note that the distinction
between nonshared and shared environmental influences can
be subtle. Parenting practices could be conceived of as shared
between all siblings in the family, to the extent that the same
parenting style might make siblings more similar to each other
(shared environment), but perceptions of parenting may be
unique to each individual child and thus work to make chil-
dren growing up with the same parent less similar to each other
(nonshared environment).
Evidence from decades of work consistently shows that
almost any psychologically interesting phenotype has a genetic
component, a finding so ubiquitous that it has been called the
'First Law' of behavior genetics. This includes almost every
major personality trait or domain. Across varying samples,
cultures, population cohorts, and personality measures, the
heritability of any typical personality trait is around 50%,
with the rest of the variance coming mostly from nonshared
environmental influences. Very little personality variation can
be accounted for by the effects of the rearing family environ-
ment, beyond the contribution of shared genetics. Much of this
work has been conducted on traits from the Big Five personal-
ity domains (extraversion, openness, agreeableness, conscien-
tiousness, and neuroticism). In their review of the literature on
twin studies of the Big Five traits, Bouchard and Loehlin found
that the heritability of all five domains was around 40-50%.
The consistency of this finding across the traits, which differ
widely in the aspects of human behavior that they capture, is
compelling evidence of the ubiquity of genetic influences on
human individual differences.
Several points related to these findings are worth consider-
ing. First, moderate to sizeable estimates of genetic influence
have also been found for dimensional measures of pathologi-
cal personality and categorical personality disorder diagnoses.
Second, with only a few exceptions, (e.g., altruism), estimates
of shared environment tend to hover close to 0, suggesting that
influences from the family, beyond genetic inheritance, have
little or no impact on personality. However, we cannot rule
out the possibility that shared environmental influences act in
an interactive manner on personality, as opposed to having
a main effect. If interactional or transactional processes are at
work, such that some family-level factors enhance and some
suppress genetic effects on personality, the influence of such
factors could appear in either heritability or nonshared envi-
ronmental estimates.
Third, most heritability estimates are based on the
assumption of additive genetic effects, the idea that individ-
ual genetic contributions are fungible and 'add up' in their
influence on phenotypic variation. More complex genetic
effects (such as genetic contributions interacting with each
other, a phenomenon known as epistasis) are possible and
may also be important in understanding the underlying
molecular architecture of personality. Fourth, even though
a substantial portion of the etiology of personality is esti-
mated as nonshared environmental effects, research has
been relatively unsuccessful in linking personality variation
to specific environmental effects or risk factors, a topic we
return to below.
Finally, much of the work to date on personality and behav-
ior genetics has concentrated on adult twin samples using
self-report methods. More recently, research has extended
downward to child and adolescent samples, using both self-
and observer report of personality. One study, using a sample
of 9- 10-year-old twins, found substantial heritability estimates
for only two (Self-directedness and Harm Avoidance) of the
scales from the Junior Character and Temperament Inventory.
Notably, they found substantial shared environmental effects
for Novelty Seeking and Cooperativeness. A different study
used a sample of toddlers and found that the variation in
inhibitory control (IC), a dimension of temperament, was
38% genetic and 62% nonshared environment using observer
ratings, but 58% genetic, 26% shared environment, and 16%
276 Behavior Genetics of Personality
nonshared environment using parent ratings, suggesting both
developmental and rater effects may be important in under-
standing the etiology of personality.
Adoption Studies
Most behavior genetic research on personality has utilized the
twin study design, but adoption studies also provide a geneti-
cally informative perspective on the etiology of personality.
Families in which there is an adopted child, like the twin
design, provide a natural experiment for understanding genetic
and environmental influences on personality variation. Con-
sider, for example, a family in which there are both biological
children of the parents and adopted, nonbiologically related
children (thus, not a kinship adoption). The adopted child or
children will share 100% of their family environment with
their adoptive parents and nonbiological adopted siblings,
but 0% of their genes.
Adoption studies, like twin studies, find substantial non-
shared environmental effects and little influence of the shared
environment; however, they tend to produce smaller heritabil-
ity estimates than do twin studies. There are two major possi-
bilities for the discrepancy in findings between twin and
adoption studies. One possibility is that identical twins reared
together are highly correlated on phenotypes like personality
because they are raised in environments that encourage imita-
tion (e.g., being dressed alike by their parents). Twin studies
operate under the equal environments assumption (EEA),
which posits that the environments of MZ twins are no more
alike than DZ twins; if this assumption were violated, MZ twin
correlations could be inflated, thus increasing heritability esti-
mates. A powerful research design, focusing on twins reared
apart, has proved valuable in supporting the likelihood of the
EEA. Any resemblance between twins who are reared apart
cannot be traced to the shared environment. Studies of twins
reared apart conducted in Minnesota and Sweden find esti-
mates of heritability in line with what has been found in
studies of twins reared together.
Another possibility for the discrepancy in heritability esti-
mates between adoption studies and twin studies involves the
role of nonadditive genetic effects on personality. As noted
above, most twin studies assume that genetic effects are addi-
tive, and few find evidence of dominant or nonadditive genetic
effects. It is possible that adoption studies are better at identify-
ing nonadditive genetic effects, that is, effects due to interac-
tions among specific genes. The Nonshared Environment in
Adolescent Development (NEAD) study was important in fur-
thering our understanding of the role of nonadditive effects in
the variation in personality. This study combined information
from MZ and DZ twins, full siblings, half siblings, and geneti-
cally unrelated children in the same family resulting from
remarriage. Results indicated high heritability estimates, sig-
nificant shared environmental estimates, and little effect of
nonshared environment on broad domains of adjustment.
In a follow-up study, the authors found slightly lower herit-
abilities than in the original study, and confirmed the presence
of significant nonadditive genetic effects.
While, in general, adoption studies are commensurate with
twin studies in showing little evidence of the shared environ-
ment, recent findings suggest that the adoption design may be
important in elucidating influences of the family environment
on personality variation. A recent study used a large sample of
both biologically related and adoptive families to examine
genetic and environmental influences on a range of pheno-
types, including personality traits of disinhibition and negative
emotionality (NE). Both types of families consisted of parents
and two adolescent siblings, either biologically related or, in
the adoptive families, two adopted siblings or one adopted
sibling and one biological sibling. While variation in NE was
largely due to nonshared environment and genetic effects,
there was evidence of significant shared environmental influ-
ences (20%) on disinhibition. The findings support the use
of large samples of adoptive families to detect the presence of
systematic family influences on the etiology of personality.
Multivariate Modeling of Personality
In the following sections, we argue that the approaches in
behavior genetics, particularly advanced biometric models
applied to data from twins, adoptions, or other informative
behavior genetic designs (e.g., twin family), can help us in better
understanding the etiology of personality. We begin by eluci-
dating how biometric models can improve our understand-
ing of the structure and development of personality, before
turning to exploration of overlap with psychopathology.
Structure of Personality
Extending beyond the basic univariate twin modeling of per-
sonality, research has examined the overlap between genetic
and environmental influences on several personality traits con-
sidered together. Multivariate biometric modeling of personal-
ity variables allows a researcher to decompose not only the
variance within phenotypes, but also the covariance between
phenotypes, into genetic and environmental influences. These
multivariate biometric models have important implications for
understanding the structure and organization of personality.
For example, we know that at the phenotypic level, the facets of
the five factor model (FFM) domain of neuroticism (e.g., angry
hostility, anxiety) are correlated, thus supporting their use
as indicators of a higher-order factor; biometric modeling can
be used to determine if the genetic and environmental influ-
ences on these facets mirror the phenotypic factor structure.
Biometric modeling of multiple phenotypes (i.e., two or more
personality traits) works much like the analysis of single traits,
estimating heritability and environmental proportions of vari-
ance unique to each, in addition to the amounts shared in
common between the two phenotypes. Further, this type of
model estimates the genetic and environmental correlations
between different traits. A genetic correlation, for example,
indexes the amount of overlap between the genetic influences
on two different traits, ranging from 0 to 1.0. These genetic and
environmental correlations can then be subject to factor ana-
lyses, much like phenotypic correlations between traits.
Findings from this work provide a complex picture of
how genetic influences on specific domains of personality
operate. For example, the genetic structure of personality traits
from the FFM domains parallel the phenotypic structure; yet,
in addition to genetic influences on the higher-order factors,
Behavior Genetics of Personality 111
there are also genetic influences on more specific personality
traits subsumed under the broad domains that are not
accounted for by the higher-order genetic factors. Jang and
colleagues examined the genetic and environmental structure
of the FFM traits using data from a combined German and
Canadian twin sample. They found that each of the five
domains consisted of two genetic and two nonshared environ-
mental factors, and concluded that the FFM higher-order per-
sonality traits may not represent coherent psychological
entities. A subsequent study applied the same biometric multi-
variate models to data consisting of adjectives representing the
FFM domains in a nationwide sample of American twins. One
etiologically coherent common factor explained the variation in
Extraversion and Neuroticism, suggesting that these domains
came closest to being unitary personality constructs.
Personality over Time
Genetic and environmental influences can also be studied
longitudinally. For example, one study examined the contribu-
tion of genetic and environmental influences to the stability
and change in 'withdrawn' behavior from age 3 to 12. Substan-
tial genetic influences were found at all ages, particularly at age
3 and 7, and high genetic correlations over time indicated that
generally the same genetic influences were operating at all ages.
Only modest shared environmental effects were found at any
age, although they were slightly higher for girls than for boys
and explained more of the stability of withdrawn behavior over
time in girls (14% for girls vs. 4% for boys). Nonshared envi-
ronmental effects increased over time, but smaller nonshared
environmental correlations indicated that these effects were
different at different ages. The general point is, given longitu-
dinal data on twins, developmental processes can also be
parsed into genetic and environmental aspects.
Links to Psychopathology
Approaches in behavior genetics have also been used to exam-
ine the shared etiology of personality and psychopathology.
Considerable evidence points to the phenotypic associations
between personality traits and different psychopathological
syndromes. Behavior genetics provides useful tools for exam-
ining whether these associations are due to genetic overlap
between personality and psychopathology. Extensions of the
multivariate biometric model (described above) have also
been used to estimate genetic and environmental influence
shared in common between personality traits and symptoms
of mental illness. Numerous studies have shown shared genetic
influences between personality and different forms of psycho-
pathology, including neuroticism and mood disorders and
disinhibition and externalizing disorders.
Major theories of the links between personality and psycho-
pathology have been reviewed extensively elsewhere. In brief,
the model that has received perhaps the greatest support in
both the child and adult literatures suggests that personality
traits and different syndromes of psychopathology are lower-
order components of a higher-order spectrum. For example,
the internalizing-externalizing model of psychopathology
indicates that mood and anxiety disorders are indicators of
a higher-order spectrum of internalizing, while behavior
disorders of substance use and antisocial behavior and conduct
disorders form indicators of a higher-order externalizing factor.
The personality trait of neuroticism seems to fit well within the
internalizing factor, while the personality trait of conscien-
tiousness (reversed) fits well within the externalizing spectrum.
Two types of biometric models are particularly appropriate
for examining the spectrum model of personality and psycho-
pathology. The common pathways model posits that the covari-
ation in a set of variables form a single latent phenotype (e.g.,
personality and psychopathology variables as in the internaliz-
ing factor), and this latent phenotype is influenced by one set
of additive genetic, shared environmental, and nonshared
environmental effects. In contrast, the independent pathways
model specifies direct links to the personality and psychopa-
thology variables from genetic, shared and nonshared environ-
mental influences common to all the variables and unique
to each individual variable. Researchers have recently begun
to directly test these competing models. In a twin sample
of more than 600 adolescents, Krueger and colleagues found
that a common pathway model best explained the covariation
between adolescent antisocial behavior, conduct disorder,
alcohol dependence, drug dependence, and the personality
trait of constraint (reverse scored). This finding is supported
by parallel results finding a common factor model of externa-
lizing behavior. More recently, Singh and Waldman compared
the common pathway and independent pathway models to
explain the covariation among NE and childhood externalizing
disorders in a sample of 4- 17-year-old twins. Unlike Krueger
and colleagues, they found that the independent pathways
model provided the best fit to the data, positing that
the genetic influences common to both NE and externalizing
disorders may support the role of NE as an endophenotype
that lies between measured genes on the chromosome and
syndromes of mental illness.
Biometric modeling has also been applied to the inter-
nalizing spectrum. Kendler and colleagues concluded that
the genetic factor structure of the internalizing spectrum dis-
orders, not including any personality traits, closely paralleled
the phenotypic structure. A later analysis added neuroticism,
and found two overlapping genetic factors; one factor was
shared between neuroticism, major depression, generalized
anxiety, panic disorder, and the phobias, while the other
factor explained the covariation between major depression,
generalized anxiety, and panic disorder, independent of neu-
roticism. Finally, South and Krueger found support for a
one-factor model of internalizing with indicators of depres-
sion, generalized anxiety, panic attacks, and neuroticism;
biometric modeling found estimates of moderate genetic
influences and large nonshared environmental influences,
with smaller estimates of the shared environment.
Environmental Influences
Having firmly established the importance of genetic influences
on personality variation, the field of behavior genetics moved
forward with the goal of identifying and explicating the
role of environmental risk factors that impact the etiology of
personality. One important line of research was initiated
by David Rowe, showing genetic influences on putatively
278 Behavior Genetics of Personality
'environmental' measures. Genetic influences have now been
found across a variety of environmental measures and assess-
ment approaches. This seminal finding was valuable in
showing how people, to some extent, 'make' their own envir-
onments; thus, the quality of parent-child, sibling, and marital
relationships, for instance, is due, in part, to the genetically
influenced traits of the individuals in those relationships. This
is known as gene-environment correlation, or the degree to
which a genotype influences the likelihood of exposure to a
specific environment. To directly examine this hypothesis,
researches have modeled the genetic and environmental overlap
between personality and measures of the environment. As with
the multivariate personality research described above, a Cho-
lesky decomposition can be conducted with personality traits
and environmental variables to estimate any shared genetic
influences.
Research has shown that there are shared genetic influences
between personality traits and aspects of the environment.
In a sample of twins reared apart, Krueger and colleagues
showed that there was genetic overlap between personality
traits and aspects of the childhood rearing environment
as recalled in adulthood; further, these genetic influences on
personality accounted entirely for the genetic influences
on the environment, handler and colleagues later replicated
these results in a sample of twins reared together. Again, they
found genetic overlap between aspects of the rearing environ-
ment (i.e., support) and personality traits, using both self- and
peer-reports of a different personality measure than the prior
study. South and colleagues, in a sample of adolescent twins,
found genetic correlations between self-reported personality
factors of positive emotionality, NE, and constraint from
the Multidimensional Personality Questionnaire (MPQ) and
aspects of the adolescent's relationship with his or her parent
relationship. Together, these findings support the role of per-
sonality in shaping the nature of environment. Not only do
genetically influenced personality traits impact the perception
of the rearing environment, it appears that aspects of the
environment (e.g., the quality of the parent-child relation-
ship) are reactive to those same genetically influenced person-
ality traits.
Another possible form of interplay between personality
and the environment, beyond gene-environment correlation,
is called gene-environment interaction, and refers to the
fact that different environments might moderate the genetic
and environmental influences on personality. In other words,
whether or not genetic influences on personality traits are
expressed may depend on the context or experience of certain
environmental influences. Typically, when heritability is esti-
mated in a biometric model using twin samples, it is estimated
across the entire sample-specific population. Thus, the herita-
bility estimate averages over any differences that might occur
within certain subsamples in the population. Biometric mod-
eration models have been developed that allow for different
estimates of heritability depending on a person's standing
on an identified environmental moderator variable. Consider
that univariate and multivariate twin studies have, for many
years, utilized biometric models that examine whether there
are gender differences in heritability estimates. Biometric mod-
eration models extend this idea, but instead of using gender
they estimate whether heritability may differ depending on
environmental contexts like socioeconomic status, marital
quality, or the parent-child relationship. A recent study used
these biometric moderation models to examine gene X envi-
ronment interactions between personality traits and the par-
ent-child relationship in a sample of adolescent twins. They
found that the heritability of broad personality factors (posi-
tive emotionality and NE) varied as a function of the quality
of the parent-adolescent relationship, suggesting that how
much genes influence the variation in personality may depend
on the family context. This work is still in the early stages
and much more still needs to be done before we can say
with any great certainty whether gene-environment interplay
will be a major step forward in understanding the developmen-
tal, causal pathways to personality.
Summary
In this article, we reviewed the decades of research in behav-
ioral genetics that has consistently found moderately strong
estimates of genetic and nonshared environmental influences
on personality. Family influences, beyond genetic effects, have
been more difficult to find, although recent studies using
younger samples, different personality assessment methods,
and adoption studies provide some tentative evidence that
shared environment may be more important, particularly
in childhood, than previously thought. While nonshared envi-
ronmental effects have historically been substantial, researchers
have struggled to identify the specific environmental contexts or
risk factors involved in the etiology of personality. Biometric
moderation models that estimate gene X environment interac-
tion in the presence of gene-environment correlation may
provide one way of finally elucidating important environmen-
tal contexts. This has particularly strong implications for
finding measured genes (from a molecular genetic approach)
involved in personality variation. As the field of behavior
genetics of personality moves forward, there appears to be ripe
opportunities for collaboration with molecular genetics, with
methodologies from both fields helping to inform the other,
linking more refined personality inventories with measured
genes and the particular environmental contexts which modify
their expression.
Further Reading
Bouchard TJ Jr. and Loehlin JC (2001) Genes, evolution, and personality. Behavior
Genetics 31 : 243-273.
Buchanan JP, McGue M, Keyes M, and lacono W (2009) Are there shared
environmental influences on adolescent behavior? Evidence from a study of adopted
siblings. Behavior Genetics 39: 532-540.
Carey G (2003) Human Genetics tor the Social Sciences. Thousand Oaks, CA: Sage.
Fanous A, Gardner CO, Prescott CA, Cancro R, and Kendler KS (2002) Neuroticism,
major depression, and gender: A population-based twin study. Psychological
Medicine 32: 719-728.
Gagne JR and Saudino KJ (2010) Wait for it! A twin study of inhibitory control in early
childhood. Behavior Genetics 40: 327-337.
Hettema JM, Neale MC, Myers JM, Prescott C, and Kendler KS (2006)
A population-based twin study of the relationship between neuroticism and
internalizing disorders. American Journal of Psychiatry 163: 857-864.
Hoekstra RA, Bartels M, Hudziak JJ, Van Beijsterveldt TCEM, and Boomsma Dl (2008)
Genetic and environmental influences on the stability of withdrawn behavior in
children: A longitudinal, multi-informant twin study. Behavior Genetics 38: 447-461 .
Behavior Genetics of Personality 279
Isen JD, Baker LA, Raine A, and Bezdjian S (2009) Genetic and environmental
influences on the junior temperament and character inventory in a preadolescent
twin sample. Behavior Genetics 39: 36-47.
Jang KL, Livesley WJ, Angleitner A, Riemann R, and Vernon PA (2002) Genetic
and environmental influences on the covariance of facets defining the domains
of the five-factor model of personality. Personality and Individual Differences
33: 83-101.
Jang KL, Livesley WJ, and Vernon PA (1996) Heritability of the big five
personality dimensions and their facets: A twin study. Journal of Personality
64: 577-591.
Jang KL, McCrae RR, Angleitner A, Riemann R, and Livesley W (1998) Heritability of
facet-level traits in a cross-cultural twin sample: Support for a hierarchical model of
personality. Journal of Personality and Social Psychology 74: 1556-1565.
Johnson W and Krueger RF (2004) Genetic and environmental structure of adjectives
describing the domains of the Big Five Model of personality: A nationwide US twin
study. Journal of Research in Personality 38: 448-472.
Kandler C, Riemann R, and Kampfe N (2009) Genetic and environmental mediation
between measures of personality and environment in twins reared together. Behavior
Genetics 39: 24-35.
Kendler K and Baker JH (2007) Genetic influences on measures of the environment:
A systematic review. Psychological Medicine 37: 615-626.
Kendler KS, Prescott CA, Myers J, and Neale MC (2003) The structure of genetic and
environmental risk factors for common psychiatric and substance use disorders in
men and women. Archives of General Psychiatry 60: 929-937.
Krueger RF (2005) Continuity of axis I and axis II: Toward a unified theory of
personality, personality disorders, and clinical disorders. Journal of Personality
Disorders 19: 233-261.
Krueger RF, Hicks BM, and McGue M (2001) Altruism and antisocial behavior:
Independent tendencies, unique personality correlates, distinct etiologies.
Psychological Science 12: 397-402.
Krueger RF, Hicks BM, Patrick CJ, Carlson SR, lacono WG, and McGue M (2002)
Etiologic connections among substance dependence, antisocial behavior, and
personality: Modeling the externalizing spectrum. Journal of Abnormal Psychology
111: 411-424.
Krueger RF and Johnson W (2008) Behavioral genetics and personality: A new look at
the integration of nature and nurture. In: John OP, Robins RW, and Pervin LA (eds.)
Handbook of Personality: Theory and Research, 3rd edn., pp. 287-310. New York:
Guilford.
Krueger RF and Markon K (2006) Reinterpreting comorbidity: A model-based approach
to understanding and classifying psychopathology. Annual Review of Clinical
Psychology 2: 111-133.
Krueger RF, Markon K, and Bouchard G (2003) The extended genotype: The heritability
of personality accounts for the heritability of recalled family environments in twins
reared apart. Journal of Personality 71 : 809-833.
Krueger RF, South SC, Johnson W, and lacono W (2008) The heritability of
personality is not always 50%: Gene-environment interactions and
correlations between personality and parenting. Journal of Personality
76: 1485-1522.
Livesley WJ and Jang KL (2008) The behavioral genetics of personality disorder.
Annual Review of Clinical Psychology 4: 247-274.
Neale MC and Maes HHM (1999) Methodology for Genetic Studies of Twins and
Families. Dordrecht, Netherlands: Kluwer.
Plomin R, DeFries JC, McClearn GE, and McGuffin P (2001) Behavioral Genetics,
4th edn. New York: Worth.
Rowe DC (1981) Environmental and genetic influences on dimensions of perceived
parenting: A twin study. Developmental Psychology 17: 203-208.
Rowe DC (1 983) A biometrical analysis of perceptions of family environment: A study of
twin and singleton sibling kinships. Child Development 54: 416-423.
Singh AL and Waldman ID (2010) The etiology of associations between negative
emotionality and childhood externalizing disorders. Journal of Abnormal
Psychology 119: 376-388.
Slutske WS, Heath AC, Dinwiddie SH, et al. (1998) Common genetic risk factors for
conduct disorder and alcohol dependence. Journal of Abnormal Psychology 107:
363-374.
Slutske WS, Heath AC, Madden PAF, Bucholz KK, Statham DJ, and Martin NG (2002)
Personality and the genetic risk for alcohol dependence. Journal of Abnormal
Psychology 111: 124-133.
South SC, Eaton NR, and Krueger RF (2010) The connections between personality and
psychopathology. In: Millon T, Krueger RF, and Simonsen E (eds.) Contemporary
Directions in Psychopathology: Scientific Foundations of the DSM-Z and I CD-I I .
New York: Guilford.
South SC and Krueger RF (2008) Marital quality moderates genetic and environmental
influences on the internalizing spectrum. Journal of Abnormal Psychology 117:
826-837.
Tackett JL (2006) Evaluating models of the personality-psychopathology relationship in
children and adolescents. Clinical Psychology Review 26: 584-599.
Turkheimer E (2000) Three laws of behavior genetics and what they mean. Current
Directions in Psychological Science 9: 160-164.
Turkheimer E and Waldron M (2000) Nonshared environment: A theoretical,
methodological, and quantitative review. Psychological Bulletin 126: 78-108.
Widiger TA and Smith GT (2008) Personality and psychopathology. In: John OP,
Robins RW, and Pervin LA (eds ) Handbook of Personality: Theory and Research,
3rd edn., pp. 743-769. New York: Guilford.
Relevant Websites
http://www.personality-arp.org/-Association for Research in Personality.
http://www.bga.org/ - Behavior Genetics Association.
http://www.ists.qimr.edu.au/- International Society for Twin Studies.
http://www.vcu.edu/mx/ - Mx Webpage.
Behavior Measurement in Psychobiological Research
G J Norman and G G Berntson, Ohio State University, Columbus, OH, USA
J T Cacioppo, University of Chicago, Chicago, IL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cardiothoracic impedance A noninvasive technique for the
measurement of cardiothoracic blood flow by passing an
electrical current through thorax and recording and
analyzing changes in impedance (impediment to electric
current flow). This information can be used to derive
measures of cardiac contractility, which reflects alterations
in sympathetic nervous system activity as well as cardiac
output, which indexes cardiovascular status.
Cortisol A steroid hormone typically associated with the
stress response that has broad effects on metabolic,
cardiovascular, and immunological processes.
Galvanic skin response A method for measuring the
electrical resistance of the skin which can serve as an index to
sympathetic responses.
Heart rate variability The rhythmical fluctuation in heart rate
that serves as a measure of autonomic cardiac control.
Variations within particular band widths can then be used to
parse out parasympathetic cardiac control through the
quantification of the component of heart rate variability in the
respiratory frequency band (respiratory sinus arrhythmia).
Oxytocin A hypothalamic neuropeptide that plays an
important role in various social behaviors in animals and
humans.
Introduction
Human behavior is an emergent phenomenon that results
from complex neural and environmental determinants occur-
ring across multiple levels of organization. Psychologists have
long appreciated this fact, and the application of psychophysi-
ological measures to psychological research has expanded in
recent years. Multilevel analyses, such as those commonly used
in psychophysiological research, represent a subset of interdis-
ciplinary approaches where the measures, constructs, and the-
ories cover multiple levels of organization. The growth of
multilevel analysis has provided important insights into the
reciprocal interactions between psychological and neurophysi-
ological processes.
Multilevel analyses involve more than manipulating or
measuring variables at different levels of organization. If mul-
tilevel analyses are to provide more extensive models of human
behavior, they also require an explication of the mechanisms
connecting these variables. This, in turn, requires an apprecia-
tion for the nature of the potential mappings between levels
of organization. The unique scientific dialects used by the social
and biological sciences can increase the prevalence of category
errors, where seemingly parallel concepts from different levels
of analysis reflect only partially overlapping domains, rather
than representing a one-to-one isomorphism. Although the
inherent complexity of psychophysical relations can present
unique scientific challenges, multilevel analysis capitalizes on
such complexity through a mutual calibration process whereby
knowledge at one level of analysis can refine and inform another
and reduce the incidence of category errors.
In this article, we highlight some features of contemporary
neurobehavioral measurement within the context of multilevel
analysis and provide a framework for applying such analyses to
the study of human behavior.
Some Principles of Multilevel Measurement of Behavior
In this section we highlight some of the conceptual founda-
tions of the multilevel measurement of phenomena relevant to
human behavior and provide the rationale, and the potential
limitations, for conducting such broad-based research.
The principle of Multiple Determinism postulates that pro-
cesses occurring at one level of organization (e.g., psychologi-
cal) will have multiple antecedents within and across levels
of analysis. For example, individual differences in the psycho-
logical and behavioral manifestations of perceived stress are
dependent on both genetic and environmental factors. Conse-
quentially, isolated analysis of behavioral output may provide
information on psychological processes, but the precise deter-
minants of such responses are not necessarily inherent in this
behavior. Because of the multiple determinants across even
proximal levels, the mappings across more divergent levels of
analysis become increasingly complex. Although the ultimate
goal of multilevel analysis is to bridge levels of analysis, the
Corollary of Proximity suggests that this effort may be more
straightforward for more closely related levels since the com-
plexity of the mappings increases geometrically as one moves
linearly across levels of organization. Skip levels, and the map-
pings can become impenetrably complex. This is not to suggest
that bridging across broader levels of analysis is not possible
or desirable; rather, it is a realization that the establishment
of smaller bridges may be more efficient and can provide the
foundation for more expansive bridge-building efforts.
The principle of Nonadditive Determinism reflects the fact
that the properties of the whole cannot always be predicted
by the properties of its parts. The sources of variance from
higher-level processes are often broader than those for lower
levels of organization; so, higher-level systems tend to be more
complex. The increased complexity inherent in behavioral and
280
Behavior Measurement in Psychobiological Research 281
psychological processes can limit the utility of reductionistic
approaches in the full explication of complex phenomena. For
example, while understanding the contribution of isolated
genes to psychiatric conditions can be quite useful, in order
to understand pleiotropic effects of genetic networks one
needs to understand how the gene products interact with
each other and attempt to incorporate this emergent behavior
into more comprehensive models. Thus, phenomena of inter-
est should include observations at the highest level of organi-
zation at which it manifests, as it may not be understood by
appeal exclusively to a priori understandings derived solely
from lower levels of analysis, a concept termed the Corollary
of Asymmetry. The corollary of asymmetry does not deny the
utility of measurement at strictly lower-level analyses (e.g.,
cellular processes), but does assert that a single level of analysis
may not universally provide the optimal description of scien-
tific phenomena.
Lastly, the principle of Reciprocal Determinism asserts that
there may be mutual, reciprocal influences among levels of
organization where the direction of causation or explanation
need not always flow from lower- to higher-level phenomena.
One may be tempted to view causation as an entirely bottom-
up process where lower-level operations (e.g., genes) strictly
determine activity at higher levels (cellular, organ system, or
behavior). While such bottom-up strategies have provided
countless scientific breakthroughs, they do not provide a com-
plete picture. For example, environmental influences can
reprogram the expression of particular genes, which can then
be passed on to subsequent generations through epigenetic
inheritance. Therefore, because causal influences among levels
can be bidirectional, the Corollary of Interdependence states
that a single level of analysis may not yield a comprehensive
account of multilevel phenomena, and that no single level of
analysis applies uniformly. Importantly, this is not to suggest
that research conducted within individual levels is not important
as multilevel research is necessarily dependent on the work done
within single-level analysis. However, the corollary suggests that
multilevel analysis may be necessary for a comprehensive under-
standing of phenomena as complex as human behavior.
Application of Psychophysiological Measurement
Psychophysiological measures of neurobehavioral processes
provide a unique perspective on human behavior. Indeed,
psychophysiological measures provide a global view of the
emergent behavior of integrated neurophysiological systems
that present a picture of neurobehavioral systems that may
not be apparent through the isolated measurement of molecu-
lar or molar processes. Generally, neurophysiological measures
are of interest primarily to the extent to which they allow
one to index a psychological process (Figure 1). For example,
the analysis of sympathetic and parasympathetic regulation
of cardiac functioning is widely used to index the activity of
neurobiological processes associated with psychological states
including anxiety, attention, and arousal. In this regard, differ-
ential patterns of sympathetic and parasympathetic activation
of the heart can provide valuable information on the neuro-
biological and psychological processes that control autonomic
output. In the following section, we highlight some applica-
tions of psychophysiological measurement within the context
of multilevel analysis.
Figure 1 Summary of commonly used measurement of psychophysiological measurement. Environmental stimuli modulate neurobiological
processes, which subsequently modify psychobiological processing that can be indexed through imaging technologies (e.g., fMRI), cardiac autonomic
control measurement, skin conductance alterations, and variations in stress hormone levels.
282 Behavior Measurement in Psychobiological Research
Measurement of Autonomic Nervous System Function
in Neurobehavioral Processes
The theoretical underpinning of multilevel analysis to human
behavior has existed for decades with contemporary views
beginning to take shape at the turn of the nineteenth century.
In his essay "Evolution and dissolution of the nervous system,"
Jackson (1884) laid the foundation for multilevel characteriza-
tions of neuroarchitectural developments. Jackson argued that
the emergence of higher leveis of neuronal organization does
not involve a replacement of lower levels. Instead, higher-level
brain structures (e.g., cerebral cortex) can be characterized by a
hierarchical organizational pattern composed of simple reflex-
like circuits at the lowest levels (brainstem and spinal cord)
and distributed neural networks for more integrative computa-
tions at higher levels. Together these interacting hierarchical
structures allow neural systems to respond to environmental
and neurophysiological perturbations via rapid low-level pro-
cessing (e.g., startle reflex) allowing more rostral neural sub-
strates sufficient time to develop strategic responses.
Early hierarchical models of autonomic control were per-
haps most visibly articulated by Walter Cannon in the early
twentieth century. Cannon's work revolved around the concept
of 'homeostasis' which proposed that a primary role of auto-
nomic nervous system (ANS) was maintaining the constancy
in vital neurophysiological processes (e.g., blood pressure).
Cannon proposed that the sympathetic nervous system served
as the primary modulator of homeostatic processes with
activity in the parasympathetic branch serving to fine-tune
end-organ responses. Based in part on Cannon's work, para-
sympathetic and sympathetic outputs, which generally have
opposing actions, have been considered to be reciprocally
regulated by central systems. Reciprocal activation patterns,
where activity in one autonomic branch increases and the
other decreases, are often observed in reflex responses where
the reciprocal changes in the independent branches act in
concert to expand dynamic control of end-organ responses.
In addition to hierarchical patterns of autonomic con-
trol, descending neural projections are capable of regulating
lower-level neural structures such as the brainstem and spinal
cord through the activation of descending projections that
bypass intermediate hierarchical structures. This heterarchical
neuroarchitectural organization likely affords higher-level neu-
ral structures the ability to control autonomic and behavioral
output patterns independent of intermediate processing levels.
Importantly, heterarchical models have all the components of
hierarchical systems but have the additional capacity to orches-
trate complex behavioral responses through direct control over
widely separated levels via direct connections.
Heterarchical control is apparent in central autonomic
regulation. Although baroreflex responses typically entail recip-
rocal patterns of autonomic control, the autonomic branches
are capable of independent activation. This has necessitated an
expansion in the simple reciprocal bipolar model of autonomic
control, in which autonomic states are considered to lie along
a single continuum with maximal sympathetic (and minimal
parasympathetic) activity at one end and maximal parasympa-
thetic (and minimal sympathetic) activity at the other. Although
this bipolar model may apply to reflexive brainstem circuits, the
greater flexibility of more rostral neural systems, together with
long heterarchical pathways, requires an expanded model of
autonomic control that can account for the coactivation (or
coinhibition) of the two autonomic branches. Additionally,
there may be far greater individual differences in autonomic
responses arising from the operations of neural systems
controlling complex psychological and behavioral outputs.
At a group level, the heart rate responses of human subjects
to orthostatic stress and to standard psychological stressors
(mental arithmetic, speech stress, reaction time task) are similar.
Analysis of the separate contributions of the two autonomic
branches by the use of pharmacological blockades revealed
that activation of the baroreflex by orthostatic stress (transition
from sitting to standing) yielded a rather consistent response
across subjects, characterized by reciprocal autonomic output
(sympathetic activation and parasympathetic withdrawal). How-
ever, psychological stress is capable of producing more varied
autonomic responses across subjects, with no overall correlation
between the responses of the autonomic branches (Figure 2).
Figure 2 Orthostatic versus psychological stress. Left panel: Group mean responses to orthostatic and psychological stressors depicted as response
vectors on the autonomic plane, from baseline (intersection of horizontal and vertical dotted lines) to the stress conditions (arrowheads). Axes
depict millisecond of heart period change related to sympathetic and parasympathetic control. Right panel: Individual response vectors to the
psychological stressors revealing individual differences in the direction of response. Reprinted from Berntson GG, Cacioppo JT, Binkley PF, Uchino
BN, Quigley KS, and Fieldstone A (1994) Autonomic cardiac control. III. Psychological stress and cardiac response in autonomic space as revealed by
pharmacological blockades. Psychophysiology 31: 599-608, with permission from Blackwell.
Behavior Measurement in Psychobiological Research 283
The importance of the measurement model is illustrated by
a study of a population-based sample in Chicago (Chicago
Health and Social Relations Study, CHASRS). Based upon a
bipolar model of autonomic control, a cardiac autonomic
balance (CAB) metric was derived as a scale extending from
maximal sympathetic activation on one end of a continuum to
maximal parasympathetic activation at the other extreme (sym-
pathetic minus parasympathetic control). While CAB was not
associated with most aspects of health in a population-based
sample (CHASRS), it was predictive of diabetes mellitus
(Figure 3). In contrast to reciprocal models of autonomic
activation, as noted earlier, contemporary models of psycho-
physiological processes emphasize the bivariate structure of
autonomic regulation and the overall capacity for autonomic
flexibility and variability. Accordingly, an index of cardiac
autonomic regulatory capacity (CAR) was also derived as the
sum of activities of the autonomic branches. In contrast to
the reciprocal diagonal represented by CAB, CAR is a metric
that captures the coactivation or independent activation of the
autonomic branches. Analysis of the CHASRS sample revealed
that CAR was a better predictor of overall health status and was
a significant predictor of the occurrence of myocardial infarc-
tion (Figure 3). These results suggest that distinct patterns of
autonomic control may be associated with distinct health
dimensions.
Electrodermal Skin Response
In addition to the analysis of cardiac functioning described
above, estimates of autonomic output can be determined
through the measurement of electrodermal activity (EDA).
The basis for such measurement derives from the innervation
of the eccrine sweat glands by the sympathetic nervous system.
Importantly, the eccrine sweat glands are outside the control of
parasympathetic responses and this fact allows for relatively
unbiased estimates of sympathetic output. EDA is widely used
within psychological contexts as it is thought to reflect varia-
tions in emotional and neurophysiological arousal. Thus, EDA
has been used as an objective index of affective processes, in
controlled experimental contexts where extraneous variables
are accounted for, and has been shown, for example, to be
readily elicited by threatening stimuli.
Electrodermal and other psychophysiological measures are
also influenced by higher subcortical and cortical brain areas
associated with attention, anxiety, and decision making. These
areas include the amygdala, anterior cingulate cortex, and pre-
frontal cortex among others, and EDA responses can be elicited
by electrical stimulation of each of these structures. This
neuroautonomic coupling confers utility on psychophysiolog-
ical measures as markers of psychological processes. In fact,
visceral/autonomic responses may not only reflect but also
impact psychological processes. Since the writings of William
James before the turn of the twentieth century, somatovisceral
feedback has variously been speculated to be important in the
evocation or regulation of emotion and behavioral guidance.
Performance in decision-making tasks is associated with antic-
ipatory EDA responses prior to risky or suboptimal choices,
and this reactivity antedates conscious awareness of the out-
come contingencies and has been suggested to guide learning
and future adaptive choices. In this paradigm, damage to
Sympathetic (-PEPz) — *-
CAR
Figure 3 Left panel: Bipolar versus bivariant representations of
sympathetic and parasympathetic control. Right panel: CAR and CAB in
disease states. Data points illustrate means and standard errors of CAR
and CAB as a function of participant group, relative to the population.
Compared to other participants, subjects with a prior myocardial
infarction (Ml) had lower CAR scores, indicating lower overall cardiac
regulatory capacity, but were not highly deviate on CAB. In contrast,
those with diabetes showed a lower CAB score, reflective of a
predominant sympathetic balance, but were not highly deviant on CAR.
Adapted from Berntson GG, Norman GJ, Hawkley LC, and Cacioppo JT
(2008) Cardiac autonomic balance versus cardiac regulatory capacity.
Psychophysiology 45(4): 643-652, with permission from Wiley.
ventromedial prefrontal cortex results in dysregulated emo-
tional processes associated with abnormalities in strategic
decision-making characterized by a failure to change disadvan-
tageous behavioral patterns. While typical individuals alter
their behavior to avoid punishment, patients with ventrome-
dial prefrontal damage continue using the same pattern of
behavior, even if this behavior is repeatedly associated with
punishing outcomes. These patients also display abnormalities
in the generation of EDA responses that manifest as a reduction
284 Behavior Measurement in Psychobiological Research
in normal anticipatory arousal in the decision-making task.
The defect in anticipatory autonomic arousal may contribute
to the behavioral deficit, as somatovisceral afference may nor-
mally serve to bias emotional behavior and sculpt strategic
decision-making processes. Consistent with Jackson's view
of evolutionary representation of function, the association
between sympathetic output and decision making is likely
the result of higher-level (e.g., prefrontal cortex) structures
'listening' to more primitive lower-level processes that reflect
rapid changes in neurophysiological states in response to envi-
ronmental stimuli. Measurement of EDA or other psychophys-
iological responses may thus provide a window into the
psychological processes that reflect the activity of neurobiolog-
ical structures important in integrating emotion and cognitive
responses.
Serum Factors and Behavior
Hormones are in continuous reciprocal interaction with the
operations of the central nervous system. The acute elevation
of stress hormones represents an adaptive process that serves to
coordinate metabolic, cardiovascular, and behavioral responses
to environmental and neurophysiological perturbations. How-
ever, when stress and subsequent release of stress hormones
becomes chronic, it can predispose individuals to a myriad of
pathophysiological processes. The most well-studied stress hor-
mone system is the hypothalamic-pituitary-adrenal (HPA) axis.
The HPA consists of multiple peptide and hormonal signaling
molecules that orchestrate broad alterations in metabolic
functioning. The cascade is initiated by the detection of threat
stimuli leading to an increase in corticotropin-releasing hor-
mone (CRH) within the hypothalamus. The increased CRH
then induces the systemic release of adrenocorticotropin-
releasing hormone (ACTH) from the anterior pituitary, which
subsequently increases the release of cortisol (or corticosterone
in rats and mice) from the adrenal cortex. Cortisol levels are
continuously monitored by nearly every cell of the body through
activation of intracellular glucocorticoid receptors. Finally, the
system is subject to servo control, as it is inhibited by high levels
of cortisol binding at receptors in the hippocampus, which serve
to inhibit further production and release of CRH.
Given the broad effects of HPA function on metabolic,
immunological, and neurobiological functioning, psychophy-
siologists routinely quantify circulating levels of cortisol and
ACTH as an index of psychological stress. While this research
has provided numerous important empirical and theoretical
insights into the interface between physiology and psychology,
important interpretative caveats exist for the application of such
measures to psychophysiological science. For example, circulat-
ing levels of ACTH and cortisol are released in a pulsatile manner,
necessitating the measurement across multiple time points, or
other approaches to provide an integrated metric. Furthermore,
circulating levels of HPA hormones display large diurnal varia-
tions with highest levels typically observed in the early morning,
requiring attention to the timing of measurements. In addition,
alteration in diurnal variation in HPA axis output may itself be a
relevant metric, as it has been found to be sensitive to various
psychopathological conditions, such as depression and anxiety.
Another common method for analyzing HPA axis function
is the dexamethasone suppression test (DST). Dexamethasone
is a synthetic glucocorticoid that, when given to healthy indi-
viduals, activates the negative feedback servo mechanisms
described above, leading to significant reductions in circulating
levels of endogenous glucocorticoids. In comparison to a sin-
gle measure of glucocorticoid levels, the dexamethasone sup-
pression possesses high construct validity and minimizes
many of the measurement issues described earlier. Individuals
suffering from any number of psychopathological conditions
such as depression show diminished negative feedback indica-
tive of glucocorticoid insensitivity. The inability to appropri-
ately control HPA axis functioning can result in chronic
elevations in circulating hormone levels, which have broad
effects on structure and function in various brain regions asso-
ciated with cognitive and affective processes. Furthermore,
chronic levels of cortisol can result in the downregulation of
glucocorticoid receptors in immune cells. Since cortisol has
potent anti-inflammatory effects, the phenomena of glucocor-
ticoid resistance can diminish the ability to regulate systemic
inflammatory processes, which can have a negative influence
on psychological and neurophysiological health outcomes.
An additional methodological consideration stems from
the fact that the majority of circulating cortisol is bound
to corticosteroid-binding globulin. Bound cortisol is unable
to access intracellular glucocorticoid receptors, rendering it
biologically inactive. Thus, measurements of circulating corti-
sol levels reflect both free and bound states making interpreta-
tion of such data more difficult. However, because binding
proteins do not readily enter the salivary glands, salivary mea-
sures primarily reflect unbound cortisol. Furthermore, as com-
pared to serum measures, measurement of salivary cortisol is
easily collected and thus can be used across wide contexts and
time points.
In addition to classical stress hormones, the neuropeptide
oxytocin (OT) has recently become the focus of extensive
research in the field of psychophysiology as a result of its
potent and selective influence on social processes. OT is
synthesized within the hypothalamus and released systemi-
cally and centrally. The release of OT within the brain is
thought to prime neurobiological networks for social behavior.
OT has been implicated in the social behavior of various
species, including monkeys, rats, and humans. Indeed, func-
tional OT signaling is necessary for social recognition within
rodents. Central administration of OT can induce maternal
behavior in virgin rats and increases social interaction between
adults. Furthermore, the role of OT in human social behavior
has recently become a topic of intense interest as it has been
shown that the oxytocinergic system is remarkably sensitive to
the neural processing of social information. The experience of
early childhood neglect, for example, impairs the increase in
peripheral OT levels typically triggered by mother-infant inter-
action and early parental separation has been proposed to alter
central OT receptor sensitivity. In contrast, positive social in-
teractions, such as social support, lead to increased levels of
OT, and OT has been suggested to mediate the well-known
beneficial effects of social support on stress responsivity and
health, a topic discussed in further detail later. Intranasal
administration of OT has been shown to attenuate amygdala
activity and reduces amygdala-brainstem coupling in response
to threatening social stimuli. Similarly, OT modulates the eval-
uation of socially relevant faces by influencing the activity of
Behavior Measurement in Psychobiological Research 285
A In arousal ratings to threat
Figure 4 Oxytocin effects on threat ratings. Mean (SEM). Prepost
change scores in emotional arousal to threatening stimuli. Oxytocin
selectively decreased emotional arousal to human but not nonhuman
threatening stimuli. Reprinted with permission from Norman GJ,
Cacioppo JT, Morris JS, Malarkey WB, DeVries AC, and Berntson GG
(2010) Selective influences of oxytocin on the evaluative processing of
social stimuli. Psychopharmacology , Epub ahead of print.
the amygdala. Recently, we have demonstrated that OT has
potent and rather selective influences on affect as it decreases
emotional arousal to threatening human but not animal sti-
muli (Figure 4). The effects of OT on neurobiological func-
tioning translate into observable changes in behavior. Indeed,
intranasal OT increases trust, independent of risk taking, and
improves the ability to correctly infer the mental state of others
and increases the duration of gazes toward the eye region of
faces, an important behavior for the recognition of emotion in
social interactions. Furthermore, OT receptor gene variants are
associated with loneliness, empathy, and autism spectmm dis-
order. The recent expansion of research on the role of OT in
human social behavior, with improvements in OT assays, may
provide novel perspectives on psychophysiological relations.
Psychophysiological Relations
As discussed earlier, technological advances have allowed for
ever more precise measurement of psychophysiological rela-
tions. However, even with the use of more sophisticated mea-
sures, the question arises as to how signals are to be interpreted
and inferences drawn. Psychophysiological processes represent
interactions between psychological (x/j) and neurophysiological
(ip) domains. Psychophysiological processes entail multiple log-
ical relationships between elements, and this fact has important
implications in inferences derived from psychophysiological
measures. Important to the measurement of psychophysiologi-
cal processes, psychological relations can be characterized by five
general categories of mappings between elements in the psycho-
logical and neurophysiological domains (Figure 5).
I. No significant association between psychological and
neurophysiological domains.
II. One-to-one: psychological domain is associated with one,
and only one, neurophysiological process.
w
c
o
ra
®
o>
o
o
CL
o
Psychology Physiology
Figure 5 Schematic of psychophysiological relations consisting of (I)
No significant association between psychological and neurophysiological
domains. (II) One-to-one: psychological domain is associated with one,
and only one, neurophysiological process. (Ill) One-to-many: in which
elements in psychological domain are related to more than one element
in the neurophysiological domain. (IV) Many-to-one: in which two or
more psychological elements are associated with a single element in the
neurophysiological domain. (V) Many-to-many: where two or more
psychological elements are associated with overlapping
neurophysiological elements.
III. One-to-many: in which elements in psychological domain
are related to more than one element in the neurophysio-
logical domain.
IV. Many-to-one: in which two or more psychological ele-
ments are associated with a single element in the neuro-
physiological domain.
V. Many-to-many: where two or more psychological ele-
ments are associated with overlapping neurophysiological
elements.
As presented above, the first category includes relations that
represent irrelevant sources of variance in psychophysiological
associations and need to be accounted for in order to avoid
spurious associations between psychological and neurophysiolog-
ical domains. The second category includes one-to-one relation-
ships where the psychological domain is associated with one, and
only one, neurophysiological process. Next is the one-to-many
relation where elements in the psychological domain are related
to more than one element in the neurophysiological domain.
The fourth domain is the many-to-one relationship where two
or more psychological states are associated with a single neuro-
physiological process. The final category is the many-to-many
relationship where two or more psychological elements are asso-
ciated with overlapping neurophysiological elements.
286 Behavior Measurement in Psychobiological Research
Unfortunately, one-to-one relationships are not com-
mon in psychophysiology, whereas relationships involving
one-to-many (category III) and many-to-many (category V)
are common. The latter are far more problematic for interpret-
ing psychophysiological relations. One approach to this prob-
lem is to reconceptualize the neurophysiological elements of a
one-to-many relationship as composed of a combination of
psychological elements. Similarly, the consolidation of neuro-
physiological elements of many-to-many relationships into
particular patterns of activation may serve to simplify relations
into a many-to-one or even one-to-one relationship. However,
caution should be applied in such instances as the inherent
complexity of psychophysiological relations may not always be
reducible to one-to-one or even one-to-many relationships.
Appreciation of the distinct psychophysiological relations,
along with their strengths and weaknesses, is vital for the
appropriate interpretation of data involving broad levels of
analysis, and maximizes the utility of such measures by limit-
ing the occurrence of logical inconsistencies.
Inference in Psychophysiological Measurement
An overarching goal behind the measurement of psychophysi-
ological relations is the elucidation of basic relationships
between the mind and the body. With multilevel analysis
come particular methodological and theoretical challenges asso-
ciated with the complexity of interactions between levels rang-
ing from the molecular to the psychological. However, advances
in psychophysiology have provided important insights into
human behavior that can serve as the basis for future multilevel
analysis. Regardless of the precision of the particular methodol-
ogy used, the interpretation of such broad-based data continues
to be challenging and rife with pitfalls. Generally, the measure-
ment of neurophysiological processes is useful to the extent to
which they reflect psychological processes.
The table above details the classes of relations that may
obtain between a psychological process (i/i) and neurophysio-
logical measure (p). In a hypothesis regarding the indexing of a
psychological process (i/') by a neurophysiological measure (p)
we seek the relation if (p) then (1/1), but that only holds with a
one-to-one relationship. However, as mentioned above, one-
to-one relations are rare in psychophysiological science. Causal
hypotheses regarding the involvement of a brain structure (p)
in a particular psychological processes (i/r) are of the form i j/=f
(ip). This logical relationship postulates that the occurrence of
ip (e.g., amygdala activation) is a sufficient condition for i/i
( fear), but it does not imply that i/i is necessarily preceded by
<p. Furthermore, the activation of particular brain structures is
often used as an index of a psychological state. Thus, in this
instance the inference is of the form i j/ =/ (ip) rather than not-
\jj =/ (not-i/>). Consequently, the goal of psychophysiology
can be characterized by the conditional probability of i jj given
<p:P(il/\tp)= 1 .
One common approach that attempts to address some of
the logical issues discussed above is the subtractive method,
where experimental and control conditions differ by one ele-
ment with the particular outcome measure reflecting the
impact of that single element. Through the application of
this method, neurophysiological correlates of psychological
processes can be identified and concepts and theories can be
refined and tested. The application of imaging technology
(e.g., functional magnetic resonance imaging, fMRI) is one
example of this methodology. Indeed, differences in neural
activation patterns under experimental and control conditions
are taken to reflect the activity specifically associated with
the experimental condition. While this approach provides a
unique perspective into psychophysiological relations, it does
not allow for the determination of causal relationships, or the
application of strong inference.
One potential issue with the interpretation of the data result-
ing from the subtractive method revolves around the fact that
researchers sometimes treat their results as if neurobiological
substrates were being manipulated directly (e.g., inhibition of
prefrontal cortex) and psychological outcomes measured (e.g.,
elevations in depression). Indeed, causal hypotheses regarding
the involvement of a brain structure (p) in a particular psycho-
logical process (t (/) are of the form i j/=f (p). This logical rela-
tionship implies that the occurrence of p (e.g., amygdala
activation) is a sufficient condition for ij/ (fear), but it does not
imply that 1/1 is necessarily preceded by p. Furthermore, the
activation of particular brain structures is often used as an
index of a psychological state. However, with the application
of ever more powerful imaging technologies, many researchers
succumb to the temptation to use the inverse logical argument
generally in the form of P(p |i/') = ? Psychophysiological mea-
sures provide information on p as a function of i ji, but the
conditional probabilities are equivalent [P(ilt\p) = P(p\xl/)\ only
if there is a one-to-one relationship between the neurophysio-
logical element p and the psychological element i/c
In the interpretation of psychophysiological data, isomor-
phism between a psychological variable and neurophysiologi-
cal variables (one-to-one relationship) must be directly tested,
and deviations from isomorphism should be accounted for.
In this regard, interdisciplinary, multilevel approaches to
the measurement of psychophysiological variables are uniquely
positioned such that they can provide complementary and con-
verging evidence from direct manipulations of the neuro-
physiological variables (p as independent variable). Using this
methodology, one can bolster the confidence in the interpreta-
tion of the data by demonstrating that specific patterns of neural
activation consistently covary across a broad range of conditions
and contexts. However, while the covariation of neurophysiolog-
ical and psychological realms may increase the confidence of
inference of i/i from p, causal inferences can only be determined
through the application of converging evidence. For example, if a
clinician observes an increase in amygdala activation in response
to aversive stimuli across various contexts, and wants to deter-
mine whether or not a causal relationship exists, then the dem-
onstration that manipulations of amygdala functioning (e.g.,
stimulation, ablation) modulate fear behavior in mice would
allow for more valid inferences of causal mechanisms. Further-
more, with recent advances in transcranial magnetic stimulation
technology as well as the application of deep brain stimulators to
treat psychiatric disease has come the increased ability to apply
more in-depth research into the causal mechanisms mediating
psychophysiological relations. Thus, the application of converg-
ing lines of evidence can routinely deal with many of the weak-
nesses inherent in the subtractive method while still maintaining
the strengths of this approach.
Behavior Measurement in Psychobiological Research 287
Conclusion
The use of measures spanning multiple levels of analysis has
become increasingly common in contemporary psychological
and neuroscientific research. Advancements in technology and
conceptual understanding of human behavior will likely
increase this trend. The expansion of knowledge and measure-
ment of the molecular, systems, and psychological concomi-
tants of human behavior will necessitate a similar increase in
the application of methodologies that appropriately integrate
facts across multiple levels of analysis. While such developments
are important, it is imperative to couch these new measurements
within a theoretical framework that appreciates multilevel anal-
ysis while simultaneously avoiding category errors through an
understanding of particular limitations inherent in the measure-
ment used. Thus, the future of psychophysiological research is
promising, and as technological and theoretical constructs con-
tinue to advance across levels of analysis, so too will our under-
standing of psychophysiological processes.
See also: Clinical Assessment; Psychopathology: Diagnosis,
Assessment, and Classification.
Further Reading
Berntson GG, Bigger JT Jr., Eckberg DL, et al. (1997) Heart rate variability: Origins,
methods, and interpretive caveats. Psychophysiology 34(6): 623-648.
Berntson GG, Norman GJ, Hawkley LC, and Cacioppo JT (2008) Cardiac autonomic
balance versus cardiac regulatory capacity. Psychophysiology 45(4): 643-652.
Cacioppo JT and Decety J (2009) What are the brain mechanisms on which
psychological processes are based? Perspectives on Psychological Science 4:
10-18.
Cacioppo JT, Hawkley LC, Crawford LE, et al. (2002) Loneliness and health: Potential
mechanisms. Psychosomatic Medicine 64(3): 407-417.
Cacioppo JT, Lorig TS, Nusbaum HC, and Berntson GG (2004) Social neuroscience:
Bridging social and biological systems. In: Sansone C, Morf CC, and Panter AT
(eds.) The Sage Handbook of Methods in Social Psychology. Thousand Oaks, CA:
Sage.
Cacioppo JT and Patrick B (2008) Loneliness: Human Nature and the Need tor Social
Connection. New York: W. W. Norton and Company.
Critchley HD (2005) Neural mechanisms of autonomic, affective, and cognitive
integration. Journal of Comparative Neurology 493(1 ): 1 54-1 66.
Cunningham WA and Zelazo PD (2007) Attitudes and evaluations: A social cognitive
neuroscience perspective. Trends in Cognitive Science 11(3): 97-104.
Dantzer R, O’Connor JC, Freund GG, Johnson RW, and Kelley KW (2008) From
inflammation to sickness and depression: When the immune system subjugates the
brain. Nature Reviews: Neuroscience 9(1): 46-56.
Davidson RJ (2003) Affective neuroscience and psychophysiology: Toward a synthesis.
Psychophysiology 40(5): 655-665.
Heinrichs M, von Dawans B, and Domes G (2009) Oxytocin, vasopressin, and human
social behavior. Frontiers in Neuroendocrinology 30(4): 548-557.
Lieberman MD (2007) Social cognitive neuroscience: A review of core processes.
Annual Review of Psychology 58: 259-289.
Malarkey WB and Mills PJ (2007) Endocrinology: The active partner in PNI research.
Brain, Behavior and Immunology 21(2): 161-168.
Norman GJ, Cacioppo JT, and Berntson GG (2009) Social neuroscience. Wiley
Interdisciplinary Reviews: Cognitive Science 1(1): 60-68.
Norman GJ, Cacioppo JT, Morris JS, Malarkey WB, DeVries AC, and Berntson GG
(2010) Selective influences of oxytocin on the evaluative processing of social
stimuli. Psychopharmacology: Epub ahead of print.
Thayer JF and Siegle GJ (2002) Neurovisceral integration in cardiac and emotional
regulation. Engineering in Medicine and Biology Magazine 21(4): 24-29.
Behavioral Economics
S Dhami and A al-Nowaihi, University of Leicester, Leicester, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behavioral game theory The application of limited
rationality to strategic human interaction.
Discount factor A function of time that captures the degree
to which current consumption is more important than
future consumption.
Heuristics Rules of thumb used by individuals that
substitute for fully optimizing behavior based on classical
statistics.
Intrinsic reciprocity The desire to reciprocate based on the
notion that one is hardwired to do so.
Liberalism versus paternalism The study of the policy
response to less than perfect rationality.
Loss aversion The tendency of losses to bite more than
gains of an equivalent magnitude.
Neuroeconomics The use of the methods of neuroscience
in economics.
Probability weighting function A function that captures
the subjective probability weight assigned by an individual
to various events.
Reference point A level of outcome such that any outcome
which is numerically greater is coded as a gain and any that
is numerically lower is coded as a loss.
Social preferences The presence of bundles of goods
of others, in addition to one's own, in one's utility
function.
Utility function A function that assigns to bundles of goods
and services, some level of human satisfaction that can be
numerically measured.
Introduction
Traditional economics is based on the assumptions of fully
rational and self-interested behavior. A very large body of
evidence casts serious doubts on both these assumptions.
A serious consideration of the psychological evidence for
human motivation in economic phenomena and the con-
struction of new theoretical models based on this evidence
has given rise to behavioral economics. This new field is
growing at a rapid pace and gaining increased acceptance in
mainstream economics. In this essay, we provide a brief over-
view of behavioral economics and some of the insights that
it offers.
Decision Theories
For simplicity, let x 1; x 2 , . . ., x„ be real numbers, which we
interpret as different levels of wealth (n can be any positive
integer). We assume that we can rank the wealth levels in the
following manner: x 1 < x 2 < ■ ■ ■ < x„. Economists use the term
lottery for the following entity
£ = (,x 1 ,p 1 -,x 2 ,p 2 -r--;x n ,pn) [1]
where outcome, x f , occurs with probability, p,. So, in the lottery
£, outcome Xi arises with probability p lr outcome x 2 arises
with probability p 2 , and so on. We assume that the probabil-
ities are nonnegative and sum up to one. Thus, a lottery is
simply a description of all possible outcomes that could arise,
along with their respective probabilities.
When the probabilities are verifiable independently of
the decision maker, we have a situation of risk. On the
other hand, if the probabilities can be inferred only from
the actions of the decision maker, then we have a situation
of uncertainty. For instance, while objectively verifiable prob-
abilities of the risk from the MMR (measles, mumps, and
rubella) vaccine were shown to be low in almost all scientific
studies, many individuals assigned high subjective probabilities,
leading them to refuse the vaccination. Denote the set of all
lotteries of the form in eqn [ 1] by £. Which lottery in £ should
be chosen?
Expected Utility Theory
The most widely used theory in economics under risk is the
expected utility (ELI) theory. Under axioms on individual pre-
ferences, known as the axioms of rationality, it can be shown
that, for each decision maker, there exists a utility function u
such that one may think of u(x,) as the utility or satisfaction
arising from the outcome x,. u(x,) is simply some real number.
The utility function has the following main properties, (i) If
the individual strictly prefers an outcome x ; to another out-
come Xj then it must be the case that u(x,) > u(xy), and (ii) if
Xi > Xj then u(x,) > u(xf). Furthermore, for any lottery £ = (x 1(
Pi ; x 2 , p 2 ; ■ ■ ■ ; x„, p„), the expected utility to the decision maker,
denoted by EU(iC), is defined as follows.
EU(£) = J2'i =1 PMxi) [2]
In eqn [2] we have used the shorthand notation f= i p;u(x,) =
piufxf) + p 2 u(x 2 ) + • • • + p n ii(x n ). This is standard notation and
we shall use it throughout this essay so it is best if the reader
who has not encountered this before invests in understanding
it at this point.
The decision maker is assumed to choose the lottery that
gives the highest expected utility. EU allows for a rich pattern of
attitudes to risk. However, EU incorporates risk preferences
through the shape of the utility function (rather than through
the probabilities).
288
Behavioral Economics 289
Definition 1 Risk aversion
A decision maker is said to be risk averse if he/she prefers the
expected value of the lottery to playing the lottery. The con-
verse behavior is risk loving and indifference between the two
is referred to as risk neutrality.
It can be shown that risk-neutrality/risk-aversion/risk-
seeking are equivalent, respectively, to the utility function, u,
being linear/concave/convex, respectively. Risk aversion, aris-
ing from a concave utility function, underpins important
explanations of economic phenomena in most areas of eco-
nomics. These include the study of insurance, the design of
contract schemes in principal-agent problems, the optimal
rates of taxes, determination of wage rates etc. A lottery L p
first-order stochastically dominates lottery L q if the distribution
function of L p lies everywhere below that of L q . Decision
makers under EU do not choose stochastically dominated
options.
The axioms of rationality that underpin EU appear intuitive
and plausible. However, they are only assumptions about
human behavior. In particular, they should not be confused
with the notions of rationality as used in logic, psychology, or
law. And, in fact, a large number of rigorous tests have rejected
these axioms. This induced a major effort to propose auxiliary
assumptions to EU in order to save it, but without much success.
stochastic dominance. Following the insights of RDU, PT was
modified to cumulative prospect theory (CP). CP does not
allow for the choice of stochastically dominated options.
Since RDU is a special case of CP, it suffices to explain CP.
In CP (like PT but unlike EU and RDU) the carriers of utility
are not final levels of wealth but deviations of these from a
reference point; a well-established idea from psychology. For
instance, dip your right hand in cold water and your left hand
in warm water. After a few seconds put them both in lukewarm
water. Your right hand then feels warm and your left hand feels
cold, although they are both gauging the same temperature.
The status quo turns out to be a useful reference point in many
applications. But some other possible choices could be an
expected level, a fair level or a legal entitlement. If the outcome
turns out to be superior (respectively inferior) to the reference
point, then the decision maker is in the domain of gains
(respectively losses).
CP also incorporates experimental and field evidence,
which suggests that utility is evaluated differently in the
domain of gains and losses. First, losses bite more than equiv-
alent gains (loss aversion). Second, the utility function is con-
cave for gains and convex for losses. Thus, in each domain,
there is diminished sensitivity to gains/losses. Tversky and
Kahneman propose the following (axiomatically founded)
utility function that captures all these facts.
Behavioral Decision Theories
Behavioral economics provides several alternative, axiomati-
cally founded, theories that explain the evidence better than
ELI, yet are tractable. The EU formula in eqn [2] has two salient
features. (1) It is linear in probabilities. (2) Utility is defined
over final wealth levels. Behavioral alternatives have relaxed
both these assumptions.
Rank-dependent utility (RDU) relaxes the linearity in prob-
ability assumption. The most satisfactory decision theory in
economics, and the one with the strongest psychological foun-
dation, is the Nobel Prize winning work on prospect theory
(PT) due to the seminal paper in 1979 by Daniel Kahneman and
Amos Tversky. However, PT allows for violations of first-order
v(x)
( x y if x > 0
X(— x)° if x < 0
[3]
where x is the deviation of some outcome relative to the
reference point, y, 6, X are constants; 0<y<l, O<0<1.
X > 1 is known as the coefficient of loss aversion. Experimental
results initially suggested that y and 8 are approximately equal
to 0.88 and X is approximately equal to 2.25. However,
subsequent research showed that the parameter of loss aver-
sion is variable and depends on the context. The properties of
the value function are illustrated in Figure 1. Notice that the
reference point is at zero and the function is relatively steeper
for negative values. It is also concave for positive values but
convex for negative values.
Figure 1 The utility function under CCP.
290 Behavioral Economics
The second main feature of CP (which it shares with PT and
RDU but not ELI) is that probabilities are transformed in a
nonlinear manner, using the device of a probability weighting
function (PWF), w. One may think of w(p) as a real number,
lying between zero and one, that captures the subjective weight
assigned by the individual to the probability p. A PWF has the
property that the individual assigns a higher subjective weight
to higher probabilities, and
w(0) = 0, w( 1) = 1 [4]
that is, a probability zero is assigned a subjective weight of zero
and a probability one is assigned the maximum probability
weight of one. Many PWF's have been proposed in the litera-
ture. The most satisfactory is the one by Drazen Prelec, which
depends on two parameters and takes the following form
w(p) = exp {—/?[— In p]“}, 0<p<l, ct > 0, fl > 0 [5]
A plot of the Prelec function is shown in Figure 2. For the
particular parameter values chosen in the plot, it is consistent
with the empirically verified inverse S-shape.
Let w + , uC, respectively, be the PWF's for the domain of
gains and losses. There is evidence that w + (p ) = = w(p).
Suppose that the reference income level is y 0 and the individual
is presented with the following lottery (see our explanation of a
lottery above):
T — (y_ ffl , p—m 5 Y- m+ 1 , P—m+1 j ’ ‘ '! XO/ Poj Xl' Pi i X2' Pi] ' ' 3 Ynr Pn)
[ 6 ]
where x_ m <x- m + i <■ ■ ■ <Xo <Xi < Y 2 < ■ ■ ■ <Yn are the m + n
outcomes such that there are m outcomes in the domain of loss
(incomes lower than yo) and n outcomes in the domain of
gains (incomes higher than y 0 ).
Linder PT or CP, individuals care about the level of income
relative to the reference point. Define income relative to the
reference point by
x i = Xi - yo; i = — m, — m + 1, . . . , 0, 1, 2, . . . , n. [7]
Subtracting income relative to the reference income from
each of the income levels in the lottery in eqn [6] and using
eqn [7] we get the following lottery, which we denote by L
L = [x— m , p— m i x~ m + 1, p— m+i ; . . . ;Xo,po',Xi,pi;X2,p2', ■ ■ • i x n ,pn)
[ 8 ]
P
Figure 2 The Prelec function for (3=1 and a=0.5.
Define £ P as the set of all possible lotteries of the form
given in eqn [8], Under PT and CP, decision makers wish to
evaluate lotteries from the set £ P . They do so by forming
an appropriate value function. Definition 2 shows the value
function for the lottery L.
Definition 2 The value function under CP
The value of the lottery Le^Cp to the decision maker is given by
V{L) = [9j
Compare eqn [9] with eqn [2], First we have used the
summation notation in eqn [9] that we have explained above
in the discussion following eqn [2] . Second, the utility function
in eqn [9], i/(x,), is defined so as to have the properties
illustrated in eqn [3] above, while the utility function in
eqn [2], ufa), is not restricted to have these properties. Third,
utilities are multiplied with probabilities in eqn [2] while in
eqn [9] they are multiplied with decision weights, 7q, i = — m,
— m + 1, . . ., 0, 1, 2, . . ., n. The construction of decision weights
is crucial in all versions of prospect theory (e.g., PT, CP, and
CCP, which we describe below). We first formally describe
the construction of decision weights in Definition 3, followed
by some intuition. We shall use the summation notation,
which we have described in the text following eqn [2] above.
Definition 3 For CP, the decision weights, n i: are defined as
follows:
n„ = w + [p„)
"i = M' + (E " = iPj) - w + (£”=i+iPj)'
tt— m Ml (P— jh)
T^j — U’ (yi — _ m Pi) W (£ ; i= -mPi).
Recall that 1/(0) = 0. Hence it is immaterial what value is
chosen for n 0 . This is why n 0 is undefined in Definition 3. The
method of constructing decision weights in Definition 3 was
originally suggested in a brilliant undergraduate thesis by John
Quiggen. The idea is that the decision weights transform prob-
abilities in a cumulative fashion using the probability weight-
ing function. The decision weight in Definition 3, associated
with some outcome is the marginal increase in the probability
weight as one moves from that outcome to the next best
outcome. Decision makers who form decision weights in this
manner do not choose stochastically dominated options. This
solved an important shortcoming associated with PT.
RDU is the special case of CP when all outcomes are in
the domain of gains. ELI is the special case of RDU when, in
addition, w(p)=p (ot = yS = 1 in eqn [5]).
CP allows for a rich set of attitudes to risk, that depend not
only on the curvature of the utility function but also on the
shape of the PWF. A fourfold classification of risk is predicted
by CP. In particular, decision makers are risk averse for mod-
erate probability moderate gains (as in EU) but (unlike EU)
they are risk seeking for low probability high gains. By
contrast, decision makers are risk averse for low probability
high losses (as in EU) but (unlike EU) are risk seeking
for moderate probability moderate losses. This fourfold pat-
tern of attitudes to risk is confirmed in laboratory experiments
and explains a wide range of economic phenomena that EU is
either unable to explain or finds it difficult to explain. See, for
instance, the Asian disease example in the section on mental
Behavioral Economics 291
accounting below for separate attitudes to risk in the
domain of gains and losses.
One criticism made in the recent work by al-Nowaihi and
Dhami is that all standard PWF's used in RDU and CP are
extremely steep at the end points (in the sense that lim p ^ 0 (it>
(P)/P) = 00 < and limp-.jdl -ie(p)]/[l — p]) = oo). This makes
very low and very high probabilities extremely salient to deci-
sion makers and leads RDU and CP to make predictions
inconsistent with a range of important observed phenomena.
For example, CP (and RDU) predicts far too much demand for
insurance (as compared to the evidence) against low probabil-
ity natural hazards, such as earthquakes, floods, and hurri-
canes. Another example is that CP (and RDU and EU) predicts
that the imposition of the severest possible punishments (e.g.,
capital punishment), even with vanishingly small probability of
detection, would deter all crime. This is contrary to the evidence.
Other similar violations come from a range of other phenomena
such as red traffic light running, driving, and talking on mobile
phones and not taking up adequate health checks.
In recent research, al-Nowaihi and Dhami account for
these problems by proposing composite cumulative prospect
theory (CCP). This differs from CP in that it uses a composite
Prelec probability weighting function (CPF), illustrated in
Figure 3. It is made up of segments from three Prelec functions
over the respective ranges,
Very low
probabilities are hugely underweighted in the sense that
lim p _ 0 (ii’(p)/p) = 0. This allows CCP to predict human behav-
ior, for instance, toward insurance and law enforcement, that
is in better conformity with the evidence. CCP is much more
general than the account that we have given here; it can do
everything that EU, RDU, and CP can do, but the converse
is false.
Collectively, PT, CP, and CCP successfully explain a wide
range of otherwise unexplained phenomena. These include
why it is hard to find a taxi on a rainy day in New York,
asymmetric price elasticities, tax evasion problems, demand
for insurance, the Allais paradox, the St. Petersburg paradox,
paradoxes in the economics of crime and punishment, the
endowment effect, the disposition effect, the equity premium
puzzle, asset pricing, incentive schemes, etc.
P
Figure 3 The composite prelec weighting function (CPF).
Heuristics and Biases
The heuristics and biases research program commenced with
seminal papers by Tversky and Kahneman in the early 1970s.
Its basic premise is that individuals are not fully rational, but
they are not irrational either. Tversky and Kahneman took the
middle ground that individuals solve problems by relying on
some fast (in terms of time) and frugal (in terms of informa-
tion acquisition and processing) heuristics. Because such heur-
istics do not optimize in the classical sense, the performance of
the heuristics is not necessarily optimal.
For instance, individuals often judge distance of an object
by the heuristic of how clear it is, sharper objects are thought to
be closer. This heuristic is fast and frugal. While this heuristic
often works well, it might also result in biased and misleading
perceptions. For instance, when conditions are hazy, distances
might be overestimated. Conversely, when conditions are very
clear, distances might be underestimated.
The heuristics and biases approach demonstrates a system-
atic departure of actual behavior from that prescribed by the
laws of classical statistics. Rather, individuals employ a range of
judgement heuristics. The point or interval estimates arising
from these judgement heuristics are often in conflict with
classical inferences. Even statistically sophisticated researchers
or experienced decision makers often rely on these judgment
heuristics.
An important heuristic is the representativeness heuristic.
When individuals use this heuristic they tend to assume that
the properties of a small sample that they are familiar with is
characteristic of the large population from which the sample
was drawn. For example, an eminent colleague told us that up
to quite late in his childhood he believed that most people in
the world were Jewish members of the British Communist
Party, because that was the circle of family and friends he
grew up in. This is sometimes referred to as the law of small
numbers. At other times, individuals exhibit the reverse bias,
conservatism, that is, they underestimate the likelihood of a
sample from a given distribution. For example, people typi-
cally underestimate the likelihood of getting a consecutive run
of 3 heads in 10 random tosses of a fair coin. These heuristics
explain many puzzles such as the gambler's fallacy (avoid
betting on a number that has come up last time), hot hands
and winning streaks (because a basketball player has scored the
last few shots, he is even more likely to score this one).
The behavior of humans is inconsistent with Bayes' rule,
especially when they are presented with probabilistic informa-
tion (as opposed to frequency format information). In partic-
ular, individuals engage in base rate neglect. For example,
consider the following question. "There is a rare condition
whose probability is one in a million. There is a test whose
success rate is 99%. You tested positive. What is the probability
that you have this condition?" The typical answer is 99%. The
true answer based on classical statistics is 0.01% (the reader
can check this by applying Bayes' rule).
Anchoring is one of the most robust and important heur-
istics that people use. In situations where people do not know
the answer to a question, they are unduly influenced by an
initial suggestion or anchor. Interestingly, the anchor could be
completely random and unrelated to the answer. For example,
a roulette wheel is spun. The resulting number is, say, 50.
292 Behavioral Economics
The subjects are then asked "how many countries are members
of the UN?" The answers would then frequently be found to
cluster around 50. Furthermore, once individuals fix an anchor
in their minds, they adjust too slowly, and insufficiently,
toward the target.
There is also a range of other judgement heuristics that the
space constraint does not permit us to discuss further. These
include hindsight-bias, false consensus, attribute substitution,
conjunction fallacy, availability heuristic, affect heuristic, and
taking a necessary condition to be sufficient.
Mental Accounting and Framing
The idea of mental accounts is a significant and important part
of behavioral economics with wide ranging implications.
Richard Thaler, whose name is most closely identified with
this concept in economics, defines mental accounting as fol-
lows. "I wish to use the term mental accounting to describe the
entire process of coding, categorizing, and evaluating events."
Elsewhere, he provides another definition that is useful in
financial situations: "mental accounting is the set of cognitive
operations used by individuals and households to organize,
evaluate, and keep track of financial activities."
Let us begin with two features of mental accounts.
1. Money is not fungible across the accounts. So, for instance,
one might put aside money in a separate account for chil-
dren's education. This would be unnecessary if money were
fully fungible across mental accounts.
2. The objective of the individual is not necessarily to maxi-
mize total financial wealth. An important objective that is
highlighted in the mental accounts literature is to limit the
size of loss in individual accounts.
Much of traditional theory makes the frame-invariance
assumption. By contrast, individual decisions are sensitive to
the framing of problems. In a famous experiment, Kahneman
and Tversky use a representative sample of physicians and
pose them the following question framed in two alternative,
but classically equivalent, ways that, respectively, stress the
positive and the negative aspects. The question posed was
the following.
Example 1 Imagine that the United States is preparing for the
outbreak of an unusual Asian disease, which is expected to kill
600 people. Two alternative programs to combat the disease
have been proposed. Assume that the exact scientific estimates
of the consequences of the programs are as follows:
Positive Framing: If program A is adopted, 200 people will
be saved. If program B is adopted, there is a one-third proba-
bility that 600 people will be saved and a two-thirds probabil-
ity that no people will be saved. Which of the two programs
would you favor?
Under positive framing, the solution is framed in terms of
'lives saved.' The result was that 72% of physicians chose A, the
safe strategy and 28% chose program B, the risky strategy.
Negative Framing: If program C is adopted, 400 people will
die. If program D is adopted, there is a one-third probability
that nobody will die and a two-thirds probability that 600
people will die. Which of the two programs would you favor?
For negative framing, only 28% of the physicians voted for
the safe strategy C while 72% of them opted for D, the risky
strategy.
The problem in each of the two frames (positive and nega-
tive) is identical. Thus, if A is preferred to B then C must
be preferred to D (or vice versa). While the aim of Kahneman
and Tversky was to show that individuals are risk averse for
gains and risk loving for losses (see the discussion of prospect
theory, above), the reframing of the same question leads
to completely different choices. Framing matters, and it can
determine which mental account an item of expenditure or
income is put into.
The evidence from models of consumption suggests that
the marginal propensity to consume out of distinct mental
accounts could be different. Thus, the effect of salary composi-
tion, for instance, between wage and bonus could have an
effect on the employee's saving decision. Different pension
plans that have otherwise identical effects in standard econom-
ics on account of perfect fungibility of money have distinct
effects in the presence of nonfungibility and different marginal
propensities to consume from different mental accounts.
Hence, in the mental accounting framework, the effects of
pension plans depend on which mental account is used to
pay for the pension contributions.
Why does the ticking of the taxi meter reduce one's pleasure
from the taxi ride? Why do we normally prefer to prepay for a
vacation? Why do we often prefer to pay for consumer durables
in instalments? These simple facts fall outside the ambit of the
workhorse life cycle model in macroeconomics. Prelec and
Lowenstein argue that consumption that has already been
paid for can be enjoyed as if it were free in the current period,
which potentially increases its current enjoyment, hence,
explaining prepayment of vacations. Similarly, the thought of
future payments reduces current utility from the consumption
of a good/service (hence, the unpleasantness arising from a
ticking taxi meter). Consumer durables on the other hand lead
to a stream of future benefits, often for several years, after the
date of purchase. Hence, paying in instalments for them is not
aversive as compared to a vacation, which is enjoyed for a small
duration of time. The framework suggested by Prelec and
Loewenstein can also explain the growing use of debit cards,
use of tokens at casinos, dependence of mortgage payments
on property values, popularity of fixed fee payment schemes,
Club Med vacations and gift giving among couples, etc.
Individuals tend to break down complex problems
into simpler ones that require less cognitive ability. Read,
Loewenstein and Rabin term this as choice bracketing. An
army General who focuses only on the battle at hand (narrow
bracketing) might lose sight of winning the war (broad brack-
eting). Recall the problem of finding a taxi on a rainy day in
New York. One possibility is that taxi drivers set a daily target
for earnings (narrow bracketing) rather than an income target
over a longer period (broad bracketing). The target is more
quickly achieved on a busy rainy day leading to too many
quits, too soon. The optimum for a taxi driver is, however, to
work longer during rainy days because customers are more
plentiful and the effective wage, higher.
Choice bracketing also takes the form of deciding on the
horizon over which one would evaluate one's portfolio. So, for
Behavioral Economics 293
instance, if one holds stocks and evaluates one's portfolio
too often then one would observe too many upward and
downward movements in the return to equities relative to
bonds, creating feelings of anxiety. Equities might then
require a higher premium, relative to the predictions of
standard theory, if they are to be held in a portfolio. This
explains an important puzzle, the equity premium puzzle,
namely, an excess of return on equities over bonds that
cannot be justified on the grounds of risk alone. Choice
bracketing can also explain strategies used at alcoholics
anonymous and the desire for risk aggregation. Alcoholics
under treatment are usually asked to take one day at a time
(narrow bracketing) because the possibility of lifetime absti-
nence (broad bracketing) is too daunting. Similarly, when a
range of risks are aggregated (broad bracketing) they might
well seem less onerous because a low outcome under one
form of risk could be mitigated by a high outcome under
another form of risk.
In standard economic theory, actual choice is the result of
well-defined initial preferences and well-defined subjective
probability distributions over uncertain events. By contrast,
Ariely, Loewenstein, and Prelec show that the initial framing
of problems or situations can have an inordinate effect on ones
initial preferences. In turn, this determines the individual's
subsequent choices. One is then led to question the very
notion of humans having well-defined initial preferences and
subjective probability distributions over events. This helps to
explain a range of phenomena, such as the initial hype before
mega sporting events, or the advertisement blitz before the
introduction of a new product.
An assortment of other mental accounting phenomena
exists. An important category of such problems arises when
individuals take actions to (mentally) justify sunk costs
that have been incurred in the past. Hence, there is a tendency
to hold on to loss making stocks for too long and sell winners
too early. The reason is that individuals create a mental account
for a particular category of expenditure. Sunk costs push
this mental account into the red, which causes anxiety. Indivi-
duals then take wasteful actions to avoid closing accounts that
are in red.
Recent research has also highlighted the mental accounting
of goals. Individuals often form goals and targets, for example,
completing term papers, referee reports, quitting an addiction,
losing a prespecified amount of weight by a certain date and so
on. While goals can be motivational, they can also be counter-
productive. On account of loss aversion, missed goals impose
large losses, and the individual might also engage in emotional
repair following a missed goal, etc.
Mental accounting has also been applied to retirement
savings. Of particular importance is the role of status-quo
options. It has been found that if the default is that the indi-
vidual is signed-into a pension plan (as opposed to the
converse) then the take-up of pension plans increases. The
number of options offered in an investment plan is also impor-
tant. Too many options can make choice more difficult. Indi-
viduals also use the diversification heuristic to choose among
alternative pension vehicles, which motivates individuals to
spread their money evenly among the available options even
when the prediction of standard economics is to engage in an
uneven spread.
Emotions
In economics, it has become standard to focus on emotionless
deliberation. However, much recent research shows that this
view is not supported by the evidence. Emotions are central to
humans and affect much of human deliberation.
We often worry about an uncertain future. Will my pension
contributions be sufficient for my retired life? Will early retire-
ment turn out to be a good decision? Should I buy or sell my
stocks in the current economic environment? These and other
temporal economic decisions produce a feeling of apprehen-
sion and unease: We shall simply refer to these feelings as
anxiety. Anticipation of a (stressful) future event can create
anxiety that in some cases is worse than the event itself. Think,
for instance, of an impending visit to a dentist for treatment.
Such feelings of anxiety have two features: (1) they are caused
by anticipation of future events, and (2) they are aversive in the
sense that we dread such future events. As a result, one might
delay a visit to the dentist, with detrimental consequences
for health.
Future events need not, however, always be unpleasant. We
might savor a pleasurable experience such as waiting for a
dessert at the end of a meal. Or, we might delay a pleasurable
vacation. Pleasant events provide examples where an individ-
ual seems to exhibit a negative discount rate; a preference for
the future rather than the present. Insofar as feelings of anxiety
and savoring impact on intertemporal choice in important
ways, we need to amend the standard model.
In one model by Caplin and Leahy, uncertainty is intro-
duced into a model of anticipal utility. In an interesting appli-
cation they derived equilibrium prices of assets whose returns
are ex-ante random. The randomness in the returns creates a
current feeling of anxiety. Hence, the ownership of a risky stock
reduces its price and increases its return relative to the no-
anxiety case and relative to assets whose returns are not too
random, such as bonds. This provides one possible explana-
tion for the equity premium puzzle.
Most dynamic models assume that future tastes are known.
However, tastes can change over time due to a variety of
reasons, for example, habit formation, moods, social environ-
ment, maturity etc. Projection bias arises when individuals
make predictions of their future tastes that are not realized in
equilibrium. Evidence shows that people are disproportion-
ately influenced by their current tastes in making predictions
of their future tastes. For instance, when ordering food at the
beginning of a meal we must predict how hungry we will be at
the end of the meal. We often overestimate our future hunger
and order more than we can eat. One can also make forecasting
errors in mainstream economics but, in the dominant rational
expectations view, there are no such errors in the absence of
random shocks and, on average, such errors are zero. Both
features are violated under projection bias.
Individuals who have projection bias make time inconsis-
tent choices, without being aware of it. Under the influence of
projection bias, individuals might end up buying goods in a
hot state that they might regret later. In a hot state they might
end up marrying or proposing a marriage/divorce without
sufficient deliberation. Public policy in some cases provides a
cooling-off period before which applications for marriage and
294 Behavioral Economics
divorce cannot be filed. Projection bias also provides an alter-
native explanation for why individuals get addicted. First, peo-
ple underestimate the (negative) effect of current consumption
on future utility. Second, individuals might under-appreciate
the formation of habit, for example, addiction to a harmful
product. By contrast, in mainstream economics, individuals
know the entire time path of costs and benefits of addiction
and if they choose to be addicts, they do so rationally. This
view is not in accord with the evidence.
Fairness and Reciprocity
Traditional economic theory relies on the assumption of self-
interested behavior (or selfish preferences). This is contra-
dicted by the evidence. A substantial fraction of individuals
exhibit social preferences, that is, care about the consumption
and well being of others. These preferences are more general
than simply altruism (unconditional kindness) and spite
(unconditional unkindness). In practice, however, altruism is
unable to explain a very wide range of behaviors. For instance,
it does not explain why many individuals are conditional
cooperators, that is, cooperate only when others cooperate.
Intrinsic reciprocity, on the other hand, takes the form of
responding to kindness (unkindness) with kindness (unkind-
ness) even in the absence of obvious longer term gains.
Purely selfish preferences are unable to explain a range of
phenomena from many diverse areas such as collective action,
contract theory, the structure of incentives, political economy
and the results of several experimental games. Many of these
phenomena can be accounted for by a model which recognizes
that individuals have social preferences.
1. Wages are usually not lowered in a recession by firms for
fear that this action could be construed by workers as unfair
and can affect their morale and effort levels.
2. Voters voluntarily vote, when pure self interest might dic-
tate them not to take the trouble of voting (almost all single
votes are not pivotal, except in small committees, and vot-
ing is costly).
3. In an experiment, most respondents thought that a hard-
ware store's actions to increase the price of shovels follow-
ing a snowstorm was unfair. However, they did not perceive
an increase in price following an increase in the cost of
inputs as unfair.
4. Why does bargaining stall, with adverse consequences in a
range of situations, such as civil litigation, labor unrest,
domestic strife, and clashes among religious and ethnic
groups? Individuals confuse what serves their self interest
with what is fair. This self-serving bias can lead to a bargain-
ing impasse.
Social preferences have been found in a range of experimen-
tal games that allow for small group or bilateral interaction.
These experiments include the gift exchange game, the ultima-
tum game, the dictator game, the trust game, the public good
game, among others. The most commonly replicated of these
experiments is the ultimatum game.
The ultimatum game is a game played between two players
who could share a cake (or prize) of some fixed size, say 100,
that is finitely divisible into 100 equal parts. The first player,
the proposer, proposes to give a share of the cake s between 1
and 100 to the second player, the responder. If the responder
accepts the offer, the cake is split according to the proposed
split. If the responder rejects the offer, the cake is lost to
both players. Under selfish preferences the subgame perfect
outcome (which is the appropriate notion of the game theo-
retic equilibrium here) is 5=1, 1 — s = 99. The basic idea being
that if the responder does not engage in this game then he
gets nothing, so he would accept the smallest indivisible unit of
the cake. The experimental results reject the prediction 5=1,
1 — s = 99. The gist of the results is as follows. (1) The mean
offer is 5 = 30 — 40. (2) The median offer is s = 40 — 50. (3)
There are rarely any unfair offers, s = 0— 10 or over-fair
offers s >50. (4) Low offers are often rejected. These results
continue to hold for high stake games. The outcome is easily
explained by assuming that individuals have social preferences.
In public good games, the purpose is to explain human
cooperation in undertaking a task for the common good. All
individuals simultaneously make a voluntary contribution to a
common pool. The experimenter then returns a fraction of the
total sum of money to each participant; this captures the
essence of a public good. The prediction of the standard
model with selfish preferences is that everyone should free
ride, that is, depend on others to make the contribution and
make zero contributions themselves. The experimental results
are that individuals start with a reasonably high level of contri-
bution. However, in successive rounds the level of contribu-
tions drops off to a level that is consistent with the model of
selfish preferences. Does the standard model stand vindicated
then? Not necessarily. It does not explain why subjects start off
with initially high levels of contribution. Furthermore, if these
subjects start afresh then they again start with high levels of
contributions. Negative reciprocity is a leading explanation.
Most individuals start off by contributing reasonably high
amounts but some do not. In subsequent rounds, negative
reciprocity kicks in and everyone lowers their contribution
amounts.
An important variant of the public good game is the public
good game with punishment. What if contributors are allowed
to punish the noncontributors at the end of the experiment.
The prediction of standard theory is that individuals should
treat bygones as bygones and not engage in any punishment.
Anticipating this, noncooperators will not be deterred. The
results of experiments do not support this prediction. Allowing
for the possibility of punishment, contribution levels reach
almost the first-best level and noncontributors are heavily
punished. The public good experiments indicate that subjects
are conditional cooperators. They cooperate when they think
that others will cooperate. Furthermore, their preferences allow
for the material payoffs of others (social preferences), thus
giving them an incentive to punish even in games with a finite
horizon. The implications of this result are deeper than just the
current context.
Humans take account of the role of intentions. Humans do
not punish unkind behavior if it was unintentional. However,
intentionally unkind behavior can get severely punished. These
issues are formally modeled in psychological game theory.
Some experimental evidence in certain contexts indicates
that the split between individuals with selfish and social pre-
ferences is 40-60. An important and interesting issue for
Behavioral Economics 295
theoretical and empirical research is to examine the implica-
tions of heterogeneity of preferences in the population. A range
of theoretical and experimental work indicates that even a
minority of individuals with social preferences can signifi-
cantly alter the standard predictions.
Conditional reciprocity, inequity aversion, and the role of
intentions explain several phenomena in contract theory such
as the following. Why don't incentives always work? Why do
we often observe cooperation in finitely repeated relation-
ships? Why do we observe bonus contracts? Why are contracts
often incomplete? Too great a use of the carrot and stick policy
is likely to be perceived as hostile intent by workers, who could
conditionally reciprocate by reducing effort. The standard
model with purely selfish preferences is unable to provide a
convincing explanation.
Time Discounting
Decisions that have a time dimension are central to many
economic problems. The basic idea in most modern theories
of time preference is the tension between immediate gratifica-
tion and delayed, but greater, gratification. For example, if an
individual or nation desires greater consumption in the future,
then current consumption has to be reduced to increase saving
and investment. This trade-off determines the consumption
profile of individuals and nations. Models which are separable
in time and outcome are sometimes called discounted utility
models (DU).
Given two dates, t and t + T, let, c t = ( c t , c t+ !,..., c t+T ), be a
stream of dated consumption levels, or a consumption profile.
A consumption profile could either be a historical record of the
actual consumption levels, or a plan for future consumption
formulated at a time, t 0 < t. The DU model states that the utility
from this consumption profile is,
tf(ct) = X^=o D(t + T M Ct +0 I 10 !
where D is the discount function and u is the single-period
utility function also known as felicity. Note that D is a function
of time only and u depends on time only through the depen-
dence of c on time.
An important question is the following. Suppose that
the consumption profile, c„ is optimal, in the sense that
it maximizes eqn [10] subject to the constraints facing the
consumer. Is the consumption profile c t+s = [c t + s , c t+s+1 , . .
c t+T ) also optimal in the sense that it maximizes U(c t + S ) =
'Er=sD{t+z)u(c t+r )l If this is the case then, c, is said to be
time-consistent otherwise it is said to be time-inconsistent.
Generally, optimal consumption plans are time-
inconsistent. For example, as Richard Thaler argued in an
influential example, a consumer may, now, prefer to have
two apples in 51 days time to one apple in 50 days time but,
come day 50, the consumer may prefer one apple that day to
two apples the following day.
A special case of the DU model is the exponential dis-
counted utility model (EDU) of Paul Samuelson, 1937. In
this case the discount function takes the form
[11]
where 9 is a constant parameter known as the discount rate
and S=e~ e is known as the discount factor. Thus, we also
have D(t) = <5 f , 0 < <5 < 1 . This model has proved extremely
tractable and useful for many purposes. In particular, optimal
consumption plans are always time-consistent if, and only if,
the discount function is given by eqn [11], However, EDU is
rejected by the evidence.
Anomalies of EDU include the following. Under EDU,
the discount rate between a time period of a given length
remains the same irrespective of when the time period begins.
However, the supposition of a constant discount rate is vio-
lated in many experiments; a finding that is known as the
common difference effect. The most famous illustration of
the common difference effect is Thaler's apples example,
mentioned above. This led George Loewenstein and Drazen
Prelec, in 1992, to propose the following hyperbolic discount
function that can address the common difference effect:
D(t) = (1 + a > 0, p > 0. [12]
Other anomalies of EDU include the following. Gains are
more salient than losses (the sign effect), greater magnitudes
are more salient than smaller ones (the magnitude effect),
discounting over an interval depends on how that interval is
subdivided (subadditive discounting), choices may appear to
be cyclical (apparent intransitivity of preferences), choices may
depend on the spread or monotonicity of the consumption
sequence despite different sequences having the same dis-
counted utility, and so on.
Loewenstein and Prelec explained the sign effect by
proposing a per-period utility function or felicity ( u in
eqn [10]) with greater elasticity for losses as compared
to gains. They explained gain-loss asymmetry by proposing
that the elasticity of the felicity for losses exceeds that
for gains. They explained the magnitude effect by assuming
that the elasticity of the felicity is increasing. A functional
form for the felicity that incorporates all these features
was given by Ali al-Nowaihi and Sanjit Dhami. The combi-
nation of such a value function with the discount function
[12] may be called the hyperbolically discounted utility
model (HDU). This can explain the important violations of
the EDU model.
Why do human and nonhuman subjects exhibit hyperbolic
discounting? The work of Partha Dasgupta and Eric Maskin
examines the evolutionary basis for this phenomenon. A wide
range of puzzles that are largely unresolved with EDU can be
explained by HDU. These include the following. Why does
consumption track income so closely? Why do individuals
under-save for retirement? Why is there a sharp drop in con-
sumption at retirement? Why do individuals hold illiquid
assets and credit card debt simultaneously? Why have national
savings rates been declining? Why is marginal propensity to
consume asset specific?
Unlike the ED LI model, where the discount rate is constant
across any two time periods of equal length, in the HDU
model, the discount depends on the distance from the present,
say, time t. In particular, under HDU, the discount rate over the
interval [t, s], s >t is higher than the discount rate over the
equal sized interval [s, s+(s— £)]. A higher discount rate
implies a lower discount factor and, so, a lower weight on the
D(t) = e~ 0t , 8 > 0,
296 Behavioral Economics
future. In this sense, the decision maker is said to be present-
biased relative to the EDU model.
A second class of models that allows for a departure of the
present from the future (relative to the EDU model) is the
model of multiple selves. In this approach, the same individ-
ual, over time, can be viewed as a collection of different selves,
one for each time period. Furthermore, the current self cannot
perfectly control future selves by command. However, the
current self can potentially take actions that can alter the con-
straints faced by future selves.
An important issue in the present-bias models is the degree
of awareness that a current self has about the degree of present
bias of one's own future selves. Experimental evidence is, at the
moment, not conclusive about the degree of self awareness
of the individual. It seems, however, that completely sophis-
ticated behavior, that is, full awareness of one's future self-
control problems and completely naive behavior, that is, no
awareness of future self-control problems are both unreason-
able relative to the intermediate case of partial awareness.
There are several applications of present biased preferences.
Applied to the life cycle model it turns out that the present-bias
of individuals leads to greater current consumption and lower
savings relative to the time consistent behavior found in the
EDU model. The current selves can, however, take certain
actions to bind the future selves.
One possibility is for the current self to invest in illiquid
assets, such as housing, that constrains the ability of the
future selves to engage in excessive consumption. This can
explain, for instance, why individuals simultaneously hold
illiquid assets, such as housing (certainly illiquid as com-
pared to financial assets), and credit card debt, a phenome-
non that EDU cannot explain. The current self buys an
illiquid asset to discipline the self-indulgent behavior of the
future selves. The future self is now faced with mortgage
payments (or a lack of cash due to the purchase of the
house) and, so, this alters his/her budget constraint. How-
ever, the easy availability of credit cards could induce the
future self to borrow and, so, successfully, relax the budget
constraint. In so doing, the future self might believe that
his/her future selves will be more responsible, which might
well turn out to be a naive assumption.
Neuroeconomics
Neuroeconomics studies the neural activity in the human brain
when one makes economic choices. One of the leading neu-
roeconomists, Colin Camerer, defines the subject matter in the
following manner: neuroeconomics is a specialization of
behavioral economics that plans to use neural data to create a
mathematical and neurally disciplined approach to the micro-
foundation of economics.
There is a wide variety of techniques used in neuroeco-
nomics to measure brain activity. Some of these techniques
are so invasive, for example, single neuron recordings, that
they can only be used on animals. However, this is of value,
as humans share a primitive mammalian brain with other
mammals and so the lessons learnt from animal neuroeco-
nomic experiments can provide valuable information. Other
techniques, for example, functional magnetic resonance
imaging, FMRI, are less invasive and are able to measure the
hemodynamic response, that is, the changes in blood flows to
different parts of the brain while pinpointing, with varying
accuracy, to relatively small and specific areas. Establishing
causality is an issue with many of these techniques. Such is
the speed of progress in neuroscience that it is now possible
to be more sure of direct causality using very new techniques
such as transcranial magnetic stimulation, TMS, whose
method can be described as follows. Suppose that we suspect
that some region, R, of the brain is responsible for some
behavior, B, that is relevant to economics. Then TMS can
temporarily curtail/stimulate area R. If this curtails/enhances
behavior B then there is indeed a possibility that region R is
responsible for behavior, B.
It is also possible to fruitfully apply older techniques that
rely on psychophysical measurements, such as pupil dilation,
heart rate, skin conductance etc. For instance, pupil dilation is
enhanced when one sends false or misleading information.
This has obvious implications for economics, for example, in
predicting the truthfulness of messages in an asymmetric
information game. When contributors punish noncontribu-
tors activation in the dorsal striatum (which is the brain's
reward circuit) informs why the outcomes in the public
good game with and without punishment are so different.
Existing neoclassical models are unable to offer a convincing
explanation.
Clearly, neuroeconomic variables have a direct role to play
in economic models. But they can also be useful indirectly.
Economic models are generally estimated by hypothesizing
that some endogenous variable, y, is determined by a set
of exogenous variables x 1: x 2 , . . x N and some noise term, s.
Typically, data is available on only a subset of the variables,
x lr x 2 , . . ., x s , s< N. The omitted variables are then, necessarily,
subsumed within the noise term, s. It is possible, however, that
some neuroeconomic variables, z u z 2 , ■ . ., z m , influence y and
are correlated with the omitted variables and, hence, can be
used as instruments for them.
Usually, there are competing economic explanations for the
same phenomenon. One can then ask which explanation is
more neurally plausible, in the sense of being consistent with
known findings from neuroeconomics. Thus, potentially, neu-
roeconomics could deliver an additional criterion for selecting
among competing explanations.
We regard behavioral economics as an extension of
mainstream (neoclassical) economics, using insights from
psychology and sociology, into areas (and there are many)
where mainstream economics has empirically failed. The
neuroeconomic findings are suggestive of a dual-system func-
tioning of the brain in which system 2 (the neocortex) deter-
mines cognition and system 1 (the limbic system) determines
emotion. Thus, one may hypothesize that decisions made
by system 2 would be in accord with rational choice
theories while the interaction between the two systems
would be in accord with alternative behavioral theories. Of
course, the two systems need not be in conflict. Thus, both
systems would, for example, urge us to breathe, eat and drink,
seek shelter, and avoid hazardous situations. And, indeed,
neuroeconomic findings have supported a range of rational
choice models. Support for the rational choice models, how-
ever, seems to arise largely in environments that are stationary
Behavioral Economics 297
and where evolution has had enough time to make an impact.
Support for behavioral alternatives is more widespread; and
include, for example, models of fairness and reciprocity, pros-
pect theory, models of limited iterative thinking and hyper-
bolic time discounting.
Behavioral Game Theory
Game theory deals with situations in which a set of individuals
(called players) take actions or play strategies that affect not
only their own payoffs (or utilities) but also those of others.
The fundamental concept in game theory is that of a noncoop-
erative equilibrium, proposed by John Nash in 1950, hence,
also known as a Nash equilibrium. In such an equilibrium,
each player plays a strategy that is best for himself in terms of
maximizing payoffs or utility (i.e., a best reply), given the
equilibrium strategies of the other players. Nash proved that
every finite game (finite number of players and finite number
of actions) has a Nash equilibrium.
It is permitted for players to have a probability distribution
over the play of their actions. To emphasize this point, some-
times strategies are referred to as mixed strategies. If each action
is played with strictly positive probability, then the strategy is
strictly mixed. By contrast, in a pure strategy, each action is
played with probability 0 or 1 . A strategy x dominates a strategy
y, for player i, if x gives higher payoffs for i than y, irrespective
of the strategies of the other players. One very simple method
of searching for a Nash equilibrium is that of iterated elimina-
tion of dominated strategies.
A dynamic game is a game played over time in which
players have the opportunity of observing and learning from
the previous actions of other players. A simple way to find a
Nash equilibrium of a dynamic game is by backward induc-
tion: find a Nash equilibrium for the subgame played in
the final period. Then find a Nash equilibrium in the penul-
timate game, assuming that players, come the final period
subgame, will play the Nash equilibrium already found for
that subgame. Carry on till the initial subgame. Backward
induction has the added attraction that it does not allow
threats that are not credible (i.e., those that would never be
used). Such an equilibrium is called a subgame perfect
equilibrium.
Behavioral game theory has developed as a response to the
violations of the predictions of (mainstream) game theory.
Contrary to the prediction of game theory, players do not
engage in iterated deletion of dominated strategies if this
involves more than 2—3 steps of iteration. Within the context
of two steps of iterated eliminations of dominated strategies,
the outcome has also been shown to be sensitive to framing
effects.
A test of the iterated deletion of dominated strategies is
in the context of the Abreu-Matsushima mechanism design
solution. This is a mechanism that is implementable by
using very few steps of iterated elimination of dominated
strategies. The results of Martin Sefton and Abdullah Yavas
show empirically that the predictions of this mechanism fail.
In conjunction with similar evidence from elsewhere, this casts
doubt on the implementability of other mechanisms (most of
which are even more complex).
The backward induction argument usually fails in
experiments.
Multiple Nash equilibria arise in almost all areas of econom-
ics. An important question is whether players can coordinate
(agree) on one of the equilibria (preferably the best for all, if
such an equilibrium exists). Widespread failure to coordinate
on any equilibrium (particularly the one that maximizes joint
payoffs) is observed in experimental results. Preplay communi-
cation does not necessarily enhance coordination. History plays
a powerful role in influencing the degree of coordination. Only
under some conditions, for example, gradual growth of group
sizes, successful coordination in small groups, knowledge of
the history of such coordination, etc., does coordination emerge
successfully. In experiments, if generations give public advice
to their successors then coordination can be enhanced.
While players often seem to play mixed strategies they do
not do so in the proportions predicted by a mixed strategy
Nash equilibrium (MNE). Furthermore, while the aggregate
proportions (across players) are sometimes found to be con-
sistent with MNE, individual proportions of play are typically
inconsistent with a MNE. Tennis and football professionals
seem to play mixed strategies that are consistent with the
evidence. However, the serial dependence of serves, in tennis
for example, is inconsistent with the predictions of a MNE.
There is also a fascinating new literature that tests the pre-
dictions of standard game theory by directly attempting to infer
the cognitive process when subjects play experimental games.
Game theory does not make any assumptions on the cognitive
process when subjects play games. Some traditionalists argue
that a study of the cognitive process cannot confirm or reject
anything in game theory. Consider, for instance, the work of
Eric Johnson et al. They consider Rubinstein's alternative offers
bargaining game played by two players over three rounds. The
game is easily solved by backward induction. One would,
therefore, expect that subjects will look to see the last round
data such as payoffs, actions etc., then look at the penultimate
round data and finally, the first round data. The authors use
MOUSELAB software to observe if the pattern of searches made
by experimental subjects is backward, beginning with the third
period data. It turns out that subjects search forward rather
than backward, that is, looking first at the first round data, then
the second round data, and finally, the third round data.
The Nash equilibrium concept and its refinements have
helped to shed light on many interesting questions in econom-
ics. Furthermore, the machinery of game theory has forced
economists to be completely explicit about who the players
are, what moves are open to them, the sequence of their moves,
what objectives they have, what information they acquire and
when, etc. This level of rigor gives modern economics a tre-
mendous advantage over pre-game-theoretic economics. How-
ever, the empirical evidence on the Nash equilibrium and its
refinements is not very encouraging, especially when indivi-
duals interact in bilateral or small group settings. More positive
evidence is found at the level of market outcomes and in large
groups.
There are now several alternative models of behavioral
game theory that perform better than a MNE. These include
the concept of a quantal response equilibrium (QRE). QRE has
been able to explain a range of findings from experimental/
field data that the classical equilibrium concepts have found
298 Behavioral Economics
troubling to explain. In a QRE, players play strictly mixed
strategies. In contrast to notions in classical game theory,
such as trembling hand perfection, the trembles do not, in
the limit, tend to zero in a QRE. Trembles are fundamental
and remain significantly large in a QRE. This avoids a range of
problems that arise in classical theory regarding assigning
beliefs about actions that are not played in equilibrium. The
reason is that, in a QRE, all paths of play are ex-ante possible.
The main requirement of a QRE is that players form beliefs that
are correct in equilibrium. However, because players random-
ize over all their actions in a QRE, they do not necessarily play
best replies.
Forming beliefs that are correct in equilibrium, as in QRE, is
cognitively challenging. This requirement is relaxed in a second
class of behavioral game theory models, the level-fe models. In
these disequilibrium models, players' beliefs need not be cor-
rect. However, players play best responses, conditional on
these, possibly incorrect, beliefs.
Game theorists steeped in classical theory might well react
indignantly and ask how players could possibly hold on to
beliefs that turn out to be incorrect? There are two responses.
(1) Level-fe models are motivated by actual experimental evi-
dence which indicates disequilibrium beliefs. (2) Even when a
process of learning might correct the beliefs of players over
time, there are still good reasons to pay attention to level-fe
models. First, many important decisions that economic agents
make in real life are of the nature of one-off decisions or
decisions that are not repeated often enough. Second, in
many experiments where long-term learning has a positive
impact, the economic environment is held fixed. In real life,
however, the economic environment is often too variable to
permit the clean learning results of experiments.
Level-fe models introduce heterogeneity among players by
distinguishing between types of players, LO, LI, L2, . . . The
generic type is denoted by Lk. Type LO, the only nonstrategic
type, typically chooses to simply randomize among the various
available actions. In general, any type Lk, fe > 1 assumes that all
other players are of type Lk — 1 only. In it's cognitive hierarchy
version, type Lk, k > 1 assumes that there is a distribution of all
the lower types of players. Fixing beliefs in this manner, each
type of player then chooses a best response over its available
actions. Type LO, sometimes known as the anchoring type, need
not be physically present in the population. It can exist purely
as a model of other players in the minds of type LI players.
Consider the p-beauty contest. In the simplest such game,
subjects are asked to guess numbers between 0 and 100. The
subject whose guess is closest to 2/3 times the average wins a
prize. The Nash equilibrium prediction is for each player to
choose 0 (the reader should check this).
However, the level-fe model predicts that people will play
the p-beauty contest as follows. A type LO player simply rando-
mizes and so chooses the number 50. A type LI player thinks
that all others are type LO so chooses the number | (50) ~ 33.
A type L2 player thinks that all others are type LI and so
chooses the number 50 (I) 2 ~ 22, and so on. Thus, if players
were indeed of types L0, L 1 , L2, ... then we would expect to see
spikes in the data at the numbers 50(§)°, 50(f) 1 , 50(f) 2 , . . . This
is a remarkably precise and unambiguous prediction. More-
over, this prediction has been verified in several empirical
works even when played with very high ability subjects.
Level-fe models can explain a range of phenomena that are
difficult to explain with other models. These include results from
auctions and market entry games, among many others. The
evidence from MOUSELAB experiments indicates that types LI
and L2, in the level-fe models, make up 67-89% of the popula-
tion. This data is hard to reconcile with QRE models suggesting
that the level-fe models have better psychological foundations.
There is a range of other behavioral game theory models.
These include the noisy introspection model, models that con-
sider the role of memory explicitly, models of coarse thinking and
mental categorization, analogy based equilibrium models, cursed
equilibrium and so on. Some are more speculative than others.
We have not touched on behavioral models of learning
because of space constraints and the fact that the technical
requirements for understanding even the basics are beyond
the scope of the current contribution.
Liberalism and Paternalism
A fundamental principle of classical liberalism is that people
are the best judges of their own welfare. This lies, explicitly
or implicitly, at the heart of traditional public economics.
Thus, the role of government is limited to upholding the law,
enforcing contracts and correcting for market failures, when
practicable. Of course, it has always been recognized that some
well-defined groups, children for example, are not fully able to
judge their own interests. The opposite view that, in the main,
people are not the best judges of their own welfare, is known
as paternalism. Under paternalism, an institution is needed
to judge people's welfare for them, for example a government,
political party or religious establishment.
Paternalism has not often had a beneficial impact in human
history. However, there is strong empirical evidence from
behavioral economics that people are, by and large, not ratio-
nal optimizers, unlike the assumption under liberalism. A
small selected sample of the violation of the rationality
assumption is as follows. Many individuals do not utilize the
full limits of their 401k pension plans despite the fact that the
company will match the extra contributions. There is also
inadequate diversification of portfolios. People often use a
range of judgement heuristics to make decisions. The evidence
is largely (but not always) inconsistent with the use of Bayes'
rule, rational expectations, and even strict optimization. Peo-
ple become obese without any apparent liking for the condi-
tion of obesity or addicts who then often need to be helped by
others to get rid of the habit. There is also an overutilization of
payday loans. These are loans that are repaid when the next
paycheck arrives; the short-term interest rates on these loans
are much higher than loans available elsewhere.
In some cases, the textbook rational behavior might not be
observed on account of lack of information, or limited cogni-
tive abilities. In other cases, it could be due to a current-bias in
preferences, which lead to self-control problems and dynamic
inconsistency. Examples are various forms of addictive behav-
ior, the desire to quit, and the subsequent relapses that are
often difficult to reconcile by using the traditional rationality
framework. People exhibit preference reversals. Preferences are
often not well-defined, are often incomplete, and dependent
on the framing of information.
Behavioral Economics 299
Shlomo Benartzi and Richard Thaler show that only 20% of
the people in their sample prefer their own portfolio to the
median portfolio. In other important contexts such as pensions,
individuals sometimes use the 1/n heuristic to allocate their
portfolio where n is the number of options available, hence,
making outcomes very sensitive to what assets are available.
This evidence has disturbing implications for the debate
between liberalism and paternalism. Aware of this, behavioral
economists sometimes refer to paternalism in the traditional
sense as heavy-handed paternalism. By contrast, the form of
paternalism suggested by behavioral economists is sometimes
called soft paternalism or light paternalism. It is this form of
paternalism that we now briefly review.
Some view the relaxation of rationality in welfare econom-
ics as a natural progression of the relaxation of other restrictive
assumptions in economics, such as perfect competition, perfect
information, certainty etc.
At the heart of most proposals under the ambit of light
paternalism is the need to distort individual decisions/choices
as little as possible and use as little coercion as possible. The
idea is that the mistakes (or misperceptions) of the bounded
rational types will be nudged in the direction of a correction
while the minimal intrusion associated with light paternalism
should have little effect on the actions of the fully rational
types. To operationalize light paternalism, one needs to specify
what a mistake/misperception is. Furthermore, in the light of
these mistakes/misperceptions, how should we judge the out-
comes of light-paternalistic policies? The latter question
requires us to have some behavioral welfare criteria. The reason
that a traditional welfare criteria would not be suitable is that
actual choices might not be indicative of an individual's best
interests. Thus, unlike the classical analysis we cannot use
preference based measures of welfare such as equivalent varia-
tion and/or compensating variation to seek the direction in
which individual welfare is improving. As Loewenstein and
Haisley put it: " Clearly, it doesn't make sense to assess whether
someone is committing an error using a measure that is pre-
mised on the assumption that people don't commit errors."
Botond Koszegi and Matthew Rabin argue that unless one
imposes ancillary assumptions, it is impossible to infer prefer-
ences from choice even with unlimited data and under the
assumption that people are 100% rational. Consider, for
instance, the possibility that people have social preferences.
Suppose that we observe a set of choices in which an individual
shares with others when there is an option to do so. From this,
should we conclude that the individual is happy to share with
others? Or that she would feel guilty if she did not? Or that she
would prefer not to have an option to share? These distinctions
cannot be made based on the choice data.
Although additional evidence is already accumulating at
a rapid rate, behavioral economics is a relatively new field
and many important effects and channels are not fully under-
stood yet. For that reason one needs to be cautious and
careful in testing the implications of various paternalistic
policies. Some have advocated careful small-scale pilot stud-
ies as well as field studies before the actual implementation
of paternalistic policies. The evidence suggests that not all
people are subject to the relevant behavioral biases. Clearly
regulation that is too stringent might have a negative effect on
these rational types.
Many notions of behavioral paternalism have been advo-
cated in the literature that the reader might wish to follow up
on. These include libertarian paternalism attributable to
Richard Thaler and Cass Sunstein, asymmetric paternalism
attributable to Colin Camerer et al. and a pragmatic approach
attributable to George Loewenstein and Emily Haisley.
Brief Conclusions
Behavioral economics is one of the most rapidly growing fields
in economics and is gradually becoming mainstream. It relies
on a strong respect for the evidence, falsifiability, and a recogni-
tion that experiments in the lab, and neuroeconomic evidence,
can usefully complement field data. It draws its principles from
several decades of research in psychology (and sociology).
Behavioral economics has seriously questioned the rationality
assumption in economics as well as the assumption of selfish
preferences. While traditional economics did recognize the pos-
sibility of less than perfect rationality, it assumed that these
errors would cancel out in the aggregate. Behavioral economics
has, instead, shown that these errors might cumulate and snow-
ball in one direction and lead to outcomes that are very different
from those predicted by traditional economics. Behavioral eco-
nomics has also stressed the importance of emotions in making
decisions, role of framing in influencing outcomes, provided
richer theories of decision-making under risk and over time
and provided new solution concepts in strategic interactions.
In this sense it has provided new theory to explain the emerging
evidence that is both refutable, and based on empirically robust
psychological principles.
Acknowledgment
We are very grateful to Emi Mise for her heroic efforts in convert-
ing our Scientific Workplace file to a portable Latex format.
We are also grateful to Samuel Smithers for a correction.
See also: Altruism and Helping Behavior; Anxiety and Fear; Cognitive
Dissonance Theory; Consumer Psychology; Decision Making
(Individuals); Preference Judgments (Individuals); Self-Fulfilling
Prophecy.
Further Reading
Bowles S (2006) Microeconomics: Behavior, Institutions, and Evolution. Princeton:
Princeton University Press.
Boyd R, Bowles S, Fehr E, and Gintis H (2005) Moral Sentiments and Material Interests.
Cambridge, MA: MIT Press.
Camerer C (2003) Behavioral Game Theory: Experiments on Strategic interaction.
Princeton, NJ: Princeton University Press.
Camerer C, Loewenstein G, and Rabin M (eds.) (2003) Advances in Behavioral
Economics. New York/Princeton, NJ: Russell Sage Foundation and Princeton
University Press.
Fehr E and Gachter S (2000) Fairness and Retaliation - The Economics of Reciprocity.
Journal of Economic Perspectives 14: 159-181.
Frederick S, Loewenstein G, and 0’Donoghue T (2002) Time discounting and time
preference: A critical review. Journal of Economic Literature 40: 351-401 .
300 Behavioral Economics
Glimcher P, Camerer C, Fehr E, and Poldrack R (2008) Handbook of Neuroeconomics.
North Holland: Elsevier.
Kahneman D and Tversky A (1979) Prospect theory: An analysis of decision under
risk. Econometrics 47: 263-291.
Kahneman D and Tversky A (eds.) (2000) Choices, Values, and Frames. New York:
Cambridge University Press.
Loewentein G (2007) Exotic Preferences: Behavioural Economics and Human
Motivation. Oxford, England: Oxford University Press.
Loewenstein G, Read D, and Baumeister R (eds.) (2003) Time and Decision: Economic
and Psychological Perspectives on Intertemporal Choice. New York: Russell Sage
Foundation Press.
Thaler RH (1993) The WinneTs Curse: Paradoxes and Anomalies of Economic Life.
Princeton, NJ: Princeton University Press.
Thaler RH (2005) Advances in Behavioral Finance ti. New York: Russell Sage
Foundation.
Thaler RH (2008) Sunstein, Cass B, Nudge: Improving Decisions on Health, Wealth, and
Happiness. New Haven, CT: Yale University Press.
Vohs KD, Baumeister RF, and Loewenstein G (2007) Do Emotions Help or Hud
Decision-Making? A Hedgefoxian Perspective. New York: Russell Sage
Foundation Press.
Behavioral Genetics
C T Nagoshi, Arizona State University, Tempe, AZ, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Allele An alternate form of a gene at a particular place
(locus) on a chromosome.
Association Correlation between phenotypic variation and
a marker (genetic, allelic, or DNA) of variation on a
chromosome.
Dominance The ability of one allele (the dominant) to
override the phenotypic expression of another allele
(the recessive).
Epistasis Nonadditive effects of the combined actions of
two or more genes, that is, gene-gene interaction.
Gene A segment of a chromosome (deoxyribonucleic acid
or DNA) that codes for some aspect of protein synthesis.
Genotype The genetic composition of an individual.
Heritability The proportion of phenotypic variability that
can be accounted for by genetic differences.
Inbreeding Matings between closely genetically related
individuals.
Linkage The extent to which two genes near each other
on the same chromosome are transmitted together instead
of segregating and independently assorting.
Phenotype The variant of a measurable characteristic
or trait.
Pleiotropy Multiple phenotypic effects of the same gene.
Behavioral genetics, as the name indicates, is the interdisciplin-
ary research area concerned with determining if and how
genetic factors influence any of the phenomena studied in the
behavioral sciences, especially psychology. Behavioral geneti-
cists apply the concepts and methodologies of molecular, pop-
ulation, and quantitative genetics to behavior as the trait of
interest.
Introduction
Two readily apparent, related aspects of human behavior make
a compelling case for the need to study genetic influences on
behavior. These are stability and variability.
If a behavior, for example, scoring high on tests of intelli-
gence, is consistently manifested by an individual across time
and across many different situations, then the stability of
this behavior must be due to some consistent set of physical
determinants. Through consistent reinforcement and/or pun-
ishment of certain behaviors, plus generalization, a stable
behavior could have been established by environmental influ-
ences on the individual. On the other hand, the behavior could
be a manifestation of some consistent 'hard-wired' physiologi-
cal process, for example, greater efficiency of the connections in
the brain's neural network. Since all physiological/biological
processes ultimately begin from some genetic code, it could be
inferred that the stable behavior is the result of the individual
having inherited some gene or genes for the behavior.
In turn, both the lack of variability and the presence
of variability in stable behaviors also require the influence of
some set of consistent physical determinants. For example, all
physiologically normal humans raised among other humans
acquire the ability to use language. The lack of variability in
this behavior among humans, plus the absence of this behavior
in this form in other animal species, is consistent with the
supposition that all normal humans inherit a set of genes that
code for the physiological structures that are required for
language. Other stable behaviors, such as performance on intel-
ligence tests, are characterized by great variability among
humans, and this variability is maintained in spite of historical
and social changes in the environmental factors thought to
influence the behaviors. Since each individual is genetically
unique, stable individual differences in behavior are also consis-
tent with the supposition that genetic determinants may be
involved.
Although behavioral genetics is concerned with establish-
ing the genetic correlates of behavior, the discussion above
makes clear that genes, in fact, do not directly cause behavior.
Figure 1 reiterates the idea that genes are basically a biochemi-
cal process for coding the synthesis of proteins which, in turn,
are assembled into the physiological structures and processes
of the organism. These physiological structures then become
one of the determinants of behavior, and the study of this
process is the realm of physiological psychology. Thus, any
finding from behavioral genetics of significant genetic contri-
butions to behavior automatically implies that some physio-
logical mechanism is the more proximal determinant.
Behavior can also be influenced by the environment
through learning, and one dotted line shows the presumption
in physiological psychology that ultimately learning operates
by modifying the physiology of the organism. It should also be
noted that genetic influences always operate on physiology
within the context of some set of environmental influences.
The other dotted line presents the possibility (discussed below)
that genetic processes can affect environmental influences.
In Figure 1, the understanding of how genetic processes are
translated into physiology is noted to be the realm of molecu-
lar genetics/genomics. Molecular genetics is concerned with
the sequence and operation of biochemically defined genetic
material (the DNA base pairs described below) in transcribing,
transmitting, and implementing the code for the synthesis
of physiological structures and processes. Genomics focuses
on gene expression, much of which has evolved to be respon-
sive to intracellular and extracellular environmental effects.
301
302 Behavioral Genetics
Genes _
(Molecular genetics
/genomics)
(Behavioral ^
genetics) Physiology -* Environment
Figure 1 Relationship of genes to behavior.
Quantitative genetics estimates the relative contribution of
genetic versus environmental influences to variation in a trait
by comparing differential resemblances or patterns of differen-
tial variability among individuals with different degrees of
genetic relatedness. Population genetics bridges molecular
and quantitative genetic approaches, as well as evolutionary
processes, by seeking to relate gene frequencies and degree
of genetic influence with particular traits for populations of
organisms operating within particular environmental contexts.
Since behavioral genetics is basically the application of
genetics to a special set of traits, behavior, much of this article
will be an overview of genetic concepts and methodologies.
We begin with a discussion of Darwin's theory of evolution by
natural selection and of Mendel's work on the logic of genetic
transmission.
Evolution and Mendelian Genetics
The evolutionary theory of Charles Darwin (1809-1882) had a
profound impact on Western thought. Even though it provides
a basis for a unified understanding of all biological phenom-
ena, the theory was and remains controversial. In psychology,
it was integrated into the concepts of the nineteenth century
physiological psychologists, became a major inspiration for
William James' school of functionalism (the school which,
somewhat ironically in this context, led to John B. Watson's
behaviorism) and Sigmund Freud's school of psychoanalysis,
and the work of Darwin's cousin Francis Galton, was also the
starting point for the scientific study of individual differences
and genetic influences on human behavior.
Darwin's theory, first published as On the Origin of Species by
Means of Natural Selection in 1858, has three major compo-
nents, each of which has important implications for the under-
standing of human behavior:
1. Induction of variation proposes that members of a species
differ from each other at birth and through maturation
along a multitude of different trait dimensions and that
biological processes continually create these individual
differences. Not only are such differences ubiquitous for
any species, they are absolutely essential for the survival
of the species in the face of relentless changes in environ-
mental conditions. One obvious implication of this for
psychology is that, in contrast to the ideas of several early
behaviorists, individual members of a species, including
humans, are biologically unique and should not be ex-
pected to respond to the same environmental influences
on behavior in exactly the same way.
2. Maintenance of variation proposes that individual differ-
ences in the traits that vary within a species are biologically
transmissible from one generation to the next. Again, this is
absolutely essential for the survival of the species. This idea
leads to the immediate inference that behavior must also
be biologically transmissible, since stable patterns of behav-
ior are undoubtedly essential aspects of species survival.
Thousands of years of humans breeding animals for behav-
ioral traits supported the idea of genetic bases for behavior,
even for humans.
3. Natural selection assumes that there are always more mem-
bers of a species born at any one time than will ever survive
to reproduce. Darwin proposed that environmental condi-
tions at a particular time, for example, availability of food,
nature of predators and parasites, climate, etc., tend to give
some individuals born with one set of traits, an advantage
in being able to survive, reproduce, and rear offsprings to
maturity over other individuals with different levels of these
traits. Over several generations of this natural selection,
more individuals with the advantageous characteristics are
born with and transmit these characteristics, while there is
a decreased incidence and transmission of the disadvanta-
geous characteristics. Darwin argued that over a sufficient
number of generations, particularly in response to some
major change in the environment, the members of a species
may now possess such different traits from the starting
population that they constitute a new species.
As noted above, Darwin's theory clearly reinforces the
importance of genetic transmission and biologically based
individual differences as determinants of the characteristics of
organisms, including behavioral characteristics. The theory
also implies that all species, including humans, are biologically
related on some level to every other species. The similarities
and differences between and within species could be under-
stood solely in terms of the interactions between the genetic
endowments of a species and the selection pressures in the
environment. Research into the determinants of behavior in
other animal species in order to understand human behavior is
justified, because other animals and humans have some of the
same or similar physiological structures and processes that
determine behavior. Other implications include the expecta-
tion that humans retain physiological structures that were once
adaptive or neutral in reproductive fitness value in natural
environments, but which may be maladaptive in rapidly
changing modern societies (e.g., the 'fight or flight' response).
Another implication is that even those 'higher' aspects of
human behavior and mental processes that supposedly differ-
entiate us from other animals can be related to genetically
transmitted physiological functions that resulted from natural
selection.
Behavioral Genetics 303
Although natural selection was the most parsimonious and
successful proposed mechanism for accounting for species
change, at the time Darwin published his theory the way that
traits were biologically transmitted was poorly understood.
Darwin himself realized that the then current concept that
parents gave off biological particles ('gemmules') that merged
and blended in the offspring would, in fact, invalidate his
principles of induction and maintenance of variation. Due
to regression toward the mean, generations of blending traits
would result in all individuals having the same averaged trait,
that is, no variation. It was only after the work of Mendel and
the beginning of modern genetics that a mechanism for the
maintenance of variation was established.
Gregor Mendel (1822-1884) studied the types and fre-
quencies of plant characteristics that resulted when members
of a species possessing one kind of characteristic were crossed
with members having a different characteristic. For his most
famous work on pea plants, such characteristics included green
versus yellow pea color, smooth versus wrinkled pea form,
long versus short plant stem, etc. His cross-breeding studies
led him to infer two principles for genetic transmission.
The law of segregation stated that every character (the phe-
notype) is determined by the combination of two elements
(genes), one from each parent. These elements split off cleanly
(segregate) from each other in the production of the sex cells
and are transmitted intact to the offspring. Since it is possible
for one form of the gene (a dominant allele) to prevent the
expression of the other form (the recessive allele), a process
called dominance, the same phenotype could result from dif-
ferent combinations of genetic elements (the genotype). For
example, a green pea would result whether the genotype con-
sisted of both alleles coding for green color (homozygous
dominant) or a genotype with both green and yellow coding
alleles (heterozygous), since the green allele was dominant.
This was apparent when Mendel crossed 'pure' green and
'pure' yellow pea plants and found that the next generation
all had green peas. When this second generation was crossed
with each other, however, Mendel found that one-fourth of
the succeeding generation again manifested yellow peas (both
alleles coding for yellow; homozygous recessive). Thus, the
yellow alleles were not blended, the frequency of the recessive
phenotype could be predicted from the random combination
of the dominant and recessive alleles, and the variability of
the genes determining green versus yellow pea color was
maintained.
The law of independent assortment stated that different genes
determined different characteristics. These genes were ran-
domly combined in the sex cells and, hence, independently
transmitted to the offspring. Mendel found that when he
looked at the results of crossing plants for two different traits,
for example, green pea color/long stem with yellow pea color/
short stem plants, the odds of a plant possessing green versus
yellow peas were completely independent of the odds of the
plant having a long versus a short stem. This was also impor-
tant for Darwin's theory, since it meant that one trait could
be selected for without affecting some other trait.
As it turned out, modern genetic findings have turned up
several exceptions to Mendel's laws. Many genetic/physiologi-
cal and/or environmental factors can result in a genotype not
being completely manifested in the phenotype ( incomplete
expression/penetrance). Many traits determined by single genes
(one pair of alleles) do not show dominance, such that het-
erozygous genotypes may result in phenotypes intermediate
between phenotypes resulting from the two homozygous gen-
otypes. Heterozygous genotypes may produce more extreme
phenotypes ( overdominance ). Genes on the X-chromosome
(discussed below) are differentially transmitted to male versus
female offspring. Recent findings suggest that parents in a few
genetic systems may be able to determine the expression of
both alleles of an offspring ( genomic imprinting and uniparental
disomy). An important recent discovery concerns simple se-
quence repeats or gene amplification, where three-nucleotide
base-pair sequences within genes increase in number over
generations, leading to disruptions in normal gene function-
ing. With regard to independent assortment, it is now known
that genes that are close to each other on the same chromo-
some are often transmitted together ( linkage ) and that single
genes can have multiple phenotypic effects (pleiotropy) . Never-
theless, Mendel's laws more than adequately accounted for
observations of genetic transmission at the phenotypic level
long before the biochemical nature of genes was understood.
Although Mendel presented his findings at a scientific
meeting in 1865, they were 'lost' until 1900, when several
biologists (Erich Tschermak von Seysenegg, Hugo, de Vries,
and Carl Correns) rediscovered and started widely applying
the laws. In the meantime, Darwin's cousin Francis Galton
(1822-1911) was using the theory of evolution to establish
the science of behavioral genetics.
Galton immediately inferred from Darwin's theory that all
human behaviors had a biological/genetic basis and were the
result of natural selection. With his book Hereditary Genius:
An Inquiry Into It's Laws and Consequences (1869), he asserted
that social eminence (reflective of intelligence) was genetically
determined. He demonstrated this by showing how the inci-
dence of eminence in the male relatives (fathers, sons, uncles,
cousins, etc.) of eminent men was a strict function of the degree
of biological relatedness between the probands (the index cases
from whom other family members are identified) and the rela-
tives. He developed the idea of the twin research design, when he
studied more physically similar twins versus less similar twins
and asserted that the former's greater similarity on psychological
traits also indicated a genetic basis for psychological as well as
physical characteristics. He established laboratories for the sys-
tematic measurement of thousands of individuals on physical
(height, weight, etc.) and psychological (acuity of eyesight, acu-
ity of hearing, reaction time, etc.) traits in hopes of establishing
the large data bases he knew were required to delineate geneti-
cally based distributional characteristics, social class differences,
developmental trends, and familial resemblances.
Besides his enormous contributions to psychology in estab-
lishing ways of measuring psychological phenomena of all
sorts, Galton's quest for ways to demonstrate genetic transmis-
sion in relatives led to the development of new statistical
techniques. His student, Karl Pearson, developed the correla-
tion coefficient to quantify familial resemblances in some
standardized fashion, and genetic questions spurred the devel-
opment of statistical procedures commonly used in the social
sciences (e.g., Fisher's chi-square and F-tests).
Thus, much of the logic and methodologies of modern
behavioral genetics had been laid down by Galton and his
304 Behavioral Genetics
students by the time Mendel's ideas were rediscovered. Ironi-
cally, it took some time and controversy to reconcile Galton's
ideas, that accumulations of genetic elements of some sort
produce continuous, typically normal distributions of traits,
such as intelligence, with Mendel's ideas about discrete traits
under the control of only a few combinations of alleles. The
reconciliation involved the recognition that the random com-
bination (consistent with independent assortment) of several
discrete genetic elements, all of which have some influence
on the trait, would produce the continuous, normal distribu-
tions Galton saw. It is still important, though, to distinguish
between traits thought to be under the control of a single gene
versus those under the control of several genes.
Although Darwin's ideas were important for the school of
functionalism established in the United States by William James
and others at the end of the nineteenth century, and Galton's
ideas were the starting point for trait theories of intelligence
and personality, genetic explanations for behavior were largely
eclipsed by behaviorism throughout most of the twentieth cen-
tury. There was also a great amount of philosophical resistance in
the general public to the idea of biological/genetic determinants
of human behavior. For instance, there was considerable public
controversy following the publication of Arthur Jensen's 1969
paper asserting that, due to the high heritability of intelligence,
educational programs would have only minimal effects in ame-
liorating race differences in I.Q. Ironically, this same controversy
was resuscitated by the 1994 publication of Richard Herrnstein
and Charles Murray's The Bell Curve.
Several historical developments have led to the much
greater acceptance today of behavioral genetics in both main-
stream psychology and the general public. The discovery of the
structure of deoxyribonucleic acid (DNA), the fundamental
molecule of heredity, by Watson and Crick in 1953 spurred
the vast expansion of knowledge in recent years on the bio-
chemical characteristics and processes of what had up to then
been only a hypothetical genetic mechanism. Discoveries
about the biochemical workings of genes and gene products
continue and will continue to be among the most exciting
findings in science. The findings from physiological psychol-
ogy and neuroscience have greatly expanded our understand-
ing of the biological processes that determine behavior, while
the idea of learning as the sole basis for behavior no longer has
ascendancy. Meanwhile, the accumulation of findings from
pedigree, twin, and adoption studies has provided compelling
evidence for the significant heritability of intelligence, person-
ality, schizophrenia, alcoholism, etc.
Population Genetics
Prior to the development of the techniques discussed below
under Molecular Genetics for identifying markers of DNA vari-
ation, genetics research involved inferring underlying genetic
mechanisms from observed phenotypic variations. Both popu-
lation and quantitative genetics primarily analyze such infor-
mative phenotypic variation, and both fields are moving
toward the greater incorporation of DNA markers into research
designs for more precise analyses of genetic processes.
As noted above, population genetics seeks to under-
stand the frequencies and relative contributions of genes on
particular variations in traits for a group of organisms oper-
ating within a particular environmental context. An important
starting point for population genetics was the statistical der-
ivation of the Hardy-Weinberg-Castle equilibrium, which
showed that, in the absence of influences that produce disequi-
librium (discussed below), the relative frequencies of alleles
(dominant vs. recessive, etc.) and genotypes (homozygous
dominant, heterozygous, etc.) for a phenotype remain at a
constant ratio over generations. This was basically just a
restatement of Darwin's idea of maintenance of genetic varia-
tion, but in this case it was derived through mathematically
following the logic of genetic transmission defined by Mendel.
There are four major forces that can change allelic frequen-
cies, that is, the relative frequencies of one form of a gene
versus another, in a population:
Migration can change allelic frequencies if individuals who
move into or out of a population differ from the original
population in their allelic frequencies. For example, it can be
argued that the waves of immigrants who came to the United
States throughout its history were more likely to carry genes
for such phenotypes as adventurousness, hardiness, rebellious-
ness, etc., compared to the populations left behind.
Random genetic drift can affect small populations, when
chance causes some alleles and not others to accumulate over
several generations.
Mutation is the ultimate source of inducing genetic varia-
tions, as alleles are altered by errors in genetic replication and
transmission and by environmental influences. The spontane-
ous mutation rate, however, is thought to be quite low, and the
effect of mutation on allelic frequencies in most cases is minor
compared to natural selection.
Natural selection is the prime basis for genetic change
described by Darwin. Natural selection operates on the basis
of the phenotypes that give an organism some advantage or
disadvantage (relative fitness) in surviving to reproduce off-
spring who, in turn, will pass on the genes that determine the
phenotype. If a phenotype, such as schizophrenia, is selected
against by the environment, that is, the phenotype reduces the
reproductive success of the individual, and is determined by
a dominant allele, then both homozygous dominant and het-
erozygous genotypes will be selected against. In this case, nat-
ural selection will rapidly decrease the frequency of the
dominant allele. If, however, the maladaptive phenotype is
coded for by a recessive allele, as is the case for phenyl-
ketonuria (discussed below), then only homozygous recessive
genotypes will be selected against, and large numbers of het-
erozygotes (carriers) will continue to transmit the 'bad' allele to
some of their offspring. This process tends to ensure that
adaptive phenotypes are coded for by dominant alleles, while
maladaptive phenotypes are carried and maintained on reces-
sives in populations. This picture becomes more complicated
when there is not complete dominance for a gene. In the
extreme case, selection may eliminate all but one form of a
gene for an adaptive trait in a population or species, and this
trait then becomes a species characteristic (i.e., there is no
variability for the trait).
Natural selection may also act to maintain or even increase
the variability of a trait in the population, creating a so-called
balanced polymorphism. For example, sickle-cell anemia is a
typically fatal medical condition in individuals of African
Behavioral Genetics 305
ancestry. The disease is caused by an autosomal recessive gene,
thus homozygous recessive genotypes are selected against. Het-
erozygotes, however, manifest a greater resistance to malaria in
malarial environments than homozygous dominants, and both
alleles are maintained in the population, in spite of the delete-
rious effects of the recessive gene by itself. Predator-prey and
parasite-host relationships also often manifest balanced poly-
morphisms, as natural selection for adaptations in one organ-
ism are constantly being compensated for by corresponding
natural selection for adaptations in the other organism.
Since characteristics that increase the probability of being
chosen as a mate also directly affect reproductive success, sex-
ual selection is a special form of natural selection and can
produce many of the same effects described above. On the
other hand, the phenotypes chosen by mates may not be
consistent with those resulting from natural selection, that is,
as a result of sexual selection, species may come to manifest
traits that appear to be maladaptive.
Besides forces that change allelic frequencies, population
geneticists also consider two forces, inbreeding, and assortative
mating, that change genotypic (but not allelic) frequencies by
increasing the similarity of the genes that both parents transmit
to their offspring. This is important, because such parental
similarities in the genes they transmit would increase the rela-
tive frequencies of homozygous dominant and recessive geno-
types, while decreasing the proportion of heterozygotes. This,
in turn, typically has the effect of increasing the variability
of the trait in the offspring generation, since there would be
higher frequencies of the extreme genotypes. This increased
variability might counter the reduction in trait variability
resulting from unidirectional natural selection.
Inbreeding is the mating of closely genetically related indi-
viduals and, as described above, causes parents to be more
similar in the genes they transmit to offspring. Since recessive
alleles often code for deleterious phenotypes but are not
expressed, due to their being paired with a dominant, the
increased incidence of homozygous recessive genotypes result-
ing from inbreeding often results in decreased fitness for the
offspring of such matings ( inbreeding depression). The counter-
part to this is the increased fitness ( hybrid vigor) found when
more genetically unrelated individuals are mated. Some evi-
dence for both inbreeding depression and hybrid vigor effects
on intelligence have been found in studies of offspring from
first cousin versus those from cross-ethnic group marriages.
Assortative mating (which is ubiquitous for human behav-
ioral traits) is where mates are similar in their phenotypes,
whether due to mate choice or environmental factors. There
are three different ways in which assortative mating may occur:
In genetic similarity, mates select each other or are selected
for each other on the basis on having similar genes, and the
effects of this on offspring genotypic frequencies are similar to
inbreeding. This phenomenon has been demonstrated in some
animal species, but is controversial in humans.
In phenotypic assortment, mates select each other on the basis
of similarities in observable characteristics. Phenotypic assort-
ment can only produce greater genetic similarity between the
parents if there is a significant relationship between the geno-
type and the phenotype, that is, if the characteristic is heritable.
In fact, many human behaviors, such as intelligence, personal-
ity, attitudes, some psychopathology, etc., have been shown
to produce moderate to high spouse correlations indicative of
phenotypic assortment.
In social homogamy, mates are similar in their phenotypes
simply because the environments in which they meet and
mate only have individuals possessing certain characteristics.
In a highly stratified society, social class differences can
cause human mates to be highly similar in several behaviors
relative to mates from other classes, even when there is
no active phenotypic assortment. Social homogamy can only
produce greater parental genetic similarity if the phenotypes
of interest are heritable and there's a significant gene-
environment correlation (discussed below).
Phenotypes may also be found to be similar in mates due to
mates influencing each other after getting together, a phenom-
enon sometimes called contagion.
One important offshoot of population genetics is sociobiol-
ogy, which seeks to determine the evolutionary significance
(i.e., fitness value) of behavioral systems. For example, socio-
biologists are interested in the environmental conditions,
physiological characteristics of the organism, and available
genetically transmissible behaviors that cause one animal spe-
cies to be characterized by altruistic behaviors, while other
otherwise similar species are not altruistic. While behavioral
geneticists generally attempt to determine the relative contri-
bution of genetic factors on variations in behavior within a
species, sociobiologists tend to look at the influence of such
factors on variations in behavior across species. In turn, socio-
biological ideas spurred the development of evolutionary psy-
chology, which seeks to relate variations in contemporary
human behavior to a history of natural selection.
Quantitative Genetics
Quantitative genetics assesses the relative contribution of
genetic versus environmental factors (nature vs. nurture) in
determining individual variations in traits within a species or
population. If only the phenotypes are measured, and the
genetic bases for the familial resemblances discussed below
only assumed, quantitative genetics must rely heavily on statis-
tical techniques to arrive at its conclusions. It should also be
noted that, as with population genetics, quantitative genetic
findings are descriptive of relative genetic and environmental
influences averaged across a particular population operating
in a particular environmental context and subject to all of
the forces discussed above that change allelic and genotypic
frequencies.
The starting point for quantitative genetics is variability.
As discussed at the beginning of this article, stable individual
differences in behavior or any other trait must be the result of
genetic and/or environmental processes that are both unique
for each individual and consistent in their effects across time.
Heritability is defined as that proportion of the variability of
a phenotype that can be accounted for by genetic factors or
h 2 = Vq/Vp, where V c and V P are the genetic and phenotypic
variabilities, respectively. Traits without any variability, such as
the ability to use language in humans, may be entirely under
genetic control, but the lack of variability for the trait precludes
any statistical tests to determine the relative influence of genes
and environment.
306 Behavioral Genetics
The counterpart to the assumptions about variability is that,
if relatives resemble each other (i.e., do not vary) for some trait,
it must be because they share the same genetic and/or environ-
mental determinants of the variability of the trait. By compar-
ing the relative degrees of resemblance (correlations) for the
phenotype of interest across relationship types that theoreti-
cally differ in their underlying shared genetic and environmen-
tal determinants, the quantitative geneticist can infer the extent
to which genetic versus environmental variability accounted
for the variability of the phenotype.
For the commonly used study of twins reared together,
identical (monozygotic or MZ) twins are known to share all
of their genetic variability (they literally have the same genes),
as well as sharing the same family environment. Fraternal
(dizygotic or DZ) twins are like nontwin siblings in only
sharing half of their additive (additive vs. nonadditive genetic
effects are discussed below) genetic variance, but they are like
MZ twins in sharing the same family environment. Hence,
if MZ twins are more alike than DZ twins, it is assumed that
this is because the former share twice as much genetic varia-
bility as the latter. Thus, doubling the difference between the
MZ and DZ correlations should estimate the heritability. For
example, adult MZ twins reared together typically correlate
around 0.75 on tests of intelligence, while DZ twins reared
together correlate around 0.50. Thus the heritability would be
h 2 = 2 (r MZ - r DZ ) = 2(0.75 - 0.50) = 0.50, or 50% of the
variability in intelligence is accounted for by genetic factors.
Fifty percent heritability means that part of the MZ resemblance
of 0.75 is still unaccounted for. This part, 0.25 or 25%, is
presumably due to the effects of shared family environment.
Genetic influences can be classified as additive or nonaddi-
tive. Additive effects are analogous to the idea of 'gene dosage,'
in that for polygenically determined traits familial resem-
blances are predicted from the extent of the linear accumula-
tion of shared genes. For example, each parent shares on
average 50% of the additive genetic variability for a phenotype
with his/her offspring by transmitting half of the offspring's
genome. Siblings share with each other on average 50% of the
additive genetic variability, because for each of the two parents
the probability is 25% that the same genes were transmitted to
both siblings (50% to one sibling x 50% to the other). Addi-
tive genetic variability is important in that it determines the
extent to which a trait can be changed in a population through
natural or artificial selection.
The two types of nonadditive genetic effects, dominance and
epistasis, can be thought of as being analogous to statistical inter-
actions. The dominance effects described above for single genes
are interallele interactions, where the effect of one allele cannot
be predicted without knowing the effect of the other allele. Thus,
parents and offspring do not share any dominance genetic varia-
bility, since dominance effects require particular combinations
of alleles from both parents. Siblings, on the other hand, do
share 25% of dominance genetic variability, since that is the
probability of both siblings receiving exactly the same combina-
tion of alleles (at least for a two-allele gene) from both parents.
Epistasis is an intergene interaction, where the effect of one
gene cannot be predicted without knowing the effect of some
other gene or genes. Only identical twins share epistatic genetic
variability, since they inherit not only the same genes, but also
the exact same combination and ordering of the genes.
Environmental effects can be classified as shared (com-
mon), special twin, and unshared (specific). Shared environ-
mental variability represents the assumption that individuals
growing up in the same family may resemble each other,
because they are influenced by the same environmental deter-
minants of the phenotype. Adoptive siblings and adoptive
parents and their offspring provide strong tests of shared envi-
ronmental effects, since presumably this is the only factor that
would create any familial resemblances. Special twin environ-
mental variability is a special form of shared environmental
effects that assumes that twins may share more similar envi-
ronmental influences, perhaps due to being of the same age,
than nontwin siblings.
Unshared environmental effects make family members dif-
ferent from each other. This may be due to idiosyncratic events
in a person's life and differential treatment of siblings by
parents, as well as measurement unreliability. Differences
between identical twins reared together must be due to un-
shared environmental variability, since that is the only factor
MZs don't have in common.
Genetic effects may also be correlated with environmental
effects. Gene-environment correlations can be passive, where the
child is born into a family environment in which environmen-
tal conditions created by the parents are consistent with the
parents' and the offspring's genotype. They can be evocative,
where the environment changes in response to the offspring's
genotype, or active, where the offspring's genotype causes
him/her to change the environment to make it more consistent
with his/her genotype. Positive gene-environment correla-
tions can presumably magnify genetic effects, while negative
gene-environment correlations can diminish them. Gene-
environment correlations also create the possibility that mea-
sures of the family environment used by psychologists, such as
parents' socioeconomic status, may in fact be confounded by
genetic factors.
Finally, there is the possibility of gene-environment interac-
tions, where genetic effects cannot be predicted unless the
environmental conditions are known. For example, genes
determining high intelligence appear to be better expressed if
a child is raised in an enriched environment.
Table 1 presents the theoretical extents to which the factors
presented above can cause relatives to resemble each other
across various relationship types. Quantitative genetic studies
typically compare phenotypic resemblances across two and
occasionally more relationship types, and some of the most
commonly used designs will be described below. For several of
these factors, the differences across relationship types are too
small to reliably detect their effects. For instance, it can be seen
that dominance and epistatic genetic effects are basically indis-
tinguishable, and gene-environment correlations and interac-
tions are barely distinguishable from pure genetic effects.
Before describing quantitative genetic designs used to esti-
mate heritabilities for human behavior, animal behavioral
genetic studies should be mentioned. To the extent that, due
to evolutionary reasons, other animal species, such as drosoph-
ila (fruit flies), nematodes, mice, and rats, share with humans
the same genes and resultant physiological structures and pro-
cesses that determine behavior, animal genetic studies have
a number of advantages in providing information about
the genetic determinants of human behavior. Unlike human
Behavioral Genetics 307
Table 1
Components of shared variance for different familial relationships
Genetic Environmental
Gene-environment
Additive
Dominance
Epistatic
Shared
Twin
Specific
Correlation
Interaction
Identical (MZ) twins
1.0
1.0
1.0
1.0
1.0
0
1.0
1.0
Fraternal (DZ) twins
0.5
0.25
0
1.0
1.0
0
<0.5?
<0.5?
Nontwin siblings
0.5
0.25
0
1.0
0
0
<0.5?
<0.5?
Parents/offspring
0.5
0
0
1.0
0
0
<0.5?
<0.5?
Half-siblings
0.25
0
0
1.0
0
0
<0.25?
<0.25?
Uncles-aunts/nephew-nieces
0.25
0
0
0?
0
0
<0.25?
<0.25?
First cousins
0.125
0
0
0?
0
0
<0.125?
<0.125?
Adoptees
0
0
0
1.0
0
0
0
0
adoption studies, in research with animals prenatal environ-
ments can be controlled by transplanting fetuses across ani-
mals with different genotypes. Rearing environments can be
controlled to eliminate the effects of such environmental dif-
ferences or manipulated to test for gene-environment interac-
tions. Inbred strains of genetically identical animals can be
used to control for genetic variability. Heritabilities can be
estimated by assessing the phenotypic response to selective
breeding for extremes on the trait.
Family studies with humans involve obtaining parent-
offspring and/or sibling correlations. As can be seen in Table 1,
such studies cannot disentangle the effects of genetic influences
from shared environmental influences, although they can pro-
vide an 'upper limit' for heritability. The extension of a family
design to the examination of large pedigrees can provide evi-
dence for the Mendelian transmission of a trait under the
control of a single gene.
Comparisons of reared-together identical versus fraternal
twin resemblances can provide an estimate of heritability, but
cannot easily disentangle additive versus nonadditive genetic
effects. As can be seen in Table 1, MZs may resemble each other
more than DZs also due to greater shared gene-environment
correlations and interactions. In addition, these comparisons
assume that greater MZ than DZ resemblances for a trait are not
the result of MZs being treated more alike than DZs causing
more phenotypic similarity in MZs. Studies have validated this
'equal environments' assumption for intelligence and person-
ality, but it remains untested for many other phenotypes.
Adoption studies involve some combination of comparing
resemblances of biological parents with their adopted-away
offspring versus the resemblances of the adoptive parents
with the offspring or comparing adoptive versus nonadoptive
sibling resemblances. Such designs have the advantage of the-
oretically separating additive genetic and shared environmen-
tal influences, although pre- and neonatal environmental
effects cannot be controlled for. Attempts by adoption agencies
to match adoptive parents with the biological parents ( selective
placement ) could confound the analyzes, but this effect has
been found to be negligible in studies of intelligence and
personality. The typically narrower range of rearing environ-
ments for samples of adoptive parents compared to biological
parents may also attenuate shared environment effects.
In order to illustrate two other important aspects of quanti-
tative genetic analysis. Figure 2 presents a path model for the
genetic transmission of two different phenotypes (subscripts x
and y) from fathers (subscript F) and mothers (subscript M) to
their biological offspring (subscript O). In the model, each
phenotype P is determined by a genetic factor G and an envi-
ronmental factor E. The Gs of the parents are linked to the Gs
of the offspring by 1/2, reflecting the 50% of shared additive
genetic variability of parents and offspring, which in this case is
confounded by shared environment.
The first additional aspect of note is the arrows connecting
the Ps at the top of the figure. These represent phenotypic
assortative mating between fathers and mothers and, as can
be seen, these arrows allow for greater similarity of the genetic
variability transmitted by both parents to the offspring. For
instance, mother's transmission of Gmj to G 0x is augmented
by an additional path from father's G Fx through the assortative
mating path to Gmh- This must be accounted for in estimating
the heritability h 2 .
The other aspect of note is the r c arrows connecting the G x s
and G y s. These represent the genetic correlations discussed above
in terms of linkage and pleiotropy. As can be seen in Figure 1,
these genetic correlations are shared genetic effects, and they
may be indicative of functional genetic relationships. For
example, a significant genetic correlation between performance
on tests of verbal ability and performance on tests of spatial
ability would indicate some common physiological/genetic
process for these abilities.
Several important caveats should be reiterated. As was the
case for population genetics, heritabilities reflect the relative
contribution of genetic factors on the variability of a trait for
just the particular sample of individuals studied and the partic-
ular environmental conditions in which they are living. Heri-
tability estimates may not generalize to other populations, and
they may greatly change if environmental conditions change.
Similarly, high heritabilities do not preclude a trait from being
influenced by the environment. Another caveat refers back to
the idea that high heritability is dependent on high phenotypic
variability, but high phenotypic variability is associated with
traits that have not been highly selected for, that is, are not
important for biological fitness, unless there is a balanced
polymorphism.
Molecular Genetics
The DNA molecules in every living cell consist of two strands of
phosphate and deoxyribose sugar groups held at a fixed dis-
tance from each other by pairs of nitrogenous compounds
called bases. Each of the four bases is attached to one strand
308 Behavioral Genetics
IV
Figure 2 Path model of bivariate parent-offspring genetic transmission.
and can only pair with the corresponding base on the other
strand in a particular combination: adenine with thymine and
guanine with cytosine. The molecule itself is held together by
its double helix structure, as discovered by Watson and Crick.
When the molecule is 'unzipped,' each strand thus is a mirror-
image of the other strand and can be used as a template for the
construction of proteins and enzymes, when three-base ribo-
nucleic acid (RNA) units carrying amino acids link up with the
DNA strand in the order dictated by the exposed base pairs.
Through this transcription and replication mechanism, each
cell programs the synthesis of the structures and chemicals
needed to maintain itself, and the DNA in the cell's nucleus
carries the program to create new cells. Also through this mech-
anism multicellular organisms transmit the program for the
creation of new organisms. These strands of nuclear DNA are
called chromosomes, and a gene is simply a segment (locus) of
the chromosome that codes for some protein or enzyme. In
sexually reproducing organisms, the 'blueprint' for the organ-
ism resides in one or more pairs of chromosomes (26 pairs in
humans), and each parent transmits one half of each pair of
chromosomes in the sex cells to the offspring. This is the bio-
chemical basis for Mendel's hypothetical genes and alleles.
Autosomal genes are those that are on the 'normal' chro-
mosomes not involved in determining gender. When the gene
(allele) from one parent pairs up with the gene from the other
parent at a locus, one gene may biochemically 'turn off the
transcription mechanism of the other gene. This is what Men-
del observed as dominance.
Chromosomes that determine gender are often different
from the other chromosomes in that both members of the
pair may not be equivalent in size and, hence, amount of
genetic material. In humans, females carry one pair of so-called
X- chromosomes, while males carry an X-chromosome paired
up with a much smaller Y-chromosome, which carries almost
no genetic information other than to turn on the process for
altering the developmental course of the embryo to become
a male. Traits coded for by genes on the X-chromosome in
males (which is transmitted by the mother) will thus be
expressed, including traits on recessive genes that might other-
wise have been inactivated by a dominant gene on the other
X-chromosome that females receive from the father. The results
of so-called X-linkage are seen, for example, in the higher
incidence in males of the relatively common form of congeni-
tal mental retardation called the fragile-X syndrome, where a
Behavioral Genetics 309
piece of the X-chromosome is 'loose' or broken off and the
genes on that piece are disrupted.
One theme that emerges from an understanding of the
mechanisms of genetic transmission is the Darwinian idea of
the importance of maintaining variability in a species. Sexual
reproduction itself, although costly to the organism in terms of
energy use, lost opportunities for food gathering, and vulnera-
bility to predation, is clearly meant to create new genetic com-
binations in the offspring by combining genes from both
parents. The process of splitting the chromosome pairs for
the sex cells also involves recombination of genetic material,
as the members of the pair cross over each other and exchange
genes. Mutations that spontaneously occur as a result of DNA
replication errors or environmental insults can alter the trans-
mitted genes.
Another theme that emerged early on from molecular
genetics was that some human behaviors, particularly mental
retardation, were clearly the result of simple genetic mechan-
isms. Errors in chromosomal replication and division at times
result in missing or duplicated chromosomes in the sex cells
transmitted to the offspring. The offspring then end up with a
missing member or piece of a member of a chromosome pair
or with an extra chromosome (trisomy). These chromosomal
defects are usually fatal for the fetus, when they occur on the
larger chromosomes, and usually result in some degree of
mental retardation, when they occur on the smaller chromo-
somes of a viable offspring. Besides the fragile-X syndrome
noted above, another common form of mental retardation is
Down's syndrome, which is a product of a trisomy on chro-
mosome 21.
Another good example of a simple genetic mechanism
responsible for mental retardation is phenylketonuria, which
is caused by a recessive autosomal gene that fails to code for the
synthesis of an active enzyme responsible for the breakdown of
the amino acid phenylalanine. The build-up of phenylalanine
in the infant depresses the transport of amino acids to the
brain and hinders normal nervous system development. The
pleiotropic effects of this genetic defect include hyperactivity,
irritability, and moderate to severe mental retardation. This
syndrome is also interesting in demonstrating that genetically
determined traits can be modified by the environment. Infants
who are detected early with the double recessive genotype can
be put on a low phenylalanine diet, and this diet has been
shown to be effective in virtually eliminating the deleterious
effects of the gene.
A technique called linkage analysis is used to confirm the
relationship of the transmission of genes or genetic markers
with the transmission of specific traits in families. This involves
identifying several large family pedigrees (two or more genera-
tions, with large numbers of siblings in each generation) or
large numbers of sibling pairs with a high incidence of the trait
of interest (e.g., some form of mental retardation). Each indi-
vidual in a pedigree is assessed for the occurrence or absence of
the trait, and the pattern of transmission of the trait through
the pedigree can be tested against the pattern expected by
different kinds of Mendelian single-gene transmission modes,
for example, autosomal dominant, X-linked recessive, etc.
In addition, each individual is also tested for the co-occurrence
of some other trait (a genetic marker), for example, a blood
group type, known to be controlled by a single gene at an
approximate locus on a particular chromosome. If the trait of
interest consistently co-occurs with the linkage marker, that is,
there is a genetic correlation, then the traits are not indepen-
dently assorting and must be close to each other on the same
now-identified chromosome. The deviation from expected
Mendelian ratios for independent assortment, indicative of
linkage, is expressed as a so-called LOD (logarithmic odds)
score.
Linkage analysis is a special case of an association study,
where the aim is to correlate markers of DNA variation (partic-
ularly allelic variation) with phenotypic variation across indi-
viduals. The simplest association study design is a two-group
comparison ( case control) of individuals differing in some
phenotype of interest. While this design is most advantageous
in terms of ease of recruiting subjects and consequent general-
izability, its internal validity is compromised by numerous
factors, including the problem of population stratification,
where DNA differences between groups may result from irrele-
vant DNA differences in different genetic lineages within
the groups. Sibling pair designs that compare siblings who are
similar versus different in the phenotype of interest elimi-
nate the problem of population stratification but cannot delin-
eate the mode of genetic transmission. Family data, such as
parents + child trios, are a special case of an association study,
which allows combining linkage and allelic association infor-
mation. The full pedigree design described above is the most
sensitive for finding genetic associations, but recruiting such
pedigrees is difficult, which often limits generalizability.
Today markers of DNA differences typically involve micro-
satellite markers, involving many alleles, or more commonly
single nucleotide polymorphisms (SNPs) based on a nucleotide
base difference across individuals on a short strand of DNA.
SNPs are detected through microarrays of DNA probes that
fluoresce when they find their match (hybridize) in a sample
of DNA (typically amplified through a replication process
called polymerase chain reaction). With the mapping of the entire
human genome in 2001 and the concomitant and subsequent
developments in biochemical and computing technologies for
high-throughput genotyping, research is moving away from
candidate gene studies that look at a few promising genes to
genome-wide association studies that use tens of thousands of
SNPs to detect associations across all the known genes.
By now, however, molecular genetic studies have confirmed
that for nearly all behavioral traits of interest, whether for
normal behavioral variation or psychopathology, multiple
genes are associated with their etiology. Furthermore, the
Human Genome Project determined that, instead of the
100 000-140 000 genes formerly thought to be in the human
genome, there are only 25 000 genes, and research has shown
that the same genes are often involved in diverse phenotypes, for
example, in schizophrenia and bipolar disorder. These findings
have emphasized the importance of searching for quantitative
trait loci (QTLs), small genetic effects for polygenically deter-
mined, continuously distributed phenotypes, but to be able to
detect such small effects requires massive sample sizes. Genetic
effects must clearly be looked at from a developmental perspec-
tive that not only considers what genes are involved, but also
when they are turned on and off in development, how they
interact with other genes, and how they interact with environ-
mental factors over the lifespan of the individual.
310 Behavioral Genetics
Behavioral Genetics and Psychology
After a contentious start, behavioral genetics has become an
accepted part of mainstream psychology. By now significant
heritability has been well-established for most stable individ-
ual differences in behavioral domains of interest to psy-
chologists, and association studies are rapidly linking these
phenotypes to particular genes. However, the early hope of
finding simple genetic mechanisms for these behaviors,
mechanisms which could be easily targeted for physiological
interventions to eliminate psychopathology or enhance nor-
mal behavioral attainments, was clearly overly optimistic. Per-
haps this should not have been surprising, given the iterative,
dynamic, and systemic nature of Darwinian evolution. Ironi-
cally, behavioral genetics has confirmed the importance of
the environment as a determinant of behavior, but has led to
a more nuanced understanding of how that environment
works in conjunction with the genes and physiology of the
individual.
See also: The Behavior-Genetics of Intelligence; Evolutionary
Psychology; Human Intelligence; Intellectual Disabilities.
Further Reading
Plomin R, DeFries JC, McClearn GE, and McGuffin P (2008) Behavioral Genetics, 5th
edn. New York: Freeman.
Rutter M (2006) Genes and Behavior: Nature-Nurture Interplay Explained. Hoboken, NJ:
Wiley.
Behavioral Medicine
J A Jimenez and P J Mills, University of California, San Diego, La Jolla, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behavioral intervention Theory-based strategies
and activities aimed at promoting healthy
behaviors.
Behavioral medicine An interdisciplinary field concerned
with the development and integration of behavioral and
biomedical science, knowledge and techniques relevant to
health and illness, and application of this knowledge, and
techniques relevant to prevention, diagnosis, treatment, and
rehabilitation.
Ecological models Models proposing that behaviors are
influenced by the interactions between multiple levels of the
social environment: intrapersonal, sociocultural, policy, and
physical environmental factors.
Health behavior Psychological processes, such as values,
perceptions, beliefs, personality traits, and mood, as well
as overt behavioral patterns, actions, and habits that relate
to health.
Health promotion Any planned combination of educational,
political, environmental, regulatory, or organizational
mechanisms that support actions and conditions of living
conducive to individuals, group, or communities.
Health psychology The study of psychological processes
related to health and health care.
Primary prevention Efforts to forestall the onset of illness
or injury during the prepathogenesis period.
Psychoneuroimmunology The study of interactions
among the behavioral, neural, endocrine, and immune
systems on maintenance of healthy functioning, and
development and progression of illness.
Secondary prevention Efforts that occur after individuals
are experiencing symptoms or who have been diagnosed
with a disease.
Tertiary prevention Efforts aimed at individuals who
are already severely affected by a disease (e.g., impairment
or disability).
Introduction
Much progress was made in the twentieth century to improve
health and prolong life. At the turn of the century, the leading
causes of death were mostly communicable in nature, how-
ever, advances in science produced improvements in saniti-
zation and hygiene, medication therapies, and vaccines to
control - and in some cases eradicate - infectious diseases.
As a result, in many industrialized and some developing
nations there has been a shift in disease patterns. Currently
cardiovascular diseases, which have complex etiologies and are
associated with numerous lifestyle behaviors, are the leading
cause of death worldwide. Modifiable behaviors such as expo-
sure to tobacco products, consumption of high-fat diets, and
physical inactivity are responsible for approximately one-third
of all deaths within the United States.
Understanding how lifestyle behaviors work in concert with
genetic, biological, and psychosocial factors to protect against
or exacerbate disease requires a multifaceted approach. Behav-
ioral medicine is an interdisciplinary field that studies the
interplay among psychosocial, biological, and behavioral path-
ways on the natural progression of disease. Perhaps what sets
behavioral medicine apart from other related fields is that
in addition to studying the mechanisms of disease, it focuses
on transferring this knowledge to behavioral interventions
that can be delivered in medical and community settings.
Behavioral medicine brings together theoretical knowledge
and methodological innovations from the social sciences
(psychology, sociology, anthropology, epidemiology, and health
promotion) as well as the biomedical sciences (psychiatry,
physiology, endocrinology, immunology, pharmacology, nutri-
tion, biology, and chemistry). Moreover, the field of behavioral
medicine unites academic researchers as well as health pro-
fessionals, such as physicians, nurses, dentists, and health
educators.
Brief Historical Context
The concept of behavioral medicine predates the formal
discipline of psychology, and can be traced to Hippocrates.
Its evolution is a result of innovations in related disciplines,
such as medicine, psychology, psychiatry, and psychosomatic
medicine. There were several landmark events that helped
shape the field as we know it today. A few of these are men-
tioned herein. In 1800s, Willhelm Wundt, the founder of
experimental psychology, and Emil Kraepelin began to inte-
grate psychology within a medical setting. By 1869, all symp-
toms that were derived from the mind were classified as
psychogenetic; thus, there was a formal delineation between
symptoms of physical origin versus those of psychological
origin. Claude Bernard introduced the concept of homeostasis,
which refers to the body's ability to self-regulate its internal
milieu in order to maintain healthy functioning. Bernard's
work focused on the role of the nervous system in maintaining
homeostasis. Building on Bernard's research, in 1932 Walter
Cannon produced a detailed account of the relationship
between the body's reactivity to stress and negative emotions,
such as anger and fear. He later coined the term fight or flight
response to describe the body's physiological reactions when
confronted with stressful stimuli. Based on this seminal work,
Hans Selye and Harold Wolf in the 1950s would go on to study
the physiological responses to prolonged stress, and find
that chronic stress weakened the body, making it vulnerable
311
312 Behavioral Medicine
to infection. This work would not only contribute to the field
of behavioral medicine, but also to other related fields such as
biopsychology and psychoneuroimmunology (PNI).
By the 1930s, the field of psychiatry had emerged as an
important medical specialty. In 1936, the Menningers, a well-
known family of psychiatrists, noted that aggressive men were
more prone to cardiovascular conditions. Physicians interested
in mind-body interactions were compelled by the psychoana-
lytical approaches of Sigmund Freud. There was an interest in
understanding how conscious and unconsciousness emotional
states contributed to the development of pathologies. The
assumption was that if the psychological conflict could be
resolved then its associated physiological manifestations
would also diminish, and eventually disappear. In 1938, the
psychiatrist Helen Flanders Dunbar was appointed as the first
editor of a new journal - Psychosomatic Medicine - which later
became the official publication of the American Psychosomatic
Society, founded in 1942. Dunbar and her contemporaries
contributed valuable knowledge about the interplay between
external stress, psychological conflict, and organic disease;
however, the field of psychosomatic medicine was initially
criticized for its lack of scientific rigor and failure to provide
efficacious treatments. As a result, psychosomatic medicine
began to adopt more rigorous research methods, and integrated
the sociobehavioral and biological sciences. Behavioral medi-
cine was developed, in part, as a response to the initial short-
comings and narrower focus of psychosomatic medicine.
Prior to the 1970s, the reigning paradigm on disease
and health was focused on biological explanations; however,
increasing awareness about lifestyle behaviors and environmen-
tal factors ushered in a new paradigm on health and disease.
George Engel called for a medical model that accounted for
the interactive process between biological, psychological, and
social factors on health and disease progression. Engel asserted
that in situations of loss the 'giving up' response may precede
the development of illness among individuals who already
had a genetic predisposition for a given illness. This new
model was referred to as biopsychology, an integrative approach
that demonstrated the interdependence between contextual
factors, coping, and biological pathways. Concurrently, Skin-
nerian psychology fell out of favor, and Albert Bandura's
social learning theory gained popularity. Social learning
theory and other cognitive behavioral models that followed
focused on the role of individual choice, and how these choices
were influenced by perceptions, emotions, and social norms.
Thus, the changing paradigm highlighted the role of personal
responsibility, cognitions, behavior, and the social environ-
ment in which the behavior took place.
The Emergence of Behavioral Medicine
While there is no one event that marked the creation of
behavioral medicine, the 1970s gave way to a culmination
of events that defined and formalized the field. In 1974,
Canada produced a national report on the role of lifestyle
behaviors, and environmental influences on health. In the
same year, the United States passed the Health Information
and Health Promotion Act, which created what is now known
as Office of Disease Prevention and Health Promotion. In
1979, the United States produced the first Healthy People
report, which was a summation of the literature to-date on
the relationship between personal behavior and health status.
Healthy People gave policymakers and the general public a
user friendly guide to lifestyle behaviors that may reduce dis-
ease risk. It was also in the late 1970s that the US government
recognized the field of behavioral medicine and initiated
mechanisms for the funding of behavioral medical research.
In 1979, the National Institutes for Health (NIH) established a
study section on behavioral medicine, and soon after other
government agencies followed suit. In 1984, the Centers for
Disease Control and Prevention (CDC) established the Behav-
ioral Risk Factor Surveillance System (BRFSS), which is a state-
based telephone survey that collects information on health risk
behaviors, preventive health practices, and health care access
primarily related to chronic disease and injury. At first only 15
states administered the BRFSS, but by 1994 the system was
nationwide. In 1995, the NIH established the Office of Behav-
ioral and Social Sciences Research (OBSSR) in recognition of
the dynamic relationship between behavior and social factors
on morbidity and mortality. The purpose of OBSSR is to stim-
ulate behavioral and social sciences research throughout NIH
and to integrate these areas of research into other NIH centers.
Formalization of the Field
Behavioral medicine was formally recognized as a new field in
1977 at the Yale Conference for Behavioral Medicine. In 1978,
the United States National Academy of Science's Institute
for Medicine developed the definition that remains today,
which centers on the integration of behavioral and biomedical
sciences, and application of this knowledge to prevent, diag-
nosis, and treat illness. In 1979, Ovide Pomerleau and Paul
Brady established one of the earliest behavioral medicine
research centers at the University of Pittsburg, and in the
same year they established the Society of Behavioral Medicine
(SBM). At first the society's main emphasis was on medication
habits associated with stress, obesity, and smoking; however,
they received criticism for this narrow focus and widened
their focus to include a larger breadth of health problems.
Today SBM entertains a wide range of topics, including:
adolescent health, aging, anxiety, cardiovascular disease, chronic
pain, depression, diabetes, eating disorders, environmental
health, headaches, HIV/AIDS, insomnia, minority health,
quality of life, sexually transmitted diseases, social support,
and substance abuse. The SBM sponsors an annual meeting
to bring together researchers and clinical practitioners. The
society's official journal is the Annals of Behavioral Medicine,
which publishes original empirical articles - including reports
of randomized controlled trials, observational studies, or other
basic and clinical investigations - and integrative reviews of
the evidence for the application of behavioral interventions
in health care. Since the establishment of SBM, an inter-
national federation of over 20 behavioral medicine societies
across the globe formed the International Society Behavioral
Medicine (ISBM). The first ISBM meeting was held in Sweden
in 1990, and the first issue of their journal International Journal
of Behavioral Medicine was published in 1994. The goals of
ISBM are to stimulate research and practice as well as dissemi-
nate behavioral medicine research via the society's journal,
and international teaching seminars.
Behavioral Medicine 313
Understanding Interactive Pathways of Disease
During the latter half of the twentieth century, many lifestyle
behaviors and psychosocial variables were found to be asso-
ciated with the development and progression of chronic dis-
eases. This stimulated further scientific investigation into the
mechanisms or pathways of disease. Without understanding
the underlying catalysts of disease it is difficult, if not impossi-
ble, to prevent, diagnose, and treat the condition. There are
many pathways of disease; herein we note some of the most
widely researched behavioral, psychosocial, and biological
mechanisms associated with chronic disease, as well as high-
light current and future research directions.
Behavioral Pathways
As disease patterns have shifted, so has the emphasis on under-
standing the influence of lifestyle and behaviors on health
outcomes. Currently, tobacco, high-fat diets, and physical inac-
tivity are the leading actual causes of death in the United States.
Actual causes of death are defined as lifestyle and behavioral
factors that contribute to mortality. While it will not be dis-
cussed in detail here, it is fair to note that lifestyle behaviors
also interact with each other, which in turn can have additive
or multipliable effects on risk of disease development or exac-
erbate already existing conditions. That is to say that many
times risky behaviors cluster together, and their compounded
effects can diminish health and well-being.
Tobacco exposure
The World Health Organization (WHO) reports that it is the
second leading cause of death worldwide. Exposure to tobacco
products is associated with the development of cardiovascular
disease, cancers of the mouth, lung, and esophagus, as well as
respiratory conditions including emphysema, bronchitis, and
chronic obstructive lung disease. In 2006, the US Surgeon
General reported that there is no risk-free level of tobacco
exposure. While the effects of first- and second-hand smoke
have been well documented and are associated with multiple
negative health outcomes, there is a growing body of litera-
ture that suggests that third-hand smoke may also be of
concern. Third-hand smoke is currently defined as residual
tobacco smoke contamination that remains after the cigarette
has been extinguished.
The WHO reports that while global tobacco consumption is
leveling off and even decreasing in some countries; more peo-
ple are smoking, and smokers are consuming more cigarettes.
It is projected that by the year 2030 industrialized nations
will have approximately 3 million smoking related deaths,
and it is projected that developing nations will have approxi-
mately 7 million deaths. The WHO estimates that smoking
prevalence will continue to increase as a function of popula-
tion growth; however, while prevalence rates may go down, the
actual number of smokers worldwide may increase.
Physical inactivity and diet
There has been a dramatic increase in prevalence rates of
obesity and type 2 diabetes mellitus. In the United States in
particular, many researchers claim that obesity has reached
epidemic proportions. This is disconcerting because obesity
and diabetes are major risk factors for many chronic condi-
tions, including cardiovascular diseases. Concurrent with the
increase in obesity, there has been a rise in the prevalence of
type 2 diabetes mellitus. Unlike type 1 diabetes mellitus, which
is an autoimmune disease that develops in childhood, type
2 diabetes often develops in adulthood as a result of obesity
and poor lifestyle behaviors. Type 2 diabetes usually begins as
an insulin resistance disorder in which the cells do not use
insulin properly, and the pancreas gradually loses its ability to
produce insulin. If not tteated, patients can suffer innumerable
adverse consequences, including impaired circulation to the
lower limbs (which may result in chronic unhealing wounds
and/or amputation), glaucoma, and cardiovascular diseases.
Obesity and diabetes are associated with poor diet and
physical inactivity. The WHO estimates that physical inactivity
causes 2 million deaths per year. With regard to diet, there has
been a push to increase fruit and vegetable consumption,
however, many individuals partake in high-fat diets, and this
coupled with physical inactivity contributes to the rising rates
of obesity.
Psychosocial Pathways
Psychosocial factors have been associated with lifestyle beha-
viors as well as biological outcomes. In particular, the field of
health psychology, which is encompassed within the field
of behavioral medicine, has made significant contributions to
understanding the role of core psychological processes on
health, including social support, stress and coping, personality
disorders, and adaption to chronic illness. The field of Health
Psychology seeks to advance contributions of psychology to
the understanding of health and illness and encourages the
integration of biomedical information with current psycholog-
ical knowledge. Psychosocial factors can be grouped into three
categories: intrapersonal, interpersonal, and societal. The con-
structs within these groupings are interrelated and interdepen-
dent in ways that may affect health outcomes. While many
psychosocial factors have been studied, we will only attend
to the most widely studied constructs related to health and
disease processes in this article.
Intrapersonal factors
Intrapersonal characteristics refer to self-attributes, and these
can include personality Uaits and mood disorders. The con-
struct of self-esteem has been related to many health-protective
behaviors and favorable clinical outcomes. It refers to evalua-
tions of self-worth and the extent to which people value
or approve of themselves. Self-efficacy, a personal belief in
one's ability to perform a behavior or set of behaviors in a
way that will produce a desired outcome, has also been related
to increased frequency of health-protective behaviors. For
example, one's ability to use condoms correctly has been asso-
ciated with increased intentions to use condoms as well as
frequency of use. Personal control can be defined as personal
expectancies related to the extent to which individuals believe
an outcome is within their control. Personal control has also
been associated with increased conduct of health-protective
behaviors, including weight loss and exercise. However, while
theoretically higher levels of self-esteem and personal control
are generally considered to be favorable, the literature presents
314 Behavioral Medicine
inconclusive results, that is, these constructs are not always
associated with health, and in some cases they have been
associated with negative health outcomes.
Negative emotions have been widely associated with
health-damaging behaviors and poor clinical outcomes.
Depression is a mental disorder that presents with low mood,
loss of interest or pleasure, feelings of guilt or low self-worth,
disturbed sleep or appetite, low energy, and/or poor concen-
tration. The WHO reports that depression is the leading cause
of disability worldwide and that its prevalence is quickly rising.
In particular, depression is comorbidity for many chronic
diseases. Some researchers argue that the prevalence of comor-
bid depression is a result of limitations on physical and,
many times, cognitive functioning associated with living with
a chronic disease. However, there is a body of evidence to
suggest that depression shares many biological pathways with
chronic diseases such as heart failure and other cardiovascular
conditions. Thus, in some cases comorbid depression is not
merely an emotional reaction to the challenges of living
with a chronic disease, but also the result of biological mecha-
nism that explains mood change. Increased depression symp-
tomatology has been associated with increased risk of
smoking, physical inactivity, and nonadherence to medication
as prescribed by a physician.
Hostility has also been associated with negative health
outcomes. Hostility is composed of cognitive, affective, and
behavioral components: (1) cognitive aspects include negative
beliefs about and attitudes toward others; (2) affective compo-
nents include the frequency, and intensity of unpleasant emo-
tions ranging from irritation to rage; and (3)behavioral aspects
of hostility include actions or intentions to either verbally
or physically harm others. Cognitive aspects of hostility
have been associated with coronary heart disease, and affective
aspects associated with hypertension. Hostility is also asso-
ciated with a variety of health behaviors, including smoking,
alcohol consumption, and obesity.
Interpersonal factors
Interpersonal factors refer to the influence of social interactions
on health and well-being. This can include formal and infor-
mal exchanges with family members, friends, neighbors, and
coworkers. Perceived social support has been widely discussed
in behavioral medicine literature. Social support refers to the
social resources that individuals perceive are available to them.
They are usually classified into one of three main categories:
emotional, instrumental, and informational support. Emo-
tional support refers to the availability of one or more persons
who can listen sympathetically, as well as provide signs of
caring and acceptance. Theoretically this type of support
should enhance self-worth, reduce negative affect, and moti-
vate coping behaviors. Instrumental support refers to tangible
aid and practical help, such as help with childcare and provi-
sions of transportation or money. Theoretically, this type of
support should help individuals solve practical problems,
which allows for increased rest and relaxation and a reduction
of psychological stress. Finally, informational support refers to
receiving information about resources. An example of this type
of support might be helping a friend develop an effective
course of action to resolve a problem or suggesting a resource
(shelter, counselor, community organization, etc.). In general,
social support is thought to have positive effects on health
because it 'buffers' stress. Thus, those who report higher levels
of social support hypothetically should have better health out-
comes. Studies indicate that higher levels of social support,
particularly emotional support, is widely associated with psy-
chological well-being, including lower levels of depression.
Higher levels of social support have also been linked to lower
physiological (e.g., blood pressure and/or heart rate) responses
to stress.
Social conflict is another interpersonal dimension that may
affect health; however, it has received much less attention than
social support. Social conflict refers to negative social interac-
tions that may occur within social relationships including
arguments, criticism, and unwanted demands. Social conflict
has been related to elevated cardiovascular and neuroendo-
crine biomarkers.
Societal factors
Societal factors refer to education, income, occupation, race,
and ethnicity as well as neighborhood and community con-
text. In the 1980s, there was a growing consciousness that
lifestyle factors and disease were unevenly distributed across
the US population. Moreover, some populations were experi-
encing greater prevalence and incidence, as well as morbidity
and mortality associated with particular diseases. Many epide-
miological studies, such as the Whitehall Study conducted
by Michael Marmot, have demonstrated that lower socio-
economic status (SES) or position is associated with higher
prevalence of health-damaging behaviors, and poorer health
outcomes. Traditionally, SES has been defined by education,
income, and occupation. These variables have been assessed
individually, and have also been combined to form composite
scores. While the pathways between SES and health are com-
plex, and still not well-defined, increased funding opportu-
nities and the subsequent research in this area have provided
some important findings. Lower SES has been associated with
higher smoking prevalence, lower consumption of fiber and
fresh fruits, physical inactivity, and heavy drinking. Moreover,
those living in lower SES environments may experience more
chronic stress associated with financial strain, crowding, noise
exposure, and unsafe neighborhoods. There are vast differences
in health and quality of life across different racial and ethnic
groups. In particular, there is a growing body of research
showing that minority populations have less access to quality
healthcare, and many groups have a higher prevalence of risk
behaviors associated with chronic disease. Within the United
States, there has been an interest in chronic psychological stress
associated with the perception of being treated unfairly based
on race or ethnicity. This construct has been linked to hyper-
tension, self-reported poor health, breast cancer, and mood
disorders, as well as obesity, smoking, infrequent condom
use, and substance use. Prolonged exposure to discrimination
may prepare the body to be more physically reactive to stressful
situations. Increased reactivity and engagement in unhealthy
behaviors may moderate the relationship between perceived
discrimination and health. With regard to environmental
factors geomapping software, and census data have helped
researchers to look more closely at neighborhood and commu-
nity characteristics, including the availability of fruit and vege-
tables and the number of fast food restaurants. The integration
Behavioral Medicine 315
of biodemography, a field concerned with the synergistic
effects of demographic and biological determinants on disease
and aging patterns, into behavioral medicine has helped to
reveal how environmental context interplays with behaviors,
psychological, and biological factors.
Health throughout the life course
In order to better understand these factors, many researchers
have adopted a lifespan approach. From this perspective,
health is viewed as a product of risk behaviors, protective fac-
tors, and environmental agents that are encountered through-
out the lifespan. The lifespan approach provides a framework
for interpreting how experiences during early years of life can
later affect health and functioning in adulthood. Moreover, this
approach allows for us to calculate cumulative, additive, and
even multiplicative impacts of risk behaviors and protective
factors on health outcomes. Improvements in longitudinal
study design and statistical methodologies have allowed behav-
ioral medicine research to study multiple risk and protective
factors on the natural progression of disease over time.
Biological Pathways
An important aspect of behavioral medicine is that it links
behavioral and psychosocial factors to physiological activity
and biological outcomes. Since its inception, the field of
behavioral medicine has been keenly interested in the sympa-
thetic nervous system (SNS). This is because the SNS has far
ranging relevance to almost all areas of health, including acute
and chronic stress, mood, psychiatric disorders, neuroimmune
regulation, cancer, and cardiovascular diseases. However, fur-
ther examination of mind-body interactions indicates that
there are many reciprocal relationships among the central
nervous system, the autonomic and neuroendocrine system,
and the immune system that affect health.
Psychoneuroimmunology
The field of PNI provides an integrative framework to
study complex interactions amongst neural, endocrine, and
immune systems, as well as behavioral and psychosocial fac-
tors on maintenance of healthy functioning, and development
and progression of illness. PNI has helped advance behavioral
medicine investigation of the pathophysiology of many chronic
diseases, especially cardiovascular disorders. The PsychoNeuro-
Immunology Research Society is an international organization
that supports all aspects of PNI research. Its international
journal, Brain, Behavior, and Immunity, was founded in 1987
and publishes peer-reviewed basic, experimental, and clinical
studies dealing with behavioral, neural, endocrine, and immune
system interactions in humans and animals.
The three main biological systems examined in PNI
include: ( 1 ) the central nervous system, which recognizes and
records experiences; (2) the autonomic and neuroendocrine
systems, which govern bodily functions; and (3) the immune
system, which organizes responses to infections and eliminates
tumor cells. Normal functioning of the multiple interconnected
regulatory pathways across these systems is necessary to main-
tain health; when pathways are compromised through loss of
regulation, disease can ensue. While each of these systems has
been studied separately, the innovation of PNI is that it aims to
understand how these systems work together in concert. It is
important to recognize that although the function of a single
hormone, cytokine, or neurotransmitter can be studied in isola-
tion, when taken out of the context of the broader physiological
milieu, the ability to fully understand its role is severely limited.
PNI and behavioral medicine: acute stress and
vaccination response
Behavioral medicine researchers often use findings from the
field of PNI to inform intervention design. One way that PNI
research has been applied to behavioral medicine is through
research on acute stress and vaccination response. We focus
here on some of the acute stress and vaccination response
findings because it is one of the most clinically relevant and
promising applications of PNI that has been applied to behav-
ioral medicine. Understanding SNS and endocrine effects on
immune functioning has been of considerable interest. Some
researchers have found that the acute stress response enhances
subsequent immune responses if the stress is experienced
shortly before the immune system is presented with certain
types of challenges. From an evolutionary perspective, it would
enhance survival, and subsequently fitness of the species to
have robust immune responses during times of stress, consid-
ering that stress can also suppress immune functioning, which
in turn makes the body more vulnerable to illness. Researchers
tested this hypothesis in rodent models, and found that rats
that were exposed to an acute stressor in close temporal prox-
imity to receiving a vaccination had higher levels of antibody
response than rats that experienced chronic stressors or those
who were not exposed to any kind of stressor. Additional
research on acute stress and vaccination suggests that rodents
with compromised immune systems experience higher levels
of antibodies compared to healthy rats; findings recently
replicated in humans. Studies in young healthy adults find
that concentric cycling, acute psychological stress, and weight
training exercise enhance responses to influenza and meningo-
coccal vaccination. Studies also suggest that immunocompro-
mised individuals have greater increases in levels of antibodies
produced as compared to healthy adults; thus, those with
impaired immune systems may experience greater benefits
from being acutely stressed before receiving a vaccination.
The exact mechanism of the relationship between acute stress
and the vaccination-induced immune response is still under
investigation. However, it is believed that stress-induced
redistribution of certain subclasses of lymphocytes is responsi-
ble for the enhanced antibody response.
Behavioral Interventions
By understanding the mechanisms of disease, scientists are
better able to intervene in causal pathways to prevent or slow
disease progression. Empirical evidence gained from studying
interactions between behavioral, psychosocial, and biological
factors has helped to shape and develop prevention and treat-
ment efforts. Traditionally, behavioral medicine focused on
biofeedback models for pain disorders, mood disorders, etc.
Biofeedback is a modality of treatment that monitors physio-
logical characteristics, and relays this information back to the
patient in real time. Patients learn to control and modify
316 Behavioral Medicine
the physiological responses being measured. A current example
of biofeedback is a pedometer. A pedometer indicates the
number of steps taken, and patients can monitor and use this
information to increase their physical activity. In particular, the
field of health promotion, which is encompassed within
behavioral medicine, has pioneered the design, operationaliza-
tion, and integration of many behavioral interventions into
medical and community settings. A behavioral intervention
usually refers to one or a series of activities that focus on
reducing risk behaviors (e.g., smoking), and increasing and
strengthening protective factors (e.g., spending time with non-
smokers). The time points at which these activities occur are
based upon the knowledge of disease etiology as well as the
theories of behavior change. In essence, behavioral interven-
tions are concerned with achieving behavioral changes that
maximize health and improve quality of life. Behavioral inter-
ventions have been classified using two distinct approaches.
The first approach classifies interventions based on the point at
which they occur along the health continuum: prepathogen-
esis, early stages of pathogenesis, or severe pathogenesis. The
second classifies interventions based on the level of the social
ecology that they target: interpersonal, community, organiza-
tional, or societal.
Primary, Secondary, and Tertiary Prevention Efforts
This intervention approach conceptualizes the progression of
disease on a continuum where at one end is good health and
other is death. Interventions can be conducted at any point
along this continuum. Primary prevention refers to interven-
tions that forestall the onset of illness or injury during the
prepathogenesis period. These interventions are focused on
reducing risk behaviors among individuals who do not have
the specified disease or disorder. For example, primary preven-
tion efforts include interventions to promote safe and frequent
condom use among HIV-negative individuals. Secondary pre-
vention refers to efforts that occur after the individual has been
diagnosed with a condition. In this case, secondary prevention
might include efforts to promote condom use among partners
who are both HIV positive, but have different strains of
the virus. Condom use would then reduce the chances of
coinfection by another strain of HIV. Tertiary prevention
efforts are aimed at individuals who are already severely
affected by a disease (e.g., impairment or disability). In the
case of HIV, tertiary prevention efforts target people with pro-
gressive AIDS or HIV infection, and caretakers and health care
professionals involved with the ill person. These efforts are
concerned with preventing as many of the disabling aspects
of AIDS as possible.
Levels of Intervention
Historically, behavioral interventions have been designed
based on individual-level theories, with little regard for the
influence of contextual factors on behavior change. As our
understanding of pathways of health and illness has evolved,
it has become evident that individual behaviors cannot be
treated without considering the context in which they occur.
Thus, there has been a movement to integrate individual-level
theories into the larger social ecology. It is important to note
that there are over 65 theories of behavior change (which will
not be discussed here); however, we reference ecological mod-
els because behavioral interventions are often classified accord-
ing to the particular stratum of the social ecology that
they target.
Levels of intervention can be defined as individual, com-
munity, organizational, and societal. However, some sources
further divide these levels into sublevels. The societal level is
the broadest, and it usually refers to nation-wide prevention
efforts. For example, the United States has produced several
social media campaigns to increase awareness and change
social norms about many behaviors, including smoking and
seatbelt use. Intervention strategies included billboards, televi-
sion advertisements, as well as policies that increase taxes on
tobacco products and financial penalties for not wearing a seat
belt. Interventions conducted at a societal level often have
simple messages that are relevant to the majority of the target
audience. Behavioral interventions conducted at the organiza-
tional and community levels are targeted at a specific subset
of the population who hold membership to a particular orga-
nization or community. These interventions might target
members of schools, neighborhoods, or even a particular racial
or ethnic group. With regard to the latter, within the past three
decades there have been increased efforts to reduce health
disparities. One way that public health and behavioral medi-
cine has responded has been to create and deliver culturally
relevant behavioral interventions targeted toward members of
a specific racial and ethnic community. Finally, interventions
focused on the individual are characterized by high levels
of interpersonal interactions between the targets of the
interventions and the health providers. Psychotherapy or HIV
case management programs are examples of individual-level
interventions.
Challenges in Behavioral Medicine
In spite of great developments, the field of behavioral medicine
has also encountered challenges. Behavior change can be slow
and difficult to sustain. Many times individual-level interven-
tions are not successful because there are social or environmen-
tal influences that impede the adoption of new behaviors. For
example, there are inconsistent research findings on the effects
of psychological therapies (e.g., cognitive behavioral therapy,
mediation and relaxation therapies) to modify immune path-
ways and disease outcomes. The WHO and leading national
health authorities estimate that the prevalence of depression
is rapidly increasing worldwide. Depression has also been
strongly associated with many chronic diseases. While there
are several behavioral therapies to reduce depressive symp-
toms, the prevalence continues to rise. Newer theoretical
models of behavioral medicine are guided by socioecological
frameworks, which intervene at multiple levels of the patient's
social ecology (e.g., intrapersonal, interpersonal, community,
national laws). For example, environmental interventions
such as increased parks, sidewalks, and bike lanes are promising,
and have been associated with increased physical activity and
reductions in weight.
Behavioral Medicine 317
Innovations in Behavioral Medicine
Evidence-Based Practice
Within the past decade there has been a movement to integrate
individual care with the best available practices, which are
determined by a systematic review of the literature. Histori-
cally, it has been challenging to compare across behavioral
interventions due to differences in measurement, sampling,
and theoretical fidelity. Thus, an evidence-based approach
calls for standardized study methods and evaluation in order
to compare findings across studies. Proponents of an evidence-
based approach believe it helps clinicians and health practi-
tioners maximize effective intervention strategies as well as
reduce less effective strategies, and ultimately, improve clinical
outcomes. Those against evidence-based approaches believe
that researchers are uninformed and out of touch with the
realities of clinical practice, and that stringent evaluation pro-
tocols adversely influence the ability of physicians to make
unimpeded clinical decisions.
Nevertheless, the movement for evidence-based practices
continues to grow. For example, the CDC developed a com-
pendium of best-evidence HIV behavioral interventions. The
CDC states that these interventions have been 'rigorously' eval-
uated and have shown significant effects in eliminating or
reducing HIV risk behaviors. Some US government funding
agencies and programs have started to only fund clinics and
community organizations that implement interventions that
have been evaluated and shown consistently, in literature, to
reduce risk behaviors and increase positive health outcomes.
Measurement Technology
Behavioral medicine has always been interested in obtaining
physiological and biochemical measurements. Such measure-
ments typically include cardiovascular measures such as blood
pressure, heart rate, stroke volume, and peripheral resistance,
as well as biochemical measures of the autonomic nervous
system, hormones, and immune system peptides that are
found in blood and/or urine. More recently, the validated use
of saliva for the assessment of certain neurohormones and
inflammatory products has greatly expanded the ability of
behavioral researchers to obtain biochemical measurements
in field research. Ambulatory measures (e.g., 24 h ambulatory
blood pressure monitoring and sleep actigraphy monitors)
have enabled researchers to look outside their laboratories
and monitor research participants for extended time periods
in their home environments. Newer technological advances
include the use of dried bloodspot (DBS) samples for the
collection of biochemical markers. This latter relatively nonin-
vasive methodology, which involves collecting drops of whole
blood on filter paper (typically from a finger prick), supports
the ability to assay a wide range of circulating biomarkers.
This methodology is still being refined, and protocols and
assays for many biomarkers are not as yet as reliable as what
can be determined from larger blood volumes obtained from
venipuncture.
Within behavioral interventions, there has been increased
integration of objective measures in addition to traditional
self-report instruments. For example, studies of children's
physical activity have collected self-report measures as well as
research assistant's observed playground activity or counted
number of toys that a child uses to be physically active (e.g.,
bike, trampoline, etc.). Moreover, there has been an increased
use of accelerometers, a device that participants can wear that
continuously records time and intensity of physical activity.
With regard to delivery of behavioral interventions, there con-
tinues to be increased use of web-based interventions as well as
text messaging. For example, some programs remind and
encourage participants to be physically active or take their
medication via text message. More recently researchers are
working on behavioral interventions that promote physical
activity using interactive video games, such as Nintendo's Wii.
Conclusions
As disease patterns have shifted, so has the emphasis on under-
standing the influence of lifestyle behaviors on health out-
comes. Currently, cardiovascular conditions and cancers are
the leading causes of death in many countries. Modifiable
behaviors such as exposure to tobacco products, high-fat
diets, and physical inactivity are responsible for high rates of
morbidity and mortality associated with many chronic condi-
tions. Chronic diseases have complex etiologies and are inter-
dependent upon lifestyle behaviors. Understanding how these
behaviors work in concert with biological and psychosocial
factors to protect against, or exacerbate, disease requires a
multifaceted approach. Behavioral medicine is an interdisci-
plinary field that studies the interplay among psychosocial,
biological, and behavior pathways on the natural progression
of disease. Perhaps what sets behavioral medicine apart from
other related fields is that in addition to studying the mechan-
isms of disease, it focuses on transferring this knowledge to
behavioral interventions that can be delivered in medical and
community settings. Recent advancements in the field center
on studying mechanisms of health disparities, developing
behavioral interventions to address these gaps, refinement of
measurement techniques, use of web-based and video games
to deliver behavioral interventions as well as systematic and
standardized evaluation to determine best practices. Overall,
the field of behavioral medicine is concerned with advance-
ments that promote improvements in clinical outcomes, and
quality of life.
See also: Biofeedback Therapy; Catecholamines and Behavior; Central
Nervous System; Cultural Psychology; Depression; Drugs, the Brain,
and Behavior; Hormones and Behavior; Media Influence on Behavior;
Operant Conditioning; Positive Psychology; Psychological Predictors
of Heart Disease; Psychotherapy; Social Support.
Further Reading
Adler R (ed.) (2006) In: Psychoneuroimmunology, 4th edn., vols. 1-2. New York:
Elsevier Academic Press.
Adler NE and Rehkopf DH (2008) U.S. disparities in health: Descriptions, causes, and
mechanisms. Annual Review of Public Health 29: 235-252.
Cohen S, Underwood LG, and Gottlieb BH (eds.) (2000) Social Support Measurement
and Intervention: A Guide tor Health and Social Scientists. New York: Oxford
University Press.
318 Behavioral Medicine
Edwards KM and Mills PJ (2009) Interactions among the endocrine, central nervous
and immune systems. In: Von Kanel R and Ehlert U (eds.) Psychoendocrinology &
Psychoimmunology.
Friedman HS and Adler NE (2007) The history and background of health psychology.
In: Friedman HS and Silver RC (eds.) Foundations of Health Psychology, pp. 3-18.
New York: Oxford University Press, Inc.
Gunnar M and Quevedo K (2006) The neurobiology of stress and development. Annual
Review of Psychology 58: 145-173.
Leventhal H, Weinman J, Leventhal EA, and Phillips LA (2008) Health psychology: The
search for pathways between behavior and health. Annual Review of Psychology
59: 477-505.
Mackay J and Ericksen M (eds.) (2002) The Tobacco Atlas. Geneva: World Health
Organization.
Miller GE, Chen E, and Cole SW (2009) Health psychology: Developing biologically
plausible models linking the social world and physical health. Annual Review of
Psychology 60: 501-524.
Mokdad AH, Marks JS, Stroup DF, and Gerberding JL (2005) Correction: Actual causes
of death in the United States, 2000. Journal of the American Medical Association
293: 293-294.
National Institute of Diabetes and Digestive and Kidney Diseases (2008) National
Diabetes Statistics, 2007 Fact Sheet. Bethesda, MD: US Department of Health and
Human Services, National Institutes of Health.
Ogden CL, Carroll MD, McDowell MA, and Flegal KM (2007) Obesity Among Adults in
the United States - No Change Since 2003-2004. Hyattsville, MD: National Center
for Health Statistics; NCHS data brief no. 1 .
Taylor SE (2011) Health Psychology, 8th edn. New York: McGraw Hill.
Wild S, Roglic G, Green A, Sicree R, and King H (2004) Global prevalence of
diabetes: Estimates for the year 2000 and projections for 2030. Diabetes Care
IT. 1047-1053.
Winickoff JP, Friebely J, Tanski SE, et al. (2009) Beliefs about the health effects of
‘thirdhand’ smoke and home smoking bans. Pediatrics 123: e74— e79.
World Health Organization (2008) Global Burden of Disease: 2004 Update. Geneva:
World Health Organization.
Relevant Websites
www.psychosomatic.org - American Psychosomatic Society.
www.cdc.gov - Centers for Disease Control and Prevention.
www.isbm.info - International Society of Behavioral Medicine.
http://pmbcii.psy.cmu.edu - Pittsburg Mind and Body Center.
www.pnirs.org/index.cfm - PychoNeurolmmunology Research Society.
www.sbm.org - Society of Behavioral Medicine.
Behavioral Pharmacology
W A McKim, Memorial University of Newfoundland, St. John’s, NL, Canada
© 2012 Elsevier Inc. All rights reserved.
This article is a revision of the previous edition article by William A. McKim, volume 1, pp. 365-374, © 1994, Elsevier Inc.
Glossary
Classical conditioning A phenomenon first studied by
Pavlov in which a previously neutral stimulus is paired with
a stimulus that elicits a reflexive response. Eventually the
neutral stimulus will elicit the reflexive behavior.
Discriminative stimulus A stimulus that signals that a
response will be followed by the presentation of a reinforcer.
Operant behavior Behavior that is maintained by its past
consequences rather than elicited or evoked by a stimulus or
antecedent condition.
Physical dependence The state in which a withdrawal
symptom is observed upon the abrupt discontinuation of
chronic drug administration.
Reinforcing stimulus A stimulus that increases the
frequency of a response upon which its presentation
is contingent.
Schedule of reinforcement This term describes the
relationship between the delivery of a reinforcing stimulus
and operant response or responses it follows. The most
common ones are ratio schedules where the reinforcer is
delivered when a number of responses have been emitted,
and interval schedules where the reinforcer is delivered after
a response on the basis of the passage of time.
Skinner box A chamber of a type invented by B. F. Skinner
used to study operant behavior in laboratory animals. The
chamber is isolated from external stimulation and contains
devices activated by responses of the animal which permit
delivery and recording of reinforcements.
Tolerance A lessening of the effect of a drug when the same
dose is repeatedly administered, or conversely, the necessity
of giving a higher dose with repeated administrations in
order to achieve the same effect.
Behavioral pharmacology may be defined in one of two ways.
The term was initially used by Peter Dews of Harvard Medical
School to refer to the analysis of drug effects using operant
analysis of behavior as developed by B. F. Skinner. More
recently the term is being used in a broader sense to refer to
the study of the effects of dmgs on behavior using any of the
experimental techniques of modem behaviorally oriented psy-
chology. The key word here is 'behavior,' since this expansion in
definition does not extend to include mentalistic concepts such
as emotion, motivation, or cognition unless they are defined in
terms of observable behavior. Before Dews suggested the term
'behavioral pharmacology' in the 1950s, research in this area
was commonly referred to as 'psychopharmacology' - a combi-
nation of the fields of psychology and pharmacology. 'Psycho-
pharmacology' is still used, but is generally taken to include
research involving any aspect of psychology, including the men-
talistic, and is used particularly to refer to the study of drugs in
the treatment of mental illness or psychological disorders.
It is important to note that there may be several aims of
research in behavioral pharmacology. As Travis Thompson
and Charles Schuster, the authors of the first text in the field,
put it, 'The behavioral pharmacologist is not only interested in
observing behavioral changes produced by dmgs, but analyzing
the mechanisms of the drug's effect.' In addition, Peter Dews has
pointed out that not only are behavioral techniques useful in
helping analyze the behavioral mechanisms through which dmgs
alter behavior, but just as often, dmgs may be used as tools to
help untangle complex problems in the analysis of behavior.
History of Behavioral Pharmacology
Humans have always been curious about substances that alter
the mind. Alcohol, cannabis, and opium have been used for
millennia. Aristotle commented on the effects of alcohol, and
the analgesic effects of opium were noted by the very earliest
Greek and Egyptian physicians. Up until the beginning of the
twentieth century, however, investigations of these substances
involved only verbal descriptions of the effects of the drug on
the subjective experience, more often with literary rather than
scientific intent. Such descriptions as DeQuincey's Confessions
of an English Opium Eater and Gautier's La Club des Hachichins
were fascinating descriptions of the dmg experience, but were
of limited value to science.
Systematic studies of the effects of these substances had to
await two developments. First, there was the development of
modern chemical techniques in the nineteenth century that
permitted the isolation of dmgs from natural substances, and
the synthesis of substances that do not occur naturally.
The second important development was a precise, systematic
and replicable means of describing recording, and analyzing
behavior. Three such systems were developed. In Russia, the
Nobel Prize-winning physiologist Ivan P. Pavlov discovered
and did extensive research on the conditioning of reflexes,
what we now call respondent behavior. The conditioning of
respondent behaviors is often referred to as classical conditioning.
In his laboratory in St. Petersberg caffeine and other dmgs were
administered to dogs and their effects on acquisition and extinc-
tion of conditioned reflexes were observed. In addition, Pavlov
also successfully conditioned the effects of dmgs to previously
neutral stimuli. His work produced interesting results, but
Pavlov did not pursue this line of research further.
In North America, Thorndike and his followers pioneered
the study of learning in laboratory animals using repeated
single discrete trials in an apparatus such as a maze or a mnway
etc. This method became known as instrumental learning. In the
1920s, Macht acquired a maze from Thorndike and in a series
319
320 Behavioral Pharmacology
of studies investigated the effects of a number of different drugs
on rat behavior. Once again, this work did not immediately
stimulate much further investigation until later when people
started looking for methods to screen newly discovered, behav-
iorally active drugs such as chlorpromazine.
Also in North America, B. F. Skinner and his coworkers
developed the technique known as the operant analysis of
behavior. Skinner and his students developed a technology
whereby the behavior of an animal in an isolated environment
(Skinner box) could be manipulated by altering the schedule
of reinforcement, that is, by precisely controlling the contin-
gency between performance of a specific action such as press-
ing a lever, and the delivery of a reinforcer such as food. Using
these techniques, the rate and pattern of responding of experi-
mental animals, usually rats and pigeons, could be controlled
with a precision not previously possible, and remained stable
throughout lengthy experimental sessions and over repeated
sessions. Skinner and Heron recognized early that this tech-
nique could be particularly useful in studying the effects of
drugs and in 1937 they published a paper describing the effects
of DL-amphetamine and caffeine on the behavior of rats. Their
study was published in the psychological literature rather than
in a pharmacology journal and it appears not to have been
noticed by pharmacologists.
After Skinner and Heron, there were other attempts to
investigate the effects of various drugs using operant techni-
ques, but up until the mid-1950s such studies were sporadic
and uncoordinated.
At this time there was a growing interest among pharmacol-
ogists in drugs that affected the central nervous system; how-
ever, most pharmacologists were unaware of the techniques
developed by psychologists. Consequently, behavioral research
carried out by pharmacologists mostly involved observing lab-
oratory animals after they had been given a drug, and quantify-
ing the total amount of activity in an unstructured situation. If
the drug increased locomotor activity it was taken to indicate
that the drug was a central nervous system 'stimulant'; if activity
decreased, the drug was a 'depressant.'
Even though both psychologists and pharmacologists were
doing behavioral research with drugs, up until the 1950s there
could not be considered to be a separate discipline of behav-
ioral pharmacology. The impetus to develop such a field came
in the early 1950s and arose largely from two events. The first
was the tremendous therapeutic and commercial success of
chlorpromazine and other antipsychotic drugs, and the result-
ing need to find tests in laboratory animals that were useful in
screening drugs for potential therapeutic effects on behavior of
humans. The second was the compelling demonstration by
Peter B. Dews of the usefulness of Skinner's operant techniques
to study drug effects.
Chlorpromazine
Chlorpromazine was marketed by the French pharmaceutical
company Rhone-Poulenc as an antipsychotic drug in 1952.
Initially it was used to prevent surgical shock, but it was
soon discovered that it was also effective in tteating the spe-
cific symptoms of schizophrenia and other psychoses. The
success of chlorpromazine demonstrated the tremendous eco-
nomic potential of behaviorally active drugs, and an intensive
search for new drugs and new medical applications of older
drugs began.
In the development of chlorpromazine, behavioral techni-
ques had been used to confirm that chlorpromazine and other
phenothiazines had antipsychotic properties. In fact, one of
the techniques used to identify the important behavioral prop-
erties of the phenothiazine antipsychotics was a rope-climbing
task developed by Macht in the tradition of instrumental
learning. It became obvious that there was a need for better
understanding of how drugs altered behavior, and that the
synthesis of pharmacology and behavioral techniques would
be useful, not only to pharmaceutical companies wanting to
develop and test new compounds, but as a separate field of
investigation that could lead to a better understanding of the
interaction of drugs and behavioral processes.
Operant Analysis of Behavior
In 1955, Peter Dews began publishing a series of papers in
pharmacology journals in which he reported using Skinner's
operant technology to demonstrate some remarkable and
surprising drug effects. Dews was ttained as a physician in
England and came to the United States where he worked for
Burroughs Wellcome Co. and later earned a Ph.D. in physiol-
ogy from the University of Minnesota. In 1952, he took a
position at Harvard Medical School. Earlier in his career he
had attempted to study the effects of THCs on behavior (THC
is one of the active ingredients in cannabis), but had given up in
frustration because '. . . there were no promising techniques in
pharmacology for studying behavioral effects.' He went on to
do experiments on the effects of histamines and antihistamines
on isolated smooth muscles of guinea pigs and blood pressure
of cats. These techniques involved the continuous recording of
physiological responses in real time on a polygraph.
When he arrived at Harvard, Dews went across the river to
Cambridge ". . . to see B. F. Skinner who introduced me to C. B.
Ferster who in turn took me into their laboratory. And there
they were recording something they called operant behavior as
ups and downs of a continuous line in real time. I was at home
at once and was determined to follow the lead of Skinner and
Heron 15 years before and to apply the techniques to behav-
ioral pharmacology. "
Of particular appeal to Dews was the fact that operant
techniques recorded a steady-state behavior for an extended
period of time during which the effects of a pharmacological
treatment could be continuously monitored and recorded.
Initially comparisons between drug and control conditions
during a single session were made for each individual animal.
As the techniques developed, comparisons were made between
drug and control conditions between different sessions in indi-
vidual animals.
It has been pointed out by M. Jackson Marr that the reason
that Dews felt so comfortable with operant technology was its
similarity to the methodology he was using in his physiological
investigations. This methodology had been developed many
years earlier by Claude Bernard, the French physiologist of the
nineteenth century who is considered to be the founder of
modern physiology and the field of experiential medicine.
Bernard studied the function of individual organs by continu-
ously recording their response to the exposure and then the
Behavioral Pharmacology 321
removal of various environmental and chemical changes such
as temperature and drugs.
Dews was soon to demonstrate how useful operant meth-
odology could be when he published a series of papers that are
now considered to be the seminal works of the field that was to
become Behavioral Pharmacology. Figure 1 is from Dews'
1955 paper, which shows the effect of different doses of pen-
tobarbital on the rate of key pecking of pigeons responding for
food on two different schedules of reinforcement. Pentobarbital
had a distinctly different effect depending on the schedule in
effect when the dmg was administered. At doses of 1. 0-2.0 mg,
the dmg increased response rates if the pigeon was responding
on a fixed ratio 50 schedule (FR 50 - every 50th response was
followed by an exposure to food), but decreased responding if a
15-min fixed interval was in effect (FI 15 min - the first response
after 15 min had elapsed was followed by an exposure to food).
Dews showed that pentobarbital altered the behavior of the
pigeon in a different manner depending on the schedule of
reinforcement in effect at the time the dmg was given. This
showed convincingly that dmg effects depend on the type of
behavior that is occurring rather than simply 'depressing' or
'stimulating' all behavior. This paper and several more like it
were published in pharmacology journals. Operant techniques
finally captured the attention of pharmacologists and became
the method of choice for studying the effects of dmgs on
behavior. Roy Pickens has pointed out that prior to 1955,
83% of published studies involving animal testing of dmgs
involved instmmental techniques and only 10% involved
operant techniques, but between 1955 and 1963, 42% of all
dmg studies used instmmental procedures and the number
studies using operant methodology had increased fivefold to
52%. The use of respondent behavior remained at 6-7%.
In addition to the work of Dews, in the mid- and late-
1950s, Joseph V. Brady established the first university-related
Figure 1 The log dose-effect curves for pentobarbital on the relative
response rate of pigeons responding on FI 15-min (open circles) and
FR 50 schedules (closed circles). The dotted lines indicate response
levels after a control injection of saline. Reproduced from Dews PB
(1955) Studies on behavior: 1. Differential sensitivity to pentobarbital of
pecking performance in a pigeon depending on the schedule of reward.
Journal of Pharmacology and Experimental Therapeutics
113(4): 393-401, with permission from Aspet.
laboratory using operant technology for the study of drug-
behavior interactions at Walter Reed Army Medical Center
and later at the University of Maryland. Brady stimulated the
development of behavioral pharmacology by training many
young researchers and urging the pharmaceutical industry
and the federal government of the United States to support
this new field.
In 1956, a conference was held on 'The Techniques for the
Study of the Behavioral Effects on Dmgs' sponsored by the
New York Academy of Science. It was chaired by Dews and
Skinner and the participants included D. S. Blough, Richard
Herrnstein, H. F. Hunt, Neil Miller, William Morse, Murray
Sidman, and many others who had made significant contribu-
tions to the field at that time. Also in 1956, Skinner made a
formal call for the development of a new science of behavioral
pharmacology. The new science got its formal start when the
Behavioral Pharmacology Society was founded. In about 1955,
a group interested in pharmacology started having informal
evening dinner meetings during the annual meeting of the
Eastern Psychological Association. The Society evolved a year
or two later.
Behavioral Pharmacology in Europe
Early interest in the behavioral analysis of dmg effects was not
confined to North America. In Britain, interest in the field was
stimulated by a pioneering symposium held in London spon-
sored by the Ciba Foundation in 1963. It was attended by
many prominent European researchers of the time as well as
those from North America, including Len Cook and Peter
Dews. The proceedings were edited by Hannah Steinberg
and published in 1964. Since that time, researchers such as
D. E. Blackman, David Sanger, Susan Iverson, Trevor Robbins,
and Ian Stolerman and their students have had an extensive
impact on the field.
The operant tradition of behavioral analysis of dmg effects
also spread to the continent. Marc Richelle, after spending a
year at Harvard with B. F. Skinner returned to Belgium in 1959
where he set up a behavioral pharmacology laboratory and
began publishing in European journals. Other prominent
Belgian behavioral pharmacologists include C. Niemegeers
and F. Colpaert. Other European pioneers in the field include
R. Dantzer in France, J. Slangen in Holland, and Daniel Bovet,
G. L. Gatti, and Georgio Bignami in Italy. The continued expan-
sion of behavioral pharmacology in Europe was marked by the
founding of The European Behavioural Pharmacology Society
in 1986. In 1989, the journal Behavioural Pharmacology was
started and the official journal of the European Behavioural
Pharmacology Society.
Graduate Training in Behavioral Pharmacology
Until the 1990s, most researchers in behavioral pharmacol-
ogy came from one of two backgrounds; either they had
been trained as psychologists and acquired an understand-
ing of pharmacology, or they had been trained as pharma-
cologists and taught themselves the techniques of behavioral
analysis. For several decades now, a number of graduate
programs have been developed which combine training in
both traditions.
322 Behavioral Pharmacology
Environmental Influences on Drug Effects
As James Barrett has put it "... a drug is not simply a molecule
with static unitary behavioral effects. The same dose of a drug
can exert an array of effects on behavior depending on the
immediate consequences of responding, the prior experience
of the organism, and the total environmental context in which
the behavior takes place."
In the years since Dews's early studies, many behavioral
pharmacologists have explored the complex interaction
between environmental factors and the effect of various
drugs. In addition to schedule of reinforcement, these factors
include rate of responding, extent of stimulus control, the
nature of the reinforcer, and the behavioral and pharmacolog-
ical history of the organism.
Schedule of Reinforcement
Dews's 1955 study showed that the effects of a barbiturate
depend on the schedule of reinforcement. It has been demon-
strated that each class of drugs has a unique pattern of effects
on various schedules. Drugs such as cocaine and amphet-
amine, for example, have effects quite different from barbitu-
rates. When various doses of amphetamine are administered to
animals responding on an FI and FR schedule similar to those
used in the Dews experiment, a dose range is encountered
where FI responding is increased considerably above control
levels and FR responding is nearly eliminated - exactly the
opposite of the barbiturates.
Rate of Responding
The rate at which an operant response is emitted has traditionally
been the primary dependent variable analyzed by operant
psychologists, and early in the development of the field it
was discovered that response rate had a pervasive influence on
the effect of many drugs. In one of his early experiments with
amphetamine, Dews noticed that the drug seemed to be interact-
ing specifically with the rate of responding. He noticed that the
drug increased low rates of responding, while at the same time it
decreased responding in situations that normally engendered
high rates. This phenomenon has become known as the rate
dependency effect. It appears to be responsible for the observa-
tion described earlier that responding on FI (which normally
engenders comparatively low rates of responding) is increased,
and responding on FR (which normally engenders high rates) is
slowed by amphetamine. In fact, it is likely that drugs often affect
different schedules differently because different schedules pro-
duce different base rates of responding.
Rate-dependent effects are caused by many different drugs in
many different species responding on a wide variety of tasks. At
one point, in fact, rate-dependent effects were thought to be so
universal that many researchers were more interested in excep-
tions to rate-dependent effects than in rate effects themselves.
Other Factors
Many other environmental influences have been shown
to interact dynamically with drug effects, but the pervasiveness
of rate has made it necessary to eliminate the influence of
rate before the role of these other factors can be demonstrated.
For example, different reinforcing events must be presented
in such a manner that they engender identical rates and patterns
of behavior before it is possible to show that in some cases a
drug will differentially alter behavior controlled by different
reinforcers. James Barrett trained monkeys to respond on an FI
5-min schedule. The reinforcing event was either food or a brief
electric shock. (Shock can serve as a reinforcer if the organism
has a suitable history of responding.) Responding on both
schedules was identical in terms of rate and pattern, but the
effect of several drugs was different depending on whether the
reinforcing event was food or shock. Pentobarbital, for example,
slowed the food reinforced performance, but created a consider-
able increase in the shock reinforced behavior.
Drugs as Discriminative Stimuli
A large and productive branch of contemporary behavioral phar-
macology deals with drugs as discriminative stimuli. Investiga-
tions into the discriminative properties of drugs originated with
research in the instrumental learning tradition and the early
neurophysiological theories of Donald Hebb at McGill University.
For years there had been anecdotal accounts that suggested
that events experienced in a drugged state might not have the
ability to control behavior when the organism was undrugged
and vice versa. This phenomenon is called dissociation or state-
dependent learning.
While investigating dissociation, Donald Overton, then a
graduate student at McGill University's Psychology Department,
did a series of experiments with rats using an instrumental
learning task; shock avoidance in a T-maze. Overton was easily
able to demonstrate that rats that learned to avoid the shock in
the maze when drugged with pentobarbital were unable to
avoid the shock later when given a placebo and vice versa.
To explore the extent of dissociation, Overton wanted to
determine whether information learned on drug days (turn
right) would interfere with information acquired on saline
days (turn left) and vice versa. On alternate days, the rats
were administered pentobarbital and one arm led directly to
safety. On other days, they were given a saline placebo and the
other arm was safe. Overton discovered that the rats very
quickly learned to make the appropriate response depending
on whether they were drugged or not. In other words, he
showed that the drug administration was acting as a discrimi-
native stimulus which controlled the direction the rat would
turn at the choice point of the maze on the first trial each day.
He also found that rats would learn to make the appropriate
response at doses much lower than those required to cause
complete dissociation.
Since Overton's early experiments, research on the discrimi-
native stimulus properties of drugs has rapidly expanded, and
the electrified T-maze has been replaced with the Skinner box.
Herbert Barry III, then at Yale, is generally credited with applying
operant techniques to the field. Barry trained experimental ani-
mals to press a lever for food in a two-lever Skinner box. On days
when a drug is administered, presses on one lever are reinforced,
and on saline days, the other lever is reinforced. Animals rapidly
leam to respond to the appropriate lever depending on whether
they had been injected with a drug or saline.
Behavioral Pharmacology 323
Using these techniques, it has been demonstrated that most
centrally acting drugs have discriminative stimulus properties,
although some classes of drugs such as the barbiturates appear
to be more easily discriminable than other classes. Such drugs
can acquire discriminative control at least as rapidly and, in
some cases, more rapidly than more conventional stimuli like
noises and lights. It has also been shown that as well as dis-
criminating between a drug and saline, laboratory animals can
discriminate between different doses of the same drug and
between different drugs.
In addition to determining whether a drug can act as a
discriminative stimulus, behavioral pharmacologists have also
been able to test for generalization between drugs. In this situa-
tion, an animal is trained to discriminate between saline and a
drug and then is tested with a different drug, a technique similar
to generalization. Animals will usually generalize responses
between drugs of the same pharmacological class, thus making
it a valuable in drug screening and drug development.
The drug discrimination paradigm has been successfully
used to investigate many aspects of drug action including
the temporal and kinetic features of drug action, reversibility
of receptor binding, stereo-specificity and other structure-
activity relationships, agonist-antagonist interactions, and
receptor sensitivity changes, to mention a few. It continues to
be applied to new and interesting problems including investi-
gations of genetic manipulations on responses to drugs.
In 1978, The Society for the Stimulus Properties of Drugs
(SPDD) was founded by Donald Overton, Herbert Barry III,
and John Roseceans. In 2008, a special edition of the journal
Psychopharmacology was issued to mark its 30th anniversary.
Ian Stolerman and others have maintained a drug discrimi-
nation database online for many years. It contains a searchable
record of all drug discrimination studies. The project is now
supported by NIDA in the United States and in 2009 was
combined with a similar database for drug self-administration
studies.
Drugs as Reinforcing Stimuli
Research with Nonhumans
Early research was generally not able to demonstrate that labo-
ratory animals would administer drugs to themselves in the
same manner humans do. Most of these studies involved
the oral administration of drugs, usually alcohol, and even
though some studies showed that animals could demonstrate
a preference for some drugs, they were seldom ever able to
show that animals would make themselves physically depen-
dent on a drug, that is, voluntarily consume enough drug that
withdrawal symptoms would develop if the drug were discon-
tinued. Physical dependence was believed to be the defining
characteristic of addiction at the time. Also, these studies were
not able to demonstrate the compulsiveness usually associated
with human drug-seeking behavior. Because of these failures
and other widespread beliefs about the nature of drug addic-
tion, it was generally believed that drug addiction was
restricted to humans.
In 1962, three groups of researchers announced at a
meeting of The Committee on Problems of Drug Dependence
in Ann Arbor, Michigan, that they had developed the now
classic technique which permits laboratory animals to admin-
ister drugs to themselves intravenously. The technique involves
implanting a cannula into a vein of a laboratory animal
and attaching the cannula to a motor-driven pump. The
pump is connected to programming apparatus so that it can
be operated by the animal when it depresses a lever. Thus,
the animal can administer an intravenous injection to itself.
Using this method, Travis Thompson and Charles Schuster
of the University of Maryland, and Gerald Deneau, Tomoji
Yanagita, and M. H. Seevers at the University of Michigan
showed that monkeys would self-administer morphine, and
James Weeks and James Collins demonstrated that rats would
self-administer morphine.
The importance of these experiments was that they demon-
strate quite clearly that the administration of a drug can be a
reinforcing event in the same manner as food to a hungry
animal or water to a thirsty one. They showed that there is
nothing special or different about drug taking and that the
general principles that have been learned about operant behav-
ior are applicable to drug self-administration. Furthermore, they
showed that drug use is not a result of a pathology or a disease
and is not independent of normal psychological processes.
One of the earliest important findings using this technique
was that physical dependence is not necessary for drug self-
administration. Travis Thompson and Roy Pickens demon-
strated that drugs such as cocaine would be self-administered
even though they do not normally induce physical depen-
dence. This made it clear that drugs are self-administered
because they are positive reinforcers, not negative reinforcers,
that is, they are not self-administered to delay or relieve
unpleasant withdrawal symptoms.
Research with Humans
A number of very successful attempts have been made to apply
operant learning techniques to the study of human drug tak-
ing. Pioneers in this field are Jack Mendlelson and Nancy
Mello. At The Massachusetts General Hospital and later at
McLean Hospital in Belmont, Massachusetts, they studied
drug self-administration of human volunteers living on a
research ward. Volunteers, all with previous drug experience,
could earn drugs by performing some activity such as riding an
exercise bicycle or pushing a button. Using these techniques, it
is possible to meet the scientific standards achieved previously
with laboratory animal subjects and also satisfy ethical and
safety standards required by governments and institutions.
Drugs studied in such situations include heroin, metha-
done, cocaine, diazepam, barbiturates, THC, nicotine, and
caffeine. These studies have also made it possible to concur-
rently observe the effects of self-administered drugs on vari-
ables such as health, mood, and human social behavior, and
provide a model to evaluate drug abuse treatment strategies.
Similarities Between Human and Nonhuman
Self-Administration
It has been shown that most drugs consumed by humans are
also self-administered by nonhumans. These include alcohol,
nicotine, caffeine, barbiturates, opiates, anxiolytic or benzodi-
azepine, amphetamines, and THC. It is also clear that the
324 Behavioral Pharmacology
patterns of behavior generated by different schedules of drug
reinforcement are similar in both humans and nonhumans to
those generated when more traditional reinforcers are used,
and that other phenomena such as extinction, discriminative
stimulus control, chaining, second-order schedules, etc., are
also observed.
Drug administration patterns over time appear to be very
similar for humans and nonhumans. For example, a human
volunteer drinking alcohol in a research ward in a hospital and
a rhesus monkey self-administering ethanol through a cannula
both show an erratic pattern of intake and abstinence with
occasional self-inflicted withdrawal symptoms: a pattern very
similar to that shown by alcoholics in the natural environ-
ment. Moreover, both humans and nonhumans administering
opiate drugs show a smooth gradual increase in intake from
day to day without severe daily fluctuations, and the cocaine
pattern in both humans and animals involves cycles of intake
and abstinence when the drug is freely available.
Factors that Affect Reinforcing Properties of Drugs
The drug self-administration studies have provided much use-
ful information into the nature of drug use and abuse. For
example, it provides a method for assessing which drugs are
the most reinforcing and consequently, the most likely to be
abused. Several methods have been used to answer this ques-
tion. In one, known as the progressive ratio schedule, the
schedule demand increases as the session progresses. For exam-
ple, a dmg may be available on an FR 25 schedule, meaning
that the animal must make 25 responses for an infusion. After
the first infusion, however, the schedule changes to an FR 50
and then to an FR 100 and so on. Eventually, the schedule
value will reach a point where responding will not continue.
This is known as the breaking point. Drugs that are highly
reinforcing would be expected to have high breaking points.
Experiments such as these generally show that cocaine and
amphetamine have higher breaking points than other drugs.
Analogous results have been found in choice experiments
where the animal must make a choice between two levers,
each of which delivers an infusion of a different drug or a
different dose of the same drug.
Apart from assessing the inherent abuse liability of a drug,
techniques such as the progressive ratio and choice have been
used to assess the effect of situational and pharmacological and
even genetic variables on reinforcing strength of drugs. For
example, the influence of such variables as the extent of previ-
ous exposure to the same or a different drug, the relief of
unpleasant symptoms such as withdrawal sickness, drug
dose, task demand, stress and the extent of other motivations
such as hunger have all been investigated.
Behavioral Economics
The use of operant procedures has facilitated the application of
the principles of behavioral economics to drug and abuse. For
example, the choice procedure has also been used to present
animals with a choice between a drug and some other rein-
forcer such as food. Generally, it has been shown that acquisi-
tion of drug self-administration and total consumption of a
drug will be reduced when other attractive reinforcers such as
sweetened water and social interaction are made simulta-
neously available. This observation may well explain why
drug abuse appears to thrive in environments where there is
unemployment, poverty, and boredom, that is, environments
where there are few other sources of reinforcement to compete
with drugs as reinforcers of behavior.
Experiments such as these have shown that many of the
rules of microeconomics apply to drug use both by humans and
nonhumans. For example, if the cost of a drug infusion increases
(the animal is required to make more responses to earn the
infusion), fewer and fewer infusions will be earned. This is
similar to the demand function described by economists where
consumption of a commodity decreases as the price increases.
In the early 1970s, people doing research on drug self-
administration started the International Study Group Investi-
gating Drugs as Reinforcers (ISGIDAR), which meets once a
year in conjunction with the College on Problems of Drug
Dependence (CPDD) in the United States and publishes a
newsletter. ISGIDAR also maintains a database of research of
self-administration studies which is now available online in
conjunction with the Drug Discrimination Database.
Conditioned Drug Effects
One area of research of modern behavioral pharmacology has
its origin in classical or respondent conditioning rather than
operant analysis of behavior. This involves the study of the
conditioning of drug effects. Many years ago, a student of
Pavlov's laboratory in Leningrad gave a dog an injection of
apomorphine and then sounded a tone when the dmg began
to have its effect of causing the dog to salivate. After a few
pairings of the tone and the dmg, the tone by itself caused
the dog to exhibit this effect of the drug, that is, it caused the
dog to salivate. In Pavlov's terminology, the dmg was the UCS
(unconditioned stimulus), the salivation was the UCR (uncon-
ditioned response), the tone was the CS (conditioned stimu-
lus), and the salivation produced by the tone was the CR
(conditioned response).
Since that time, there have been many demonstrations
of conditioned dmg effects. Shepard Siegel of McMaster
University in Hamilton has shown that in many cases, the CR
is in fact, opposite to the UCR. In other words, if the effect
of the dmg is in one direction, an increase in body tempera-
ture, for example, then the conditioned effect of the dmg will
be in the opposite direction, a decrease in body temperature.
Siegel and other researchers have shown that this conditioned
effect in the opposite direction, called a conditioned compensa-
tory effect, follows the principles of classical conditioning. Siegel
also demonstrated that in many cases this compensatory
response is responsible for the development of tolerance for
dmgs. If a stimulus reliably precedes the delivery of a dmg, the
conditioned compensatory response will occur in response to
the stimulus. Because this response is physiologically opposite
to the effect of the dmg, the normal dmg effect is diminished.
This lessening of the dmg effect after repeated adminisuations
is a type of tolerance. Siegel has shown that tolerance to many
effects of dmgs will disappear if the stimulus condition nor-
mally present when the dmg is taken is no longer present,
that is, there is no stimulus to evoke the conditioned
Behavioral Pharmacology 325
compensatory effect. This is an example of a type of tolerance
called behavioral tolerance to distinguish it from other types
of tolerance that arise from metabolic and physiological
mechanisms.
The reinforcing effects of a drug can also be conditioned to
environmental stimuli. Sights, sounds, and even places that
have reliably been associated with drug administration can
acquire reinforcing properties which can also control behavior
in the same manner as the drug. Animals will bar press for
lights that have been paired with a reinforcing drug, and will
actively seek out and return to places where they have experi-
enced the effect of the drug in the past, an effect called place
conditioning. Place conditioning techniques have been used to
investigate the reinforcing effects of many drugs.
See also: Classical Conditioning; Operant Conditioning.
McKim WA (2007) Drugs and Behavior: An Introduction to Behavioral Pharmacology,
6th edn. Englewood Cliffs, NJ: Prentice-Hall.
Special Issue on Behavioral Pharmacology (1991) Journal of the Experimental Analysis
of Behavior 56(2): 167-423.
Winger G, Woods JH, Galuiska CM, and Wade-Galuska T (2005) Behavioral
perspectives on the neuroscience of drug addiction. Journal of the Experimental
Analysis of Behavior 84(3): 667-681 .
Relevant Websites
http://www.behavioralpharmacologysociety.org/ - Behavioral Pharmacology Society
(BPS).
http://www.behavioralpharmacologysociety.org/ - European Behavioral Pharmacology
Society (EBPS).
http://www.isgidar.org/ - International Study Group for the Investigation of Drugs as
Reinforcers (ISGIDAR).
http://www.sspd.org.uk/ - Society for the Stimulus Property of Drugs (SSPD).
http://www.dd-database.org/content.php?code=HOME - The Comprehensive Drug
Self-administration and Discrimination Bibliographic Databases.
Further Reading
Barrett J, Thompson T, and Dews PB (1977) Advances in Behavioral Pharmacology,
vol. 7. New York: Academic Press.
The Behavior-Genetics of Intelligence
M A Runco, University of Georgia, Athens, GA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Adoption studies Comparisons of (a) biological mothers
and their children with (b) adoptive parents and their foster
children.
Dizygotic twins Also known as fraternal twins; they share
approximately 50% of their genetic makeup.
Heritability A statistical index of genetic contributions.
Monozygotic twins Also known as identical twins; they
share approximately 100% of their genetic makeup.
Nature A label for genetic, biological, nonenvironmental
contributions to development and behavior.
Nurture A label for environmental, experiential,
nongenetic contributions to development and
behavior.
Range of reaction It indicates an interaction of nature
(which determines a range of potentials) and nurture (which
determines how much genetic potential is fulfilled).
The nature and nurture of intelligence has been debated for
centuries. One of the more significant breakthroughs on
this topic resulted from the development and refinement of
methods in behavioral genetics. This ensured that objective
data would be brought to bear and that theories could go
beyond a simple either/or theory of 'nature or nurture' to the
more reasonable view that biology and environment both
play a role. This article briefly reviews the behavioral genetic
approach to intelligence and summarizes the key findings, as
well as the limitations and remaining issues. It has a historical
flavor which reflects the fact that methods in behavioral genet-
ics were developed before more sophisticated and accurate
chemical analyses of genes were available. Still, there is much
to learn from the behavioral genetics research on intelligence,
especially when findings agree with those obtained with the
other more recently developed methodologies.
The starting point must be the definition of 'intelligence.'
As is implied by the claim that behavioral genetics research
would allow objective data to be brought to bear on the topic,
some sort of operational definition and reliable assessment
of intelligence is required. Unfortunately, the assessment of
cognitive abilities is itself a hotly debated topic. Paradoxically,
the 'intelligence quotient,' or IQ, is both the most employed
and the most infamous of the standardized measures of
intelligence. The debate concerning its reliability, validity,
and utility has now extended through several decades. Arthur
Jensen, one of the most vocal and prolific advocates, holds that
the evidence in favor of reliable IQ testing is more than con-
vincing. Stephen Jay Gould, in one of the most informative and
least political of critiques, suggested that reliable testing is not
possible because the IQ assumes the existence of one general
intelligence, known as 'g.' Gould insisted that there is no
mathematical, psychological, or even theoretical justification
for 'g.' He concluded that 'g' is a statistical artifact, and in
particular, an artifact of subjective factor analyses. Many other
critics of IQ testing concur that IQ is only a 'hypothetical
construct' which has no inherent, objective validity.
There is little doubt that many of the tasks contained in
IQ tests can be biased by experience. Some tasks seem to draw
mostly on factual information, and thus individuals who
receive a good education or have intellectual resources readily
at hand are likely to have an advantage. Yet psychometricians
deny this is a bias. They argue that individuals who score
high on an IQ test have above average information processing
capabilities. These in turn allow the individual to perceive,
evaluate, and remember more and more accurate information
from their experiences, compared to other individuals in the
same population. In other words, the psychometric view is
that having experience is by itself not enough for intelligence.
The person needs the intellectual ability to benefit from expe-
rience as well. As a historical aside, this view parallels what
Donald Hebb said long ago about bright canines making better
use of the stimuli in their environments.
In another corner there are those who feel that IQ tests are
too strongly tied to the personality factors that can facilitate
or inhibit intellectual performance. Examples of this perspec-
tive are numerous, but two stand out. John Loehlin, Arthur
Lindsey, and J. Spuhler pointed to an inflation of ability result-
ing from an individual's tendency to actively seek cognitive
challenges. The second example is that of Sandra Scarr, who
gave a detailed description of how personality factors determine
whether an individual (a) maximizes intellectual opportunities
or (b) passively accepts whatever is available. To what degree the
relevant personality factors are genetically or environmentally
based is, once again, an ongoing debate, but it does appear that
IQ performance may be a function of both personality and
cognitive abilities.
Those in favor of IQ hold that, even if not a perfect measure
of cognitive capacity, IQ is still useful. Phillip Vernon wrote,
"Psychometry is quite entitled to use its own brand of opera-
tionalism, regardless of philosophical theories of scientific
method, provided it works." Certainly IQ tests offer a standar-
dized assessment. As a matter of fact, they have demonstrated a
high level of utility specifically in the area of behavioral genetics.
Behavioral Genetics Methods
The idea of calculating an accurate 'heritability' of intelligence
was advanced over 60 years ago. At that point the size of the
326
The Behavior-Genetics of Intelligence 327
human genome was being estimated and enormous variation
was acknowledged. One early estimate was that there are 2 23
alternative genomes based on the human chromosomal supply.
Variety is insured by mutation, recombination, and meiosis of
the genes; and since these can occur at different loci at different
times with different frequencies, a probabilistic index of herita-
bility was the goal. Two methods were used to that end.
In the first, identical or monozygotic (M Z) twins are
examined. MZ twins have a virtually identical genetic makeup,
and if they happen to be separated shortly after birth, any even-
tual differences between them can be attributed to environ-
mental discrepancies. The second behavioral genetic method
compares an adopted child and her or his biological parents
and the same child with his or her foster parents. Presumably,
any correlations between the former can be attributed to genetic
similarity - an implication of the Mendelian principle that the
closer the biological kinship, the higher the genetic similarity -
while any correlations between the latter can be attributed to the
environment. Typically, a regression coefficient is used in both
the within-family and the between-family studies.
In one study, using these methods Alice Mary Leahy
compared 194 adopted children to a large control group. The
two groups were matched in terms of age, sex, education, and
parental SES. The experimental group had been placed in their
foster homes before they were 6 months of age, and they had
lived there for at least 5 years. The controls were, of course,
raised by their own parents. Leahy found an average correla-
tion of 0.56 between the control's IQ scores and their parents'
educational level, parents' vocabulary, parents' IQ, and a gen-
eral index about the home atmosphere. The correlation for the
adopted children and their adoptive parents was 0.20. As a
check for selective placement, Leahy correlated the biological
mothers' education with the foster parents' occupation and
education. The results were high enough to show that the
placement agencies did in fact select the foster parents so as
to match the child's background. Consequently, the differences
between the two groups are on the conservative side, and
the correlational discrepancies between them are even more
impressive. Leahy concluded that the 'measurable environ-
ment' could only alter the IQ from three to five points in either
direction. She reported that the genetic component in IQ is far
more influential than the environmental impact.
Longitudinal research with adoptive children and their
biological and foster parents often allows collection of more
data about the samples. In one such investigation 180 children
born to unwed mothers were tested five times between 1949
and 1966. Notably, they were not typical, even for adopted
children. For example, their biological mothers had subnormal
IQ scores (an average between 83 and 86). Also, selective place-
ment was again evidenced by moderate correlations between
the biological mothers' average educational level and the
foster parents' ('midparent') average educational level. For
this research the children were first tested with the Kuhlmann
test and then with the Stanford-Binet. The correlation coeffi-
cient calculated between the children and their biological
mothers increased at each testing, peaking at 0.44. Correlations
with nonbiological mothers never exceeded 0.10. One caveat:
the average IQ of the children at 13.5 years was 107-120 points
above the mothers' average. With such low scores for the
mothers, 'regressions effects' are likely, with the children tend-
ing to score closer to the mean.
Sandra Scarr and Robert Weinberg reported similar findings
with adoptive families, but they too had atypical samples.
In particular, SES and IQ averages were one standard deviation
above the norm. This implies a restricted range, though Scarr
and Weinberg argued that the patterns of variables were 'suffi-
ciently similar' to more representative samples. They therefore
endorsed the generalization of the empirical findings. More
specifically, they felt that the 'relative magnitude' of the coeffi-
cients was accurate and valid. Using 150 adopted children and
122 matched controls they compared Wechsler adult intelli-
gence scale (WAIS) scores. The children were adopted quite
young, averaging only 2.6 months at the time of placement.
Analyses indicated that the parental educational level, family
income, and parental IQ were each highly correlated in the
biological parent-child analyses. The adoptive parent-child
correlations were very low. Family size was unrelated to the
child's IQ in the adoptive families, and negatively related
in the biological families. Birth rank was significantly related
to IQ in both families, with later born children having a
marked disadvantage. The correlation between the adoptive
fathers' IQ and the child's IQ averaged 0.15, and the correla-
tion between the mothers' and the child's IQ's averaged 0.04.
The correlations between the biological mothers' educational
level and the adoptive mothers' IQ averaged 0.20 and 0.10
with the adoptive fathers' IQ. Here again there is evidence for
selective placement.
Similar patterns and heritabilities are apparent in twin
studies. Alan Gottesman, for instance, described one study in
which MZ twins, reared apart, were found to have an average
Stanford-Binet difference of 24 points. A second study used
another measure of mental age, but still found an average
difference of 14 points on verbal tasks and an average differ-
ence of 10 points on nonverbal tasks.
Another analytic method uses additive variances. Here
within-groups variability on IQ tests is delineated. In par-
ticular, the variance attributed to genetic differences is added
to the variance due to environmental differences, and this
is added to the variance due to errors in measurement.
When this method was used variability attributable to genetic
factors was very close to 60% ! A similar analytic method
involves subtracting MZ correlations from dizygotic (DZ)
twin correlations, with the difference attributed to the envi-
ronment. As expected, given the heritabilities already cited,
the results show genetic influence to be greater than the
environmental ones.
L. Erlenmeyer-Kimling and Lissy Jarvik summarized all of
the research available when the behavioral genetic approach
was the most sophisticated methodology available. Their sum-
mary, which was according to Robert Plomin and Stephen
Petrill one of the most influential publications on the behav-
ioral genetics of intelligence, clearly supported the genetic
view. In particular:
In 14 studies with MZ twins reared together, when the
genetic expectation was 1.00 (MZ twins sharing virtually all
genes), the median correlation was 0.87. In the four studies
when MZ twins were reared apart, that coefficient dropped
only to 0.75.
328 The Behavior-Genetics of Intelligence
In 1 1 studies with DZ like-sex twins reared together, the
genetic expectation being 0.50, the coefficient was 0.56. When
unlike DZ twins were reared together it was 0.49.
In 36 studies with siblings (not twins) reared together, the
genetic expectation again being 0.50, the median coefficient
was 0.55. In three studies with siblings (nontwins) reared
apart, this dropped to 0.47.
In 12 studies with biological parents and their children, the
genetic expectation again being 0.50, the median coefficient
was in fact 0.50. This is in direct contrast to the three studies
with foster parents and their children, the genetic expectation
being 0.00, the median correlation being 0.20. In five studies
with unrelated children raised together, and thus a 0.00 genetic
expectation, the median correlation was 0.24. Finally, in four
studies with unrelated children reared apart, the median corre-
lation was — 0.01.
The median correlations were remarkably close to the
expected values in almost every case. Note especially the MZ
twins reared apart in contrast to other siblings reared apart and
unrelated children reared apart. These medians are even more
impressive considering that the data came from a diverse set
of investigations. Certainly experimenter effects can be disre-
garded. Importantly, much more recently Plomin and Petrill
reviewed the available research and again concluded that
the highest correlations of IQ are for MZ twins reared together,
and then MZ twins reared apart. All other twins, siblings, and
intrafamily correlations are lower, even when individuals were
reared together or shared the home.
Cross-cultural studies tended to result in the same correla-
tional patterns. That is, the environmental impact was
detectable, but small in comparison to the genetic influence.
Findings such as these led to conclusions about a high herita-
bility of IQ. Yet methodological weaknesses in all behavioral
genetic studies - not just the cross-cultural investigations -
bring this into question. Clearly the measurement concerns
outlined earlier, especially those focused on the IQ, are
relevant to any correlation calculated from parent-child or
intratwin IQs.
The behavioral genetic approach itself has notable limita-
tions. Adoption studies, for example, tend to be restricted. When
there is a restricted range of scores, there is not as much varia-
bility and correlations are attenuated. Additionally, as noted
above, restricted samples severely limit generalizations. If the
sample is restricted, generalizations about heritability to other
samples or a population are not warranted. Certainly the chil-
dren in adoption studies were not typical: by definition adopted
children will have unusual emotional and physical personal
histories, histories that are different from those of nonadopted
children. Their backgrounds are simply atypical, even if entirely
healthy. Also, the placement of the children is far from random.
Adoption agencies use screening techniques, checking the
prospective foster homes for income, family size, and the like.
All of these are pertinent and very possibly could exaggerate
intrafamilial correlations. A last concern along these lines is
that the biological parents are often unknown, or their mental
ages only estimated from educational level or from equally
noncomparable indices.
A major concern in the twin studies is that the home atmo-
sphere always holds some similarities. There are constants
across all homes and all families. Such similarity will probably
contribute to correlations. Finally, assortative mating is com-
mon. Surely no one chooses a mate at random! The parents
probably share some of the same child-rearing ideals and
attitudes, and thus intrafamial correlations would be increased
relative to interfamlial correlations.
Conclusions
Behavioral genetics research often leads to high heritability
estimates for IQ, but the validity of that conclusion is often
questioned because of methodological concerns, including
the meaning of the term 'IQ'. Additionally, the heritability
reported in behavioral genetics research probably has low
generalizability. Urie Bronfenbrenner was correct when he
described heritability as contextual: it is applicable to one
trait, for one population, at one point in time. And very impor-
tantly, high heritability does not exclude environmental
effects. On the contrary, both genetic givens and environmen-
tal inputs are required for human growth of any kind, but
either or both may cause differences between individuals.
Intelligence grows, but it grows only particular potentials.
Plomin and Petrill offered a careful and precise view in their
1997 overview of behavioral genetics:
The word genetics is used very narrowly to refer to DNA differences
among individuals that are inherited from generation to generation.
It does not refer to the vast majority of DNA that is the same for all of
us or to the many DNA events that are not inherited, such as
mutations in DNA in cells other than the sex cells. Furthermore,
genetic research describes what the genetic and environmental ori-
gins of differences among individuals in a particular population at a
particular time are. It does not predict what could be, nor does it
prescribe what should be. Evidence for genetic influence (what is)
does not imply that differences among individuals are immutable or
irremediable-novel environmental factors could make a difference
(what could be).
The most important term in behavior-genetics is the range
of reaction. Everyone has a different responsiveness to similar
conditions, and that is hardwired. Technically put, different
genotypes may have the same phenotype, and different phe-
notypes may be expressed by the same genotype. Some
researchers have accepted the heritability statistic as an index
of genetic influence, but with questionable methodologies and
the reliance on IQ as a dependent measure, any generalizations
must be strongly qualified. Still, it is possible to conclude that
every individual has a range of alternative developmental out-
comes. IQ and the entire phenotype are products of a bipartite
system (environment and genes) and reactions within ranges.
Behavioral genetic research suggested that the genetic contri-
bution to intelligence was larger than the environmental, and
as Plomin and Petrill pointed out in 1997, that conclusion was
consistent with research on the heritability of psychopathol-
ogy. It is fortunate that large strides are being made with more
accurate molecular genetic methodologies.
See also: Behavior Genetics of Personality; Behavioral Genetics; Human
Intelligence; Molecular Genetics and Human Behavior.
The Behavior-Genetics of Intelligence 329
Further Reading
Gould SJ (1980) Jensen’s last stand (Review of Bias in mental testing by A. Jensen).
New York Review of Books 27(7): 38-44.
Jensen A (1980) Bias in Mental Testing. Toronto, ON: Free Press.
Plomin R and Petrill SA (1995) Genetics and intelligence: What’s new? intelligence
24: 53-77.
Scarr S and Weinberg RA (1979) Intellectual similarities in adoptive and
biologically related families of adolescents. In: Willerman L and Turner RG (eds.)
Readings About Individual and Group Differences. San Francisco, CA:
W.H. Freeman.
Vernon PE (1979) Intelligence: Heredity and Environment. San Francisco, CA: W.H.
Freeman.
Wilson E0 (1978) On Human Nature. Cambridge, MA: Harvard University Press.
Big Five Model and Personality Disorders
T A Widiger, University of Kentucky, Lexington, KY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Big Five Dimensional model of personality structure,
consisting of five broad domains of neuroticism,
extraversion, openness, agreeableness, and conscientious.
Each broad domain is differentiated into six more
specific facets.
Categorical Describing personality disorders in terms of
distinct types, each having distinct etiology, pathology, and
treatment implications. Differential diagnosis is focused on
identifying which specific disorder best describes the
person's condition.
Dimensional Describing personality disorder
multifactorially, which each person characterized by a
relatively unique profile or constellation of maladaptive
and adaptive personality traits, with each component
having specific implications for etiology, pathology, and
treatment.
DSM-IV-TR The official manual of the American Psychiatric
Association for the diagnosis of mental disorders (i.e.,
Diagnostic and Statistical Manual of Mental Disorders, 4th edn.,
text revision).
Personality One's characteristic manner of thinking,
feeling, behaving, and relating to others.
Personality disorder The presence of personality traits that
result in clinically significant impairment to social or
occupational functioning or personal distress.
An alternative to DSM-IV-TR is to consider personality disor-
ders to be maladaptive variants of general personality structure.
The predominant model for general personality structure
within psychology is the Big Five, otherwise known as the
five-factor model (FFM). The FFM consists of five broad
domains: neuroticism (or emotional instability), extraversion
versus introversion, openness versus closedness, agreeableness
versus antagonism, and conscientiousness. The FFM was
derived originally through empirical studies of the trait terms
within the English language. Language can be understood as a
sedimentary deposit of the observations of persons over the
thousands of years of the language's development and trans-
formation. The most important domains of personality func-
tioning are those with the most number of trait terms to
describe and differentiate the various manifestations and
nuances of a respective domain, and the structure of personal-
ity is suggested by the empirical relationships among these trait
terms. The initial lexical studies of the English language con-
verged well onto the five-factor structure. Subsequent lexical
studies have been conducted on many additional languages
(e.g., German, Dutch, Czech, Polish, Russian, Italian, Spanish,
Hebrew, Hungarian, Turkish, Korean, and Filipino) and these
have confirmed well the existence of the five broad domains.
The five domains have been differentiated into 30 more spe-
cific facets on the basis of their development of and research
with the NEO Personality Inventory-Revised (NEO PI-R), by
far the most commonly used and heavily researched measure
of the FFM. Table 1 provides a brief characterization of both
the normal and abnormal variants of each of the 60 poles of
the 30 facets of the FFM in terms of the five-factor form (FFF),
a more elaborated version of the FFM rating form (FFMRF).
Advantages of Integrating DSM-IV-TR with the Big Five
Integrating the diagnosis of personality disorders with the
Big Five has a number of distinct advantages, including
the resolution of the many failures of the existing diagnostic
system, such as the heterogeneity among persons sharing the
same categorical diagnosis, the excessive diagnostic comorbid-
ity, inadequate coverage, the inadequate scientific base, and
the arbitrary and unstable boundary with normal psychologi-
cal functioning. Each of these will be discussed in turn.
Heterogeneity of Persons Sharing the Same Diagnosis
The DSM-IV-TR personality disorders describe prototypic cases,
yet very few actual cases are prototypic. As a result, a personality
disorder diagnosis provides a misleading description of any
particular patient's unique personality profile. Persons can in
fact be provided the same diagnosis of antisocial personality
disorder yet not share one feature in common, as any three of
seven diagnostic criteria can be used to make the diagnosis.
Heterogeneity among persons is addressed by the Big Five
through the provision of multifactorial descriptions of an indi-
vidual's psychopathology (personality disorder) profile that
matches well his or her actual constellation of maladaptive
personality traits. In addition, the Big Five or FFM allows for
the inclusion of normal and adaptive personality traits within
the profile description, thereby providing a more complete and
comprehensive description. Personality disorder diagnoses can
be quite stigmatizing, suggesting that the person's characteristic
manner of thinking, feeling, and relating to others (i.e., the
persons sense of self) is itself a mental disorder. The FFM's
inclusion of normal and adaptive personality traits provides a
means to recognize and appreciate that a person is more than
just the personality disorder and that there are aspects to the
self that can be adaptive, even commendable, despite the pres-
ence of the personality disorder. Some of these strengths may
also be quite relevant to treatment, such as openness to experi-
ence indicating an interest in exploratory psychotherapy, agree-
ableness indicating an engagement in group therapy, and
conscientiousness indicating a willingness and ability to adhere
to the demands and rigors of dialectical behavior therapy.
330
Big Five Model and Personality Disorders 331
Table 1 Adaptive and maladaptive variants of the five-factor model as presented in five-factor form
Maladaptively high
Normal high
Normal low
Maladaptively tow
Neuroticism
Anxiousness
Fearful, anxious
Vigilant, worrisome, wary
Relaxed, calm
Fearless
Angry hostility
Rageful
Brooding, resentful, defiant
Even-tempered
Won’t even protest exploitation
Depressiveness
Depressed, suicidal
Pessimistic, discouraged
Hot easily discouraged
Unrealistic, overly optimistic
Self-consciousness
Uncertain of self or identity
Self-conscious, embarrassed
Self-assured, charming
Glib, shameless
Impulsivity
Unable to resist impulses
Self-indulgent
Restrained
Overly restrained
Vulnerability
Helpless, emotionally
unstable
Vulnerable
Resilient
Feels invincible, oblivious
Extraversion
Warmth
Intense attachments
Affectionate, warm
Formal, reserved
Cold, distant
Gregariousness
Attention-seeking
Sociable, outgoing, personable
Independent
Isolated
Assertiveness
Dominant, pushy
Assertive, forceful
Passive
Resigned, uninfluential
Activity
Frantic
Energetic
Slow-paced
Lethargic, sedentary
Excitement-seeking
Reckless, foolhardy
Adventurous
Cautious
Dull, listless
Positive emotions
Melodramatic, manic
High-spirited, cheerful, joyful
Placid, sober, serious
Grim, anhedonic
Openness
Fantasy
Unrealistic, lives in fantasy
Imaginative
Practical, realistic
Concrete
Esthetics
Bizarre interests
Esthetic interests
Minimal esthetic interests
Disinterested
Feelings
Intense, in turmoil
Self-aware, expressive
Constricted, blunted
Alexithymic
Actions
Eccentric
Unconventional
Predictable
Mechanized, stuck in routine
Ideas
Peculiar, weird
Creative, curious
Pragmatic
Closed-minded
Values
Radical
Open, flexible
Traditional
Dogmatic, moralistically
intolerant
Agreeableness
Trust
Gullible
Trusting
Cautious, skeptical
Cynical, suspicious
Straightforwardness
Guileless
Honest, forthright
Savvy, cunning, shrewd
Deceptive, dishonest,
manipulative
Altruism
Self-sacrificial, selfless
Giving, generous
Frugal, withholding
Greedy, exploitative
Compliance
Yielding, docile, meek
Cooperative, obedient,
deferential
Critical, contrary
Combative, aggressive
Modesty
Self-effacing, self-
denigrating
Humble, modest, unassuming
Confident, self-assured
Boastful, pretentious, arrogant
Tender-Mindedness
Conscientiousness
Overly soft-hearted
Empathic, sympathetic, gentle
Strong, tough
Callous, merciless, ruthless
Competence
Perfectionistic
Efficient, resourceful
Casual
Disinclined, lax
Order
Preoccupied
w/organization
Organized, methodical
Disorganized
Careless, sloppy, haphazard
Dutifulness
Rigidly principled
Dependable, reliable,
responsible
Easy-going, capricious
Irresponsible, undependable,
immoral
Achievement
Workaholic
Purposeful, diligent, ambitious
Carefree, content
Aimless, shiftless, desultory
Self-discipline
Single-minded
doggedness
Self-disciplined, willpower
Leisurely
Negligent, hedonistic
Deliberation
Ruminative, indecisive
Thoughtful, reflective,
circumspect
Quick to make decisions
Hasty, rash
Excessive Diagnostic Co-occurrence
DSM-IV-TR provides diagnostic criterion sets to help guide a
clinician toward a correct diagnosis and an additional section
devoted to differential diagnosis when persons meet diagnostic
criteria for more than one disorder. The intention of this
information is to help the clinician determine which particular
mental disorder is present, the identification of which would
hopefully indicate the presence of a specific pathology and sug-
gest a specific treatment. It is evident, however, that DSM-IV-TR
routinely fails in this goal, despite the best efforts of the leading
clinicians and researchers who have authored the manual.
The authors of the diagnostic manual construct personality
syndromes and then hope that persons will share the same
precise profile and not have any traits of any other profile. This
assumption of distinct types is not supported empirically, con-
tributing not only to the heterogeneity of persons sharing the
same diagnosis but also the need to provide multiple diagnoses
to the same patient in order to fully describe all of his or her
maladaptive personality traits. The problem is further com-
pounded by the considerable overlap among the personality
syndromes included within DSM-IV-TR, a comorbidity that is
well explained by overlap in FFM personality traits (e.g., most
of the personality disorders are heavily saturated with neuroti-
cism; schizoid, avoidant, and schizotypal share traits of intro-
version; antisocial, narcissistic, and paranoid share traits of
antagonism).
332 Big Five Model and Personality Disorders
Inadequate Coverage
Clinicians provide a diagnosis of not otherwise specified
(NOS) when they determine that a person has that particular
class of mental disorder but the symptoms are not adequately
represented by any one of the individual diagnostic categories.
Personality disorder NOS (PDNOS) is often the single most
frequently used diagnosis in clinical practice, as indicated in
studies of clinical records and in a meta-analysis of NOS usage
across structured and unstructured assessments.
The FFM is considered to be a reasonably comprehensive
dimensional model of general personality structure. One of its
strengths is its ability to accommodate the personality traits
included within alternative models of personality. The FFM
also accommodates well the maladaptive personality traits
included within the DSM-IV-TR diagnostic categories, as well
as additional maladaptive personality traits not included
within the existing nomenclature. Table 2 provides a descrip-
tion of each of the DSM-IV-TR personality disorders in terms
of the FFM. For example, the FFM includes the traits of
DSM-IV-TR antisocial personality disorder (deception, exploi-
tation, aggression, irresponsibility, negligence, rashness, angry
hostility, impulsivity, excitement-seeking, and assertiveness),
and goes beyond DSM-IV-TR to include additional traits within
the widely popular Psychopathy Checklist-Revised (PCL-R),
such as glib charm (low self-consciousness), arrogance, and
lack of empathy (tough-minded callousness) and goes even
further to include traits of psychopathy emphasized originally
by Cleckely but not included within either the DSM-IV-TR or
the PCL-R, such as low anxiousness and low vulnerability or
fearlessness.
It is evident from Table 2 that there is considerable overlap
among the DSM-IV-TR personality disorders with respect to
their FFM profile. For example, many of the personality dis-
orders are characterized by facets of high neuroticism and
low agreeableness, contributing to their considerable diagnos-
tic co-occurrence. However, the FFM also identifies traits
that are relatively unique to a respective diagnosis. For exam-
ple, the FFM has the social withdrawal evident not only in the
Table 2 DSM-IV-TR personality disorders from the perspective of the five-factor model of general personality structure
PRN SZD SZT ATS BDL HST NCS AVD DPD OCP
Neuroticism (vs. emotional stability)
Anxiousness (vs. unconcerned)
Angry hostility (vs. dispassionate) H
Depressiveness (vs. optimistic)
Self-consciousness (vs. shameless)
Impulsivity (vs. restrained)
Vulnerability (vs. fearless)
Extraversion (vs. introversion)
Warmth (vs. coldness) L
Gregariousness (vs. withdrawal) L
Assertiveness (vs. submissiveness)
Activity (vs. passivity)
Excitement-seeking (vs. dullness)
Positive emotionality (vs. anhedonia)
Openness (vs. closedness)
Fantasy (vs. concrete)
Aesthetics (vs. disinterest)
Feelings (vs. alexithymia)
Actions (vs. routine) L
Ideas (vs. closed-minded)
Values (vs. dogmatic) L
Agreeableness (vs. antagonism)
Trust (vs. mistrust) L
Straightforwardness (vs. deception) L
Altruism (vs. exploitation)
Compliance (vs. opposition, aggression) L
Modesty (vs. arrogance)
Tender-mindedness (vs. tough-minded) L
Conscientiousness (vs. disinhibition)
Competence (vs. ineptitude)
Order (vs. disordered)
Dutifulness (vs. irresponsibility)
Achievement-striving (vs. lackadaisical)
Self-discipline (vs. negligence)
Deliberation (vs. rashness)
H
H
L L
L L
L
L
L L
L
L
H
L H
H H
H
L H
H H
L H
H
H
H
H
H
H H
H
L L
H
H H
H
H
H L
H H
H H
H H
H H
H
L
L L
L
L
H
L
L
L
L
L
L
H
H
H
H
H
H
PRN, paranoid; SZD, schizoid; SZT, schizotypal; ATS, antisocial; BDL, borderline; HST, histrionic; NCS, narcissistic; AVD, avoidant; DPD, dependent; OCP, obsessive-compulsive;
H, high; L, low.
Big Five Model and Personality Disorders 333
avoidant, schizoid, and schizotypal personality disorders, but
also the anxiousness and self-consciousness of the avoi-
dant, the anhedonia that is considered to be specific to the
schizoid, and the cognitive-perceptual aberrations of the schi-
zotypal. The FFM has the intense attachment needs, the defer-
ence, and the self-conscious anxiousness of the dependent, the
perfectionism and workaholism of the obsessive-compulsive,
and the fragile vulnerability and emotional dysregulation of
the borderline. The FFM also goes beyond the DSM-IV-TR
nomenclature to include (for instance) closed-mindedness
(evident in racist, prejudicial persons) and alexithymia.
Most individual persons will have personality profiles
that fail to match perfectly any one of the 10 DSM-IV-TR
prototypes provided in Table 2. The need for the frequently
used wastebasket NOS diagnosis would decrease substantially
if the DSM-V shifted to a profile description in terms of the
domains and facets of the FFM as subthreshold, atypical, and
idiosyncratic profiles of individual cases would be readily
described in terms of the 60 poles of the 30 facets of the FFM.
Scientific Foundation
One fundamental benefit of integrating the DSM-IV-TR with
the FFM is the presence of a considerable body of basic scien-
tific research to support the validity of the FFM in contrast to
the inadequate scientific foundation for at least half of the
DSM-IV-TR personality disorders. Briefly discussed here will
be genetics, childhood antecedents, temporal stability, and
universality.
Genetics
There is behavior genetic research to support the validity of the
DSM-IV-TR personality disorders but the extent of support
does pale in comparison to the FFM. In addition, research
concerning the FFM has gone beyond univariate behavior
genetics that simply indicate extent of heritability to multivari-
ate behavior genetics to provide validity for the specific FFM
personality structure. There is also quite a bit of molecular
genetic research on FFM neuroticism. In contrast, it might be
difficult to find even one molecular genetic study of a DSM-IV-
TR-TR personality disorder as the heterogenous constellations
of maladaptive personality traits that describe each of them are
antithetical to discovering a specific genetic contribution.
Childhood Antecedents
Very little is said in DSM-IV-TR on the childhood ante-
cedents for most of the personality disorders because little
is in fact known. Much has been written on childhood ante-
cedents but this literature is based largely on clinical experi-
ence, theoretical expectations, and retrospective studies. There
has been only one longitudinal prospective study providing
information concerning the histrionic, dependent, avoidant,
obsessive-compulsive, paranoid, schizoid, schizotypal, and
narcissistic personality disorders, and this project was sorely
limited by the absence of adequate assessments in childhood
and adolescence.
In contrast, there is a considerable body of research on
childhood personality traits and temperaments all of which
has been well integrated within the FFM. Rather than seek an
understanding of the childhood antecedents of personality
disorder from studies of personality disorder symptomatology
in children and adolescents, it might be more fruitful to seek an
understanding from the study of childhood temperaments. If
childhood temperaments do not provide an initial foundation
or disposition toward the development of an adult personality
disorder, what would be the genetic foundation for an adult
personality disorder?
However, the adult outcome of childhood temperaments is
well understood to be a general personality structure. It would
be rather incongruous to suggest that the temperaments of
childhood and adolescence provide a disposition or vulnera-
bility to adult personality disorder, yet they also suggest that
these personality disorders cannot themselves be understood
as maladaptive variants of these temperaments and traits
within adulthood. Researchers have even developed a model
and instrument for the assessment of FFM personality disorder
in childhood and adolescence.
Temporal Stability
Temporal stability is central to the validity of personality traits.
Fundamental to the distinction between a personality disorder
and most other mental disorders is the temporal stability of
personality. However, there is increasing concern regarding the
temporal stability of the DSM-IV-TR personality disorder diag-
noses and DSM-IV-TR personality disorder researchers are now
even suggesting that it should no longer be a defining feature of
a personality disorder.
In contrast, there is considerable support for the temporal
stability of FFM personality traits across the lifespan. A meta-
analysis of the findings of 152 longitudinal studies was con-
ducted, involving 3217 test- retest correlations that covered
various spans of time from childhood to old age. The conclu-
sion was that the FFM traits are quite consistent over the life
course. One prominent study compared directly the temporal
stability of the DSM-IV-TR diagnoses and FFM personality traits
over a 2-year span of time, and reported weak stability for the
DSM-IV-TR diagnoses but strong stability for the FFM traits.
Universality
A common distinction in cross-cultural research is between etic
and emic studies. Etic studies use constructs and measures
from one culture imported into another, determining (in
part) whether the importation reproduces the nomological
net of predictions previously obtained in other cultures. Emic
studies use constructs and measures that are indigenous to a
particular culture, determining whether a particular model of
personality structure is evident from the perspective or experi-
ences of that culture.
There is only limited etic support for the universality of the
DSM-IV-TR nomenclature and virtually no emic support. It is
perhaps noteworthy that DSM-IV-TR narcissistic personality
disorder was not included within the World Health Organiza-
tion's (WHO) International Classification of Diseases (ICD-10)
despite its presence since 1980 within DSM-III. Of course,
334 Big Five Model and Personality Disorders
it now appears that ICD-11 may simply accept whatever
nomenclature is developed by the APA so there will no longer
be any inconsistency.
The universality of the FFM is supported by extensive etic
and emic research. Emic support for the FFM is provided by the
many lexical studies of the trait terms within the indigenous
languages of other cultures (e.g., German, Dutch, Czech,
Polish, Russian, Italian, Spanish, Hebrew, Hungarian, Turkish,
Korean, and Filipino). Etic support is provided by a replica-
tion of the FFM personality structure using observer ratings
of 11985 individuals obtained in 51 different cultures.
The largest cross-cultural study of the FFM to date included
100 researchers and 17 837 participants from 56 countries.
Results indicated that the five-dimensional structure was
highly robust across major regions of the world, including
North America, South America, Western Europe, Eastern
Europe, Southern Europe, the Middle East, Africa, Oceania,
South-Southeast Asia, and East Asia.
FFM Personality Disorder Diagnosis
The FFM or Big Five personality disorder diagnosis consists of
four steps. The first step is to obtain an FFM description of the
patient. This can be accomplished most easily through the
scoring of a one-page rating form that covers all of the domains
and facets of the FFM, as assessed (for instance) by the FFF (see
Table 1), or through a self-report inventory (e.g., NEO PI-R) or
through a semistructured interview (e.g., Structured Interview
for the FFM, SIFFM). There are also many other available
measures of the FFM.
The second step is to identify impairments associated with
any elevations on a respective facet of the FFM. Table 1 pro-
vides typical impairments associated with each of the 60 poles
of the 30 facets of the FFM. More detailed descriptions are
available elsewhere (see Further Reading). This assessment is
included explicitly within the administration of the SIFFM. For
example, if persons endorse going out of their way to help
others (high altruism) they are asked if they do this at the
sacrifice of their own needs; if persons endorse confiding in
others (high trust) they are asked in the SIFFM if they have ever
been mistreated or used by others as a result.
The third step of the FFM four-step procedure is to deter-
mine whether the impairments are at a clinically significant
level warranting the diagnosis of a personality disorder. This
step is in contrast to the thresholds for the diagnosis of indi-
vidual personality disorders provided within the DSM-IV-TR
(e.g., five of nine diagnostic criteria for borderline, three of
seven for antisocial). There has never been an effort to set a
consistent diagnostic threshold for the personality disorders
within any edition of the APA diagnostic manual. In each
instance, the thresholds have been set independently of one
another. The diagnostic thresholds in DSM-IV-TR (with the
exception of antisocial) were determined by an individual
assigned to that diagnosis who selected a threshold which
seemed on the basis of a visual inspection of the criterion set
to be an optimal point of demarcation.
The DSM-III and DSM-III-R personality disorder diagnostic
thresholds (with two exceptions) were never based on any
explicit or published rationale. The two exceptions are the
diagnostic thresholds for the DSM-III versions of the border-
line and schizotypal personality disorders. Their DSM-III cutoff
points were selected on the basis of maximizing agreement
with diagnoses provided by a large sample of clinicians. How-
ever, it evident there have since been so many revisions, dele-
tions, and additions to their criterion sets that the current
diagnostic thresholds no longer relate well to the original
thresholds. There was never an effort to provide any rationale
for the diagnostic thresholds for the DSM-III paranoid, schiz-
oid, narcissistic, histrionic, antisocial, avoidant, dependent, or
obsessive-compulsive personality disorders. With the excep-
tion of the DSM-IV-TR antisocial personality disorder, no effort
has been taken to maintain a consistent threshold across revi-
sions to the diagnostic manual. Hence, with each revision,
remarkable, unanticipated, and unintended changes to preva-
lence rates have occurred.
The FFM of personality disorder, in contrast, proposes using
a uniform and consistent basis for determining when a person-
ality disorder is present. The cutoff point is modeled after the
Global Assessment of Functioning (GAF) scale provided on
Axis V of DSM-IV-TR that has itself been used in quite a
number of studies within clinical populations. A score of 71
or above on the GAF indicates a normal range of functioning
(e.g., problems are transient and expectable reactions to stres-
sors); a score of 60 or below represents a clinically significant
level of impairment (moderate difficulty in social or occupa-
tional functioning, such as having few friends or significant
conflicts with coworkers) that would warrant a diagnosis
of personality disorder. Further explication of this scale is
provided by the Global Assessment of Relational Functioning
and the Social and Occupational Functioning scales. This
approach to setting the diagnostic threshold would bring to the
diagnosis research that has been conducted with the GAF and
would provide a threshold that would be consistent across all
personality disorders as well as facilitate consistency across
all future revisions to the nomenclature.
The fourth step, prototypal matching, is an optional step for
those who still wish to provide single diagnostic terms to
describe a particular patient's personality profile. In this step,
one correlates the patient's actual FFM profile with the FFM
description of a prototypic case (see Table 2). Research has
demonstrated that these personality disorder indices are just as
valid for the assessment of the borderline, the psychopathic
and other personality disorders as any two explicit measures of
these personality disorders. In other words, one can readily
translate an FFM description back into a DSM-IV-TR personal-
ity disorder construct. In most cases, however, the quantitative
matching will serve primarily to indicate the extent to which
any single construct (e.g., antisocial) is inadequately descrip-
tive of the individual person. In the vast majority of cases, the
optimal description will be provided by the actual FFM profile
of the person rather than the extent to which the person's FFM
profile resembles a hypothetical prototype.
Feasibility
Most clinicians may understandably respond with a deep
breath of concern upon first inspection of Table 1, finding
it daunting to conceive of having to become familiar with
Big Five Model and Personality Disorders 335
both the adaptive and maladaptive variants of all 60 poles of all
30 facets of the FFM. However, there are a few points to keep in
mind. First, the language of the FFM is the language with which
one is already familiar. FFM constructs are already known to
anyone who understands his or her own language. This is not to
say that the FFM fails to accommodate sophisticated clinical
constructs. Research has indicated that constructs within the
California Q-set, the Shedler-Westen Assessment Procedure-
200, and the DSM-IV-TR personality disorder criterion sets are
well understood and articulated in terms of the FFM.
The FFM structure is also not difficult to learn as it is the
structure within which one already conceptualizes and thinks
about these traits. The FFM structure reproduces the naturally
occurring structure within one's language. It is no accident
that persons reviewing disparate trait research concerning (for
instance) gender differences, childhood temperament, health
psychology, consequential outcomes of personality, and even
animal behavior have all used the FFM as the structure with
which to organize these vast trait literatures.
The administration of the SIFFM, a structured interview
assessment of the FFM, takes approximately half the time it
takes to administer a DSM-IV-TR personality disorder semi-
structured interview, despite the fact that it is assessing normal
as well as abnormal personality traits. The reason is in part that
the SIFFM assessment of maladaptive personality traits is
appreciably less redundant than an assessment of the markedly
overlapping DSM-IV-TR personality disorder criterion sets. In
addition, much of the time spent in the administration of a
DSM-IV-TR personality disorder semistructured interview is
spent in the assessment of diagnostic criteria that are not
present. The DSM-IV-TR requires that all of the diagnostic
criteria for each personality disorder be assessed in every
patient. A substantial proportion of this time is eliminated in
the FFM of personality disorder as abnormal variants are not
assessed if the respective normal variant is not present.
Nevertheless, clinicians may still find it daunting to have to
consider (potentially) 60 maladaptive traits for each patient, in
contrast to just the ten personality disorders of DSM-IV-TR.
Figure 1 provides an abbreviated version of the FFM of person-
ality disorder. Figure 1 reduces the 60 maladaptive variants to
just 26. This reduction was achieved in part by eliminating
poles of facets that were considered to be relatively less fre-
quent or even obscure (e.g., maladaptively high openness to
esthetics) for most clinical use.
Figure 2 provides further detail on how the domain and
maladaptive variants could be assessed. Provided within
Figure 2 are draff diagnostic criteria for neuroticism and for
the eight respective maladaptive facets for that domain
included within the abbreviated version (additional domains
are covered in papers listed in the section for Further Reading).
A patient is first assessed with respect to the six facets of
FFM neuroticism. Each facet is rated on a 1-5 Likert scale,
consistent with the FFMRF and FFF. If the person receives an
elevated score for the domain of neuroticism, then the clini-
cian would assess for the presence of the six maladaptive
variants of high neuroticism (i.e., anxiousness, depressiveness,
angry hostility, emotional instability, self pathology, and help-
lessness) and would not need to assess for the maladaptive
variants of low neuroticism. If the person receives a low score
for neuroticism, then the clinician would not need to assess
for any of the maladaptive variants of high neuroticism and
would assess instead the two maladaptive variants of low
neuroticism (i.e., glib, shameless charm, and fearlessness).
If the person receives a score within the middle range, then
no further assessment would typically be necessary.
The abbreviated version does naturally fail to include all
of the maladaptive traits present within the FFM, as illustrated
within Table 1. Missing from the abbreviated version, for
instance, are such maladaptive variants as attention-seeking
(high gregariousness), gullibility (high trust), guilelessness
(high straightforwardness), exploitativeness and greed (low altru-
ism), callousness (high tough-mindedness), ruminative inde-
cisiveness (high deliberation), hedonism (low self-discipline),
and aimlessness (low achievement-striving). Any one of
these could, in some instances, be very clinically important.
One cannot have an abbreviated model without losing
Emotional
| Depressiveness |
dysregulation
Meekness
Aggression
7
Alexithymic
/
Suspiciousness
Figure 1 Abbreviated version of FFM proposal for DSM-V.
336 Big Five Model and Personality Disorders
Angry hostility
Glib,
shameless charm
Superficially charming
Emotionally glib and facile
Incapable of embarrassment or shame
Talks his/her way out of problems
Lacks feelings of guilt or remorse
Fearless
Appears incapable of feeling afraid
Acts fearless despite clear dangers
Takes risks due to fearlessness
Feels invincible or invulnerable
Remains calm when under stress
Figure 2 FFM diagnosis of maladaptive neuroticism.
information and some of this information will at times have
important clinical significance. In addition, in the abbreviated
model it will be relatively more difficult to identify complex
profiles (e.g., elevations in opposite directions on facets within
the same domain). Many clinicians will then likely prefer the
complete model, once they become familiar with the simpler,
abbreviated model.
However, even in the abbreviated version clinicians would
be alerted to the potential presence of the additional maladap-
tivity through elevations on specifically relevant facet scales
within each of the five domains. If a person receives a high (or
low) score on a respective facet for which no maladaptive
variant is provided within the abbreviated version (e.g., a
score of 5 for the facet of impulsivity for neuroticism, suggesting
a potential difficulty in control affectively driven impulses and
urges, or a score of 1 for angry hostility, suggesting the potential
for never standing up for oneself in protest or justifiable anger;
see Table 1 ) the maladaptive personality functioning could still
be assessed, if wished, through an implementation of the more
complete, full version of the FFM of personality disorder.
Conclusions
There are many advantages of shifting the classification of
personality disorder to the FFM. One straightforward advan-
tage is the provision of a description of abnormal personality
functioning with the same model and language used to
describe general personality structure. It would address
the many fundamental limitations of the categorical model
(e.g., heterogeneity within diagnoses, inadequate coverage,
lack of consistent diagnostic thresholds, excessive diagnostic
co-occurrence, and inadequate scientific base). It would con-
tain the means of providing more comprehensive and indi-
vidually specific descriptions of each patient's normal and
abnormal personality structure and, it would transfer to the
psychiatric nomenclature a wealth of knowledge concerning
the origins, childhood antecedents, stability, and universality
of the dispositions that underlie personality disorder.
See also: Avoidant Personality Disorder; Borderline Personality
Disorder; Extraversion-Introversion; Histrionic Personality Disorder;
Individual Differences in Temperament: Definition, Measurement, and
Outcomes; Personality Disorders; Personality, Structure;
Psychopathology: Diagnosis, Assessment, and Classification; Schizoid
and Schizotypal Personality Disorder; Temperament and Individual
Differences.
Further Reading
Clark LA (2007) Assessment and diagnosis of personality disorder: Perennial
issues and an emerging reconceptualization. Annual Review of Psychology
57: 227-257.
Big Five Model and Personality Disorders 337
De Clercq B, De Fruyt F, Van Leeuwen K, and Mervielde I (2006) The structure of
maladaptive personality traits in childhood: A first step toward an integrative
developmental perspective for DSM-V. Journal of Abnormal Psychology 115: 639-657.
Lynam DR and Widiger TA (2001) Using the five factor model to represent the DSM-IV
personality disorders: An expert consensus approach. Journal of Abnormal
Psychology WO: 401-412.
Mullins-Sweatt SN, Jamerson JE, Samuel SB, Olson DR, and Widiger TA (2006)
Psychometric properties of an abbreviated instrument for the assessment of the five
factor model. Assessment 13: 119-137.
Mullins-Sweatt SN and Widiger TA (2007) The Shedler-Westen Assessment Procedure
from the perspective of general personality structure. Journal of Abnormal
Psychology WO: 618-623.
Samuel DB and Widiger TA (2008) A meta-analytic review of the relationships between
the five-factor model and DSM-IV-TR personality disorders: A facet level analysis.
Clinical Psychology Review 28: 1 326-1 342.
Widiger TA (2009) Neuroticism. In: Leary M and Hoyle R (eds.) Handbook of Individual
Differences in Social Behavior, pp. 129-146. New York: Guilford.
Widiger TA, Costa PT, and McCrae RR (2002) A proposal for Axis II:
Diagnosing personality disorders using the five factor model. In: Costa PT and
Widiger TA (eds.) Personality Disorders and the Five Factor Model of
Personality, 2nd edn., pp. 431-456. Washington, DC: American Psychological
Association.
Widiger TA, De Clerq B, and De Fruyt F (2009) Childhood antecedents of personality
disorder: An alternative perspective. Development and Psychopathology
21: 771-791.
Widiger TA and Mullins-Sweatt SN (2009) Five-factor model of personality
disorder: A proposal for DSM-V. Annual Review of Clinical Psychology
5: 115-138.
Widiger TA and Trull TJ (2007) Plate tectonics in the classification of personality
disorder: Shifting to a dimensional model. American Psychologist 62: 71—83.
Bilingualism and Multilingualism
L Wei, University of London, London, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Affective filter hypothesis Emotions function as a filter that
reduces the amount of language input the learner is able to
comprehend. Negative emotions such as anxiety, self-doubt,
and mere boredom interfere with the process of acquiring a
second language. Positive emotions such as interest and
self-esteem can facilitate the process.
BAMFLA Bilingual and multilingual first language
acquisition, that is, acquiring more than one language
simultaneously as first language from birth.
Code-switching Alternation between languages in the same
conversational interaction or across different modalities
such as sign and speech.
Compound bilingual A bilingual who has learned two
languages at the same time and often in the same context.
Coordinative bilingual A bilingual who has learned two
languages in different contexts and kept the two languages
fairly separately.
Critical period hypothesis A hypothesis that claims that
there is an ideal 'window' of time to acquire language, after
which this is no longer possible, that is, language acquisition
is biologically linked to age.
Input hypothesis Also known as the comprehensible input
hypothesis, it claims that learners acquire a language only
when they receive comprehensible input.
Interaction hypothesis A hypothesis that proposes that
language acquisition is strongly facilitated by the use of the
target language in interaction.
Language contact Situations where groups of speakers of
different languages are in contact with each other. Language
contact is the main cause of bilingualism and
multilingualism.
Language ideology or linguistic ideology The explicit
and implicit attitudes regarding language that define
what is perceived as proper speech or good language
behavior. It also encompasses what people believe a
particular language can do for its users and the user
community.
Language socialization A process whereby children
and other novices in society acquire tacit knowledge
of principles of social order and systems of belief
through exposure to and participation in
language-mediated interaction. It involves both
socialization through language and socialization to
use language.
Negative motivation A response that involves the
undertaking of tasks for fear that there should be
undesirable outcomes.
Negotiation of identity A process through which people
reach agreements regarding 'who is who' in their
relationships. It helps to establish what people can expect of
one another.
Output hypothesis A hypothesis that claims that
meaningful output is as necessary to language learning as
meaningful input.
Positive motivation A response which includes
enjoyment and optimism about the tasks that the learner is
involved in.
Sign bilingualism Dual modality bilingualism of sign
language and other human languages.
Subordinative bilingual A bilingual whose two languages
are not of equal status or proficiency level, with one being
subordinate to the other in structure, cognitive processing,
and usage.
Introduction
One of the major discoveries of modern linguistic science is
that human beings have the capacity to learn as many lan-
guages as they are willing to and as they are allowed to by the
social conditions. In principle then, anybody can be bilingual
and multilingual. Indeed, the vast majority of the world's
population is bilingual and multilingual. Nevertheless, there
are different pathways to bilingualism and multilingualism,
and there are plenty of examples of people who have failed to
acquire a language sufficiently to communicate with others in
it, or those who, despite knowing other languages, function as
monolinguals. The questions then are the following:
• What are the conditions that facilitate the learning and use
of multiple languages?
• As far as individuals are concerned, what does it mean to be
bilingual or multilingual?
Pathways to Bilingualism and Multilingualism
People often assume that bilingualism and multilingualism are
for the privileged few - those who have had a good schooling
where they learned new languages and those who have traveled
a lot or lived in different countries. Education and migration,
either temporary or long-term, are clearly opportunities for
learning languages, but most people become bilingual and
multilingual within their immediate families and commu-
nities. Table 1 illustrates six different processes whereby chil-
dren can become bilingual or multilingual at home.
There are many other factors that can lead individuals on to
a path to bilingualism and multilingualism. Most of them are
external to the individual's cognitive capacities but have some-
thing to do with the broader social environment. For example,
political or military acts such as colonization, annexation,
resettlement, and federation of a country or region can have
critical linguistic effects. People may become refugees, either in
338
Bilingualism and Multilingualism 339
Table 1 Processes of bilingual acquisition
Parent
Community
Strategy
1. One person one language
The parents have different native
languages with each having
some degree of competence in
the other’s language
The language of one of the
parents is the dominant
language of the community
The parents each speak their own language to
the child from birth
2. Nondominant home
The parents have different native
The language of one of the
Both parents speak the nondominant
language/one language
one environment
languages
parents is the dominant
language of the community
language to the child, who is fully exposed
to the dominant language only when outside
the home, and in particular in nursery
school
3. Nondominant home
The parents share the same native
The dominant language is not
The parents speak their own language to the
language without
community support
language
that of the parents
child
4. Double nondominant
The parents have different native
The dominant language is
The parents each speak their own language to
home language without
community support
languages
different from either of the
parents’ languages
the child from birth
5. Nonnative parents
The parents share the same native
language
The dominant language is the
same as that of the parents
One of the parents always addresses the child
in a language which is not his/her native
language
6. Mixed languages
The parents are bilingual
Sectors of community may
also be bilingual
Parents code-switch and mix languages
a new place or in their homeland, and be required to learn the
language of their new environment. After a successful military
invasion, the indigenous population may have to learn the
invader's language in order to prosper. Colonization is exem-
plified by the former British, French, Spanish, Portuguese, and
Dutch colonies in Africa, Asia, and South America, most of
which achieved independence in the nineteenth century.
A modem example of annexation can be found in the absorp-
tion of the Baltic republics - Lithuania, Latvia, and Estonia -
into the Soviet Union after the Second World War. In the latter
part of the twentieth century, military conflicts in Cennal Africa
and the former Yugoslavia have seen the resettlement of
people of different ethnic backgrounds. Examples of federa-
tions where diverse ethnic groups or nationalities are united
under the political control of one state include Switzerland,
Belgium, and Cameroon. Individuals in these countries and
regions may have the opportunity to become bilingual or
multilingual. Similarly, natural disasters such as famine, floods,
and volcanic eruptions can be the cause of major movements of
population. New language contact situations emerge as people
begin to resettle. Some of the Irish and Chinese resettlements
in North America have been the result of natural disasters.
People are exposed to new languages as a result of the resettle-
ments. In today's world, the availability of information and
communication technologies (ICT), such as the Internet, has
led to a further expansion of the use of certain languages across
the world. People who wish to use modern technologies need
to learn and use the languages of the media.
Some of the pathways to bilingualism and multilingualism
may be more personal than environmental. For instance,
people may wish to live in a country because of its religious
significance, or to leave a country because of its religious oppres-
sion. In either case, a new language may have to be learned. The
Russian speakers in Israel are a case in point. Similarly, very
large numbers of people across the world have migrated to find
work and to improve their standard of living. The economic
factor accounts for most of the linguistic diversity of the United
States and for the increasing proportion of multilingualism
in present-day Europe.
Factors Contributing to the Success or Failure of
Becoming Bilingual and Multilingual
From the above list, one can easily conclude that becoming
bilingual or multilingual is a consequence of what the envi-
ronment provides the individual with and has little to do with
individuals' mental capacity. Yes, it is true that in theory
anyone could learn as many languages as he or she wishes to.
But the reality is that not everyone becomes bilingual or multi-
lingual even if the environmental conditions are favorable
for language learning. It is certainly a fact that not everyone
becomes equally proficient in the languages he or she learns
even if he or she is exposed to the same learning conditions.
Linguists have spent a great deal of time investigating the vari-
ous factors that lead to success or failure in becoming bilingual
and multilingual. Some of the factors appear to be learner-
internal, that is, they are conditioned by the learner's biological
and cognitive capacities, while others are more social.
Research evidence suggests that human beings' ability to
learn languages is closely related to age and memory capacity.
Infants seem to be able to acquire new languages with great
ease, whereas adults often have difficulty in achieving the
same level of fluency or accuracy in the later-learned languages.
Such facts gave rise to the 'critical period hypothesis,' which
claims that there is an ideal age to acquire a language in a
linguistically rich environment, after which the individual
will never achieve a full command of the language. However,
there is little agreement as to what that critical period is.
Research results are varied: some demonsttate that prepubes-
cent children acquire language easily, and some that older
learners have a certain advantage and can achieve native-level
340 Bilingualism and Multilingualism
proficiency. Brain plasticity has often been cited as the factor
underlying the critical period. More recently, it has been sug-
gested that the delayed development of the prefrontal cortex
in children and a concomitant delay in the development of
cognitive control may facilitate convention learning, allowing
young children to learn language far more easily than cogni-
tively mature adults and older children. Nevertheless, this is
a highly controversial and vital topic for research. Advances
in brain imaging technologies are generating new findings
all the time.
Many researchers do not believe in the existence of the
critical period for language learning at all. Instead, they point
to social and environmental factors, attitude, motivation,
input and interaction, as well as pedagogical effect, and lan-
guage policy. Individuals learn languages for very different
reasons and with very different motivations: some learn new
languages because they are married to speakers of other lan-
guages, while others seek employments that require the use of
certain languages. Community attitudes toward the language
being learned can also have a profound impact on the individ-
ual learners. Where the community has a broadly negative view
of the target language and its speakers, or a negative view of its
relation to them, learning is typically much more difficult.
At the same time, the learner's direct contact with the target
language, that is, input, has a crucial impact on the learning
process and learning outcome.
Generally speaking, the amount of input a learner receives
is one of the most important factors affecting his or her
learning. However, it must be at a level that is comprehensible
to the learner. In his study of second language learning,
Stephen Krashen advanced the idea that language input should
be at the 1+1' level, that is, just beyond what the learner
can fully understand. In other words, the input is comprehen-
sible, but contains structures that are not yet fully understood.
While there has been criticism of Krashen's theory that factors
other than structural difficulty (such as interest or presenta-
tion) can affect the actual conversion of input into intake, the
i + 1 idea has been found useful in language teaching.
In contrast, Merrill Swain proposed an Output Hypothesis
that meaningful output is as necessary to language learning as
meaningful input. However, most studies have shown little if
any correlation between learning and quantity of output. It is
likely that meaningful output is important to language
learning because the experience of producing language leads
to more effective processing of input.
A rather different perspective on success or failure of lan-
guage learning is the so-called Interaction Hypothesis, which
proposes that language acquisition is strongly facilitated by
the use of the target language in interaction. In particular, the
negotiation of meaning has been shown to contribute greatly
to the acquisition of vocabulary.
In addition to the Input Hypothesis, Krashen also pro-
posed an Affective Filter Hypothesis, which claims that the
relative success or failure of language learning is often filtered
through motivation, attitude, and other affective factors. Many
researchers have suggested that more motivated learners tend
to be more successful. Motivation involves four key aspects:
1 . A Goal
2. An Effort
3. A Desire to attain the goal
4. Favorable attitude toward the activity in question.
It has been suggested that good L2 learners have the
following characteristics. They
• are willing and accurate guessers;
• have a strong drive to communicate;
• are often uninhibited, and if they are, they combat
inhibition by using positive self-talk, by extensive use of
practicing in private, and by putting themselves in situa-
tions where they have to participate communicatively;
• are willing to make mistakes;
• focus on form by looking for patterns and analyzing;
• take advantage of all practice opportunities;
• monitor their speech as well as that of others; and
• pay attention to meaning.
In short, they are more motivated and have a more positive
attitude toward learning the language.
Motivation is a desire to achieve a goal, combined with the
energy to work toward that goal. It determines the extent of
active, personal involvement in L2 learning.
Sometimes, a distinction is made between positive and
negative motivation. Positive motivation is a response which
includes enjoyment and optimism about the tasks that you are
involved in, whereas negative motivation involves undertaking
tasks for fear that there should be undesirable outcomes. Ques-
tions then arise as to where positive and negative motivations
came from, the answer to which lies largely in individuals'
personal experience. A person growing up in a multilingual
environment where the majority of people he or she interacts
with on a routine basis speak a variety of different languages
freely and where there is no societal control over which lan-
guage should be used in what manner would, at least in prin-
ciple, have a more positive attitude toward learning new
languages and using different languages. In contrast, a person
who has been forced to learn or to drop a particular language
or experienced drama in a particular language may develop a
negative motivation. The links between personal experience
and motivation and other affective and emotional aspects of
language learning are an area worthy of further investigation.
Other factors that researchers have examined in terms of
success or failure of becoming bilingual and multilingual
include gender and personality. These factors are at the inter-
face between learners' internal and external dimensions. Evi-
dence from neuroscience suggests that boys and girls use
different parts of their brains to process some basic aspects of
grammar. Sociolinguists have shown that men and women use
languages in different ways for communicative purposes. How-
ever, there is very little research into the gender differences in
bilingual or second language acquisition. Some researchers
have suggested that there are gender differences in motivations
for language learning. It certainly seems to be the case that
more girls than boys prefer to learn modern languages in
schools and have somewhat better attainment in school-
based language examinations. Yet none of the apparent differ-
ence can be solely attributed to biological differences between
the two genders. It is more likely to be a sociocultural effect in
that language learning and use is closely related to the different
Bilingualism and Multilingualism 341
roles of girls and boys and women and men in society. Simi-
larly, some personality traits have been linked to the relative
success or failure in second or additional language learning.
Extraverts or unreserved and outgoing people tend to acquire a
second language better than introverts or shy people. But extra-
version and introversion are sociocultural constructs, rather
than biologically determined.
What Does it Mean to Be Bilingual or Multilingual?
In theory, these different routes to bilingualism would result in
different language behaviors and cognitive organizations of the
bilingual brain. But at the moment, there is very little system-
atic research linking the process of bilingual acquisition with
the outcome of bilinguality at the individual level. What we do
know, however, is that the different languages being acquired
by the child do not always develop at the same speed or to the
same level, resulting in different configurations of language
dominance in individual children and different proficiencies
in the languages.
In an early study, Uriel Weinreich proposed three types
of bilinguals representing three types of relationship between
the linguistic sign (or signifier) and the semantic content
(signified). In Type A, the individual combines a signifier from
each language with a separate unit of signified. Weinreich called
them 'coordinative' (later often called 'coordinate') bilinguals.
In Type B, the individual identifies two signifiers but regards
them as a single compound, or composite, unit of signified;
hence they are called 'compound' bilinguals. Type C relates to
people who learn a new language with the help of a previously
acquired one. They are called 'subordinative' (or 'subordinate')
bilinguals. His examples were from English and Russian:
‘book’ ‘kniga’
(a) /buk J /kn’iga/
‘book’ = ‘kniga’
(b) /buk / /kn’iga/
‘book’
/buk/
(c) /kn’iga/
Weinreich's typology is often misinterpreted in the litera-
ture as referring to differences in the degree of proficiency in
the languages. But it is a fact that the relationship between
the language proficiency and cognitive organization of the
bilingual individual, as conceptualized in Weinreich's model,
is far from clear. Some 'subordinate' bilinguals demonstrate a
very high level of proficiency in processing both languages,
as evidenced in grammaticality and fluency of speech, while
some 'coordinative' bilinguals show difficulties in processing
two languages simultaneously (i.e., in code-switching or in
'foreign' words identification tasks). It must also be stressed
that, using Weinreich's distinctions, bilingual individuals are
distributed along a continuum from a subordinate or com-
pound end to a coordinate end, and can at the same time be
more subordinate or compound for certain concepts and more
coordinate for others, depending on, among other things,
the age and context of acquisition. There are now a number
of psycholinguistic models of the bilingual lexicon, including
the Concept Mediation Model, the Word Association model,
and the Revised Hierarchical Model. These models take into
consideration factors such as proficiency level, age, and context
of acquisition and have much more explanatory power than
the earlier models for bilingual language use.
In terms of bilingual and multilingual first language acqui-
sition (BAMFLA), that is, acquiring multiple languages from
birth, both linguists and psychologists have spent a consider-
able amount of time investigating the developmental patterns
of bilingual and multilingual children. Earlier studies sug-
gested that bilingual acquisition went through three key stages:
Stage I: The child has one lexical system comprising words
from both languages.
Stage II: The child distinguishes two different lexicons, but
applies the same syntactic rules to both languages.
Stage III: The child speaks two languages that are different both
in lexicon and syntax, but each language is associated with
the person who uses that language.
Although some studies have produced evidence supporting
the model, there has been much criticism, particularly of the
claims made regarding the first two stages. This is generally
known as the 'one-system-or-two' debate, that is, whether
bilingual children begin with a fused linguistic system and
gradually differentiate the two languages, or whether they
start with a differentiated system. Part of this debate centers
around the question: What counts as evidence for differentia-
tion or fusion? Some researchers based their decision on
whether the child made appropriate sociolinguistic choices,
that is, whether the child spoke the 'right' language to the
'right' person. It was argued that awareness of the two lan-
guages as distinct plays a crucial role in deciding the issue of
differentiation, and a child's ability to make appropriate lan-
guage choices reflects that awareness. However, the argument
that bilingual children separate the languages when they are
aware that there are two systems is circular, unless some crite-
rion is provided for assessing whether awareness means any-
thing other than children separating the languages. In any case,
we need to bear in mind that a child's apparent (in)ability to
choose the right language for the right addressee is a rather
different issue from whether the child has one or two linguistic
systems. Part of the problem is the familiar one of what we can
infer about competence from performance. There now exists a
large body of literature rebutting the 'fused' system hypothesis,
arguing instead that bilinguals have two distinct but interde-
pendent systems from the very start.
Is bilingual acquisition the same as monolingual acquisi-
tion? Theoretically, separate development is possible without
342 Bilingualism and Multilingualism
there being any similarity with monolingual acquisition. Most
researchers argue that bilingual children's language develop-
ment is by and large the same as that of monolingual children.
In very general terms, both bilingual and monolingual chil-
dren go through an initial babbling stage, followed by the one-
word stage, the two-word stage, the multiword stage, and the
multiclause stage. At the morphosyntactic level, a number of
studies have reported similarities rather than differences
between bilingual and monolingual acquisition. Nevertheless,
one needs to be careful in the kinds of conclusions one draws
from such evidence. Similarities between bilingual and mono-
lingual acquisition do not mean that ( 1 ) the two languages a
bilingual child is acquiring develop in the same way or at the
same speed; (2) the two languages a bilingual child is acquiring
do not influence and interact with each other. For example,
studies have found that while 2-3-year-old French-English
bilingual children displayed patterns that characterize the per-
formance of monolingual children acquiring these languages
separately and they acquired these patterns within the same age
range as monolingual children, they (1) used finite verb forms
earlier in French than in English; (2) used subject pronouns in
French exclusively with finite verbs but subject pronouns
in English with both finite and nonfinite verbs, in accordance
with the status of French clitic subject pronouns (or agreement
markers) as against full NPs in English; and (3) placed verbal
negatives after lexical verbs in French (e.g., n'aime pas ) but
before lexical verbs in English ( do not like).
There is one area in which bilingual children clearly differ
from monolingual children, namely, code-mixing. Studies
show that bilingual children mix elements from both lan-
guages in the same utterance as soon as they can produce
two-word utterances. As with adult code switching, bilingual
children's language mixing is highly structured. The operation
of constraints based on surface features of grammar, such as
word order, is evident from the two-word/-morpheme stage
onward, and the operation of constraints based on abstract
notions of grammatical knowledge is most evident in bilingual
children once they demonstrate such knowledge overtly (e.g.,
verb tense and agreement markings), usually around 2.6years
of age and older. These findings suggest that, in addition to the
linguistic competence to formulate correct monolingual
strings, bilingual children have the added capacity to coordi-
nate their two languages online in accordance with the gram-
matical constraints of both languages during mixing. While
these studies provide further evidence for the separate develop-
ment, or two-systems, argument, they also suggest that there
are both quantitative and qualitative differences between bilin-
gual acquisition and monolingual acquisition.
The ability to switch from one language to another turns
out to be a crucial aspect of being bilingual and multilingual.
Indeed, one cannot tell who is or who isn't a bilingual until the
person uses two or more languages in the same conversation.
There is a widespread impression that bilingual speakers code-
switch because they cannot express themselves adequately
in one language. This may be true to some extent when a
bilingual is momentarily lost for words in one of his or her
languages. However, code-switching is an extremely common
practice among bilinguals and takes many forms. A long nar-
rative may be divided into different parts which are expressed
in different languages; sentences may begin in one language
and finish in another; words and phrases from different lan-
guages may succeed each other. Linguists have devoted much
attention to the study of code-switching. It has been demon-
sttated that code-switching involves skilled manipulation of
overlapping sections of two (or more) grammars, and that
there is virtually no instance of ungrammatical combination
of two languages in code-switching, regardless of the bilingual
ability of the speaker. Some suggest that code-switching is itself
a discrete mode of speaking.
One important aspect of code-switching is that the two
languages involved do not play the same role in sentence
making. Typically, one language sets the grammatical frame-
work, with the other providing certain items to fit into the
framework. Code-switching therefore is not a simple combina-
tion of two sets of grammatical rules but grammatical integra-
tion of one language in another. Bilingual speakers of different
proficiency levels in their two languages or speaking two typo-
logically different languages can engage in code-switching and
indeed vary it according to their needs.
There is a group of bilinguals who engage themselves in
cross-modality language production. This is the case with
speech-sign bilinguals who, in addition to the oral modality,
use the manual-visual modality in everyday communication.
They are special in one aspect: that is, the two different
modalities allow for the simultaneous production of the two
languages. In other words, one can speak and sign at the same
time. Research has shown that such simultaneous bimodal
production is typically exemplified by the use of lexical items
from both languages but only one set of grammatical rules,
which is usually from the spoken language. Right now, we
know relatively little about how the two linguistic systems
interact in the language production and processing of
speech-sign bilinguals. Indeed, much more work needs to be
undertaken before we can fully appreciate the complexity of
the language behavior of bilinguals in general.
It should be emphasized that code-switching is not only an
effective means of communication but also an act of identity.
Every time bilingual speakers say something in one language
when they might just as easily have said it in another, they are
reconnecting with people, situations, and power configura-
tions from their history of past interactions and imprinting
on that history their attitudes toward the people and languages
concerned. Through language choice, they maintain and
change ethnic group boundaries and personal relationships,
and construct and define 'self and 'other' within a broader
political economy and historical context. So, the issue of lan-
guage use of bilinguals and multilinguals becomes an issue of
identity and identification.
The notion of identity has gone through considerable
change in sociolinguistics. In the earlier variationist sociolin-
guistic work, identity was taken to mean the speaker's socioeco-
nomic class, gender, age, or place of origin. It is assumed that
speakers express, rather than negotiate, identities through their
language use. Several scholars later criticized such assumptions
and argued instead that identities are negotiated through social
interaction. Linguistic forms and strategies have multiple
functions and cannot be directly linked to particular identities
outside of interactional contexts. More recent studies by socio-
linguists have emphasized the negotiation of identities and
demonstrated that identities are local constructions.
Bilingualism and Multilingualism 343
Parallel to the work on multilingualism and negotiation of
identities, sociolinguists critically examine some of the con-
cepts and notions commonly used by other researchers in the
field of bilingualism and multilingualism. For example, the
very idea of code-switching raises questions as to what a lan-
guage is. Instead of thinking of languages as discrete systems,
sociolinguists tend to see multilingual speakers as actors of
social life who draw on complex sets of communicative
resources which are unevenly distributed and unevenly valued.
The linguistic systematicity therefore appears to be at least
as much a function of historically rooted ideologies (of nation
and ethnicity) and of the ordering practices of social life as
of language per se. This perspective goes beyond a focus on
mental representation of linguistic knowledge and opens up
the possibility of looking at bilingualism and multilingualism
as a matter of ideology, communicative practice, and social
process.
This particular sociolinguistic perspective has important
implications for the way researchers collect, analyze, and
interpret data. Informed by developments in anthropology,
sociology, and cultural studies, sociolinguists have examined
communicative practices within and across sites that can be
ethnographically demonstrated to be linked. Working with the
ideas of trajectories (of speakers, linguistic resources, discourses,
institutions) across time and space and of discursive spaces
which allow for, and also constrain, the production and circu-
lation of discourses, sociolinguists have examined multilin-
gual practices in a number of communities and argued that
multilingual practices contribute to the construction of social
boundaries and of the resources those boundaries regulate.
They therefore also raise the question of the social and histori-
cal conditions that allow for the development of particular
regimes of language, for their reproduction, their contestation,
and, eventually, their modification or transformation.
A further, closely related, area in which sociolinguists have
extended the work by linguists and psycholinguists on bilin-
gualism is that of the acquisition of linguistic knowledge.
Building on earlier research on language socialization, which
focused on young children acquiring their first language in
culturally specific ways, they have examined bilingual and
multilingual children's developing competence in various
speech and literacy events. Particular attention is given to the
range of linguistic resources available, or not, in bilingual and
multilingual communities and the ways in which children, as
well as adolescents and adults, learn to choose among these
resources for their symbolic value. The researchers emphasized
language socialization as an interactive process, in which those
being socialized also act as agents rather than as mere passive
initiates. This line of inquiry also demonstrates how domains
of knowledge are constructed through language and cultural
practices, and how the individual's positioning affects the pro-
cess of knowledge acquisition and construction.
Conclusion
Increased contacts with peoples speaking different languages as
a result of globalization mean that bilingualism and multilin-
gualism are becoming more commonplace a phenomenon for
today's world. People are more aware of the linguistic diversity
around them. As the article shows, there is a considerable
amount of knowledge related to bilingual and multilingual
individuals, the manner in which they acquire languages, and
the way they use them in daily life. Paradoxically, however,
much of the research still uses the monolingual as the yardstick
and often sees bilingualism as a deviation from the norm.
Many issues also require further empirical research: for exam-
ple, the organization of the bilingual brain; the structure and
organization of the bilingual's different languages; the various
processing operations involved in the perception, production,
and memorization of language when the speaker is using
one or all his or her languages; and the linguistic and psycho-
linguistic differences and similarities between code-switching,
borrowing, and interference. Advances in research technolo-
gies will clearly help the field to develop more sophisticated
methods. Research findings will have impacts on policy and
practice regarding the bilingual and multilingual individuals as
well as the communities they form and live in.
Further Reading
Bialystok E (2001 ) Bilingualism in Development: Language, Literacy, and Cognition.
New York: Cambridge University Press.
De Houwer A (2009) Bilingual First Language Acquisition. Bristol, UK: Multilingual
Matters.
Genesee F (1989) Early bilingual language development: One language or two 1 Journal
of Child Language 16: 161-179.
Krashen SD (1 985) The Input Hypothesis: Issues and Implications. New York:
Longman.
Krashen SD (2003) Explorations in Language Acquisition and Use. Portsmouth, NH:
Heinemann.
Kroll J and Stewart E (1984) Category interference in translation and picture naming.
Journal of Memory and Language 33: 1 49-1 74.
Lenneberg EH (1967) Biological Foundations of Language. New York: Wiley.
Li W (2007) The Bilingualism Reader, 2nd edn. London: Routledge.
Long MH (1981) Input interaction and second language acquisition. Annals of
the New York Academy of Sciences 379: 259-278.
Meisel JM (1989) Early differentiation of languages in bilingual children. In:
Hyltenstam K and Obler L (eds.) Bilingualism Across the Lifespan: Aspects of
Acquisition, Maturity, and Loss, pp. 13-40. Cambridge: Cambridge University
Press.
Potter MC, So K-F, Von Echardt B, and Feldman LB (1984) Lexical and conceptual
representation in beginning and more proficient bilinguals. Journal of Verbal
Learning and Verbal Behavior 23: 23—38.
Romaine S (1989/1995) Bilingualism, 2nd edn. Oxford: Blackwell.
Singleton D and Lengyel 1 (1995) The Age Factor in Second Language Acquisition.
Clevedon, UK: Multilingual Matters.
Swain M and Lapkin S (1995) Problems in output and the cognitive processes
they generate: A step toward language learning. Applied Linguistics
16: 371-391.
Volterra V and Taeschner T (1978) The acquisition and development of language by
bilingual children. Journal of Child Language 5: 311-326.
Weinreich U (1 953) Languages in Contact: Findings and problems. New York: The
Linguistic Circle of New York; Reissued by Mouton in The Hague, 1968.
Biofeedback Therapy
K W Jung, D-H Yang, and S-J Myung, Asan Medical Center, Seoul, South Korea
© 2012 Elsevier Inc. All rights reserved.
Glossary
Classical conditioning The formation of a
conditioned reflex through the temporal paring of an
unconditioned stimulus, which reflexively elicits a
response, and a neutral conditioned stimulus. After the
pairing, the formerly neutral conditioned stimulus
acquires the ability to elicit a response similar to
the one elicited by the unconditioned
stimulus.
Operant conditioning The presentation of a stimulus
dependent on the occurrence of a response for the purpose
of modifying the strength or the frequency of the response.
Reinforcer In operant conditioning, any stimulus when
presented or removed depending on the occurrence of a
response results in the modification of the response.
In a classical conditioning, the presentation of a conditioned
stimulus and an unconditioned stimulus is in a close
temporal proximity.
Underlying Learning Theory and Biologic Mechanism
Learning Theory
Learning theories have traditionally distinguished between two
types of learning. One type, called classical conditioning, in-
volves the pairing of two stimuli in a close temporal proximity.
One of the two stimuli, known as the unconditioned stimulus,
normally elicits some reflex behavior such as salivation, eye
blink, or a change in skin conductance. The other stimulus,
known as the conditioned stimulus, is neutral prior to the
pairing but acquires the ability to elicit a response similar to
the unconditioned response after the pairing process. The sec-
ond type of learning recognized by learning theorists is known
as operant conditioning. In operant conditioning, the response
is generally thought of as emitted on a more or less voluntary
basis, rather than being reflexively elicited by any particular
stimulus. Also, the operant conditioning process involves the
pairing of the stimulus dependent on the occurrence of a
response. The paired stimulus leads to an increase in the fre-
quency or the magnitude of the response and it is known as a
negative reinforce if its removal leads to an increase in the
frequency or the magnitude of the response.
Between 1928, when the two types of learning were for-
mally distinguished, and the early 1960s, it was generally
assumed that operant and classical conditioning were mutu-
ally exclusive processes. All voluntary skeletal motor behaviors
were assumed to be subject to operant conditioning, while
visceral and glandular responses mediated by the autonomic
nervous system were assumed to be involuntary and modifi-
able only through classical conditioning. It was thought
impossible that salivation to a light stimulus could be oper-
antly reinforced by food. This view was retained until the early
1960s when groups of researchers in the Soviet Union,
Canada, and the United States began a series of experiments
designed to show that responses mediated by the autonomic
nervous system can, in fact, be brought under a voluntary
control through techniques that closely resemble operant
conditioning.
Since the first suggestion regarding instrumental learning
and operant conditioning, it demonstrated the medical poten-
tial of biofeedback by showing that the normally involuntary
autonomic nervous system can be operantly conditioned with
the use of an appropriate feedback. By means of instruments,
patients acquire information about the status of involuntary
biological functions such as skin temperature and electrical
conductivity, muscle tension, blood pressure, heart rate, and
brain wave activity. Patients, then, learn to regulate one or
more of these biological states that affect symptoms.
Biological Mechanisms
Biofeedback makes physiologic change possible by means of
operant conditioning or trial-and-error learning, in which a
response is learned and performed depending on whether
that response is followed by reinforcement. For biofeedback
to be useful, four conditions must be satisfied. First, there must
be a readily detectable and measurable response such as the
bladder pressure or the pelvic floor muscle activity. Second,
there must be variability in that response with a detectable
change as opposed to total paralysis. Third, there must be a
perceptible cue such as the sensation of urgency that indicates
to the patient when control should be performed. Fourth,
because biofeedback is based on learning, it requires the active
involvement of a motivated patient.
Taking basketball as an example, the information about
where the ball went with each shot is called feedback. If both
the coach and the novice were blindfolded, there would be no
feedback, and the novice would not learn to improve. Some
tense patients may not be aware of the fact that certain muscles
are tense, and patients with neuromuscular disorders, or even
their therapists, may not be able to discriminate small increases
in the activity of paralyzed muscles or decreases in the contrac-
tions of spastic ones. They are like a blindfolded basketball
novice and coach. However, a display of the electrical activity
of the muscles (the EMG) can remove the blindfold and give
them a better feedback about what the muscles are doing. This
feedback might be in the form of a series of auditory clicks
that get faster when the muscle contracts more or a tracing of
the activity of one or more muscles as a dot draws the curve
of a graph on a TV screen. Feedback from a measuring instru-
ment that yields moment-to-moment information about a
biological function is called biofeedback.
344
Biofeedback Therapy 345
One of the advantages of biofeedback is that it allows small
changes in the correct direction to be noticed and rewarded
as success so that they gradually can be built up into larger
changes. Eventually, patients learn to perceive these changes
without the measuring instrument so that they can practice by
themselves. Biofeedback should be especially effective in those
cases where the patients cannot perceive their initial small
correct responses or even may have the wrong perception of
what they are doing.
Other advantages of biofeedback are that by making the
early signs of slight progress conspicuous, it can encourage
and motivate the patients, relieve their sense of helplessness,
and serve as a coping response to reduce symptoms of stress.
Instead of having something done to the patients, it teaches
them to do something for themselves, increasing their confi-
dence, or what has been called self-efficacy. This factor is
particularly important when biofeedback is being used to
treat symptoms that are elicited or aggravated by stress.
Early Development
Experiments on Humans
It was first shown that human subjects could learn to voluntar-
ily dilate blood vessels of the finger in order to avoid or escape
an electric shock if they were provided with visual information
about their vasomotor activity. Subjects provided with an
amplified auditory representation of their heart beat also were
trained to accelerate their heart rate to avoid an electric shock
to the ankle. Other responses generally considered involuntary
were brought under voluntary control by providing subjects
with feedback information about the targeted response and
some form of operant reinforcement. Studies conducted in
the 1960s also demonstrated operant conditioning of the skin
conductance response and brain wave activity. Voluntary con-
trol over the firing of single motor units was demonstrated by
providing subjects with auditory and visual displays of individ-
ual myoelectric potentials recorded from fine wire intramuscu-
lar electrodes.
Experiments on Animals
At about the same time that voluntary control of autonomically
mediated behavior was being demonstrated in humans, paral-
lel lines of research were in progress using various animal
species. For example, thirsty dogs were trained to increase and
decrease the flow of saliva using water as a reinforcer. Bidirec-
tional changes in heart rate and blood pressure were also shown
to be subject to operant conditioning in monkeys and rats.
Methods
The feedback instrument used depends on the patient and the
specific problem. The most effective instruments are the elec-
tromyogram (EMG), which measures the electrical potentials
of muscle fibers; the electroencephalogram (EEG), which mea-
sures alpha waves that occur in a relaxed state and the thermis-
tor, which measures skin temperature (which drops during
tension because of peripheral vasoconstriction). Recently,
fMRI (functional magnetic resonance imaging) has been used
to identify both positive and negative emotional stimuli for
stress management (neurofeedback).
Patients are attached to one of the instruments that measure
a physiological function and translate the measurement into
an audible or visual signal that the patients use to gage their
responses. For example, in the treatment of bruxism, an EMG is
attached to the masseter muscle. The EMG emits a high tone
when the muscle is contracted and a low tone when at rest.
Patients can learn to alter the tone to indicate relaxation.
Patients receive feedback about the masseter muscle, the tone
reinforces the learning, and the condition ameliorates, with all
of these events interacting synergistically.
Many less specific clinical applications (e.g., treating insom-
nia, dysmenorrhea, and speech problems; improving athletic
performance; treating volitional disorders; achieving altered
states of consciousness; managing stress; and supplementing
psychotherapy for anxiety associated with somatoform disor-
ders) use a model, in which frontalis muscle EMG biofeedback
is combined with thermal biofeedback and verbal instructions
in progressive relaxation.
Relaxation Therapy
Learning relaxation, therefore, involves cultivating a muscle
sense. To develop the muscle sense further, patients are taught
to isolate and contract specific muscles or muscle groups, one
at a time. For example, patients flex the forearm while the
therapist holds it back to observe tenseness in the biceps mus-
cle. Patients are repeatedly reminded that relaxation involves
no effort. In fact 'making an effort is being tense and therefore
is not to relax.' As the session progresses, patients are instructed
to let go further and further, even past the point when the body
part seemed perfectly relaxed.
Patients would work in this fashion with different muscle
groups, often over more than 50 sessions. Patients were also
frequently left alone while the therapist attended to other
patients.
Neurofeedback
Neurofeedback allowed, for the first time, voluntary self-
regulation of brain activity through feedback and reward.
Expectancies ran high and many premature announcements
of clinical success based on single case studies or uncontrolled
observations discredited the field early on. In the 1970s,
Miller's demonstrations of operant control of autonomic
(and central nervous system (CNS)) functions in curarized
rats, supposedly proving 'voluntary' operant regulation of
many bodily functions excluding mediation of the motor sys-
tem through curarization, turned out to be difficult to replicate.
Together with the clinical overstatements in the field of bio-
feedback, this historic incident virtually halted funding from
the public sources and blocked large controlled clinical studies
despite some indications of its efficiency. However, more recent
studies suggested that some patients with drug-resistant epilepsy
(mostly with secondarily generalized seizures) experienced a
reduction in the number of ictal events during and after training
consisting of self-regulation of slow cortical potentials (SCPs),
an effect also reported using biofeedback of skin conductance
346 Biofeedback Therapy
responses (GSR). Nagai et al. showed that learned increase in
autonomic arousal through reduction of skin conductance
decreased negative SCPs at the cortical level and thus increased
seizures thresholds confirming earlier reports.
The clinical target populations for brain-machine interface
(BMI) treatment consist primarily of patients with amyo-
trophic lateral sclerosis (ALS) and severe CNS damage includ-
ing spinal cord injuries and stroke, resulting in substantial
deficits in communication and motor function. However, an
extensive body of literature started in the 1970s using neuro-
feedback training. Such training implemented to control vari-
ous EEG-measures provided solid evidence of positive effects
in patients with otherwise pharmacologically intractable epi-
lepsy, attention deficit disorder, and hyperactivity ADHD.
More recently, the successful introduction and testing of real-
time fMRI and a near-infrared spectroscopy (NIRS)-BMI has
opened an exciting field of interest in patients with psycho-
pathological conditions.
Clinical Applications
Gastrointestinal Diseases
In the field of gastroenterology, an initial study focused on
biofeedback applications in patients with fecal incontinence.
Later, biofeedback applications for the treatment of chronic
constipation due to pelvic floor dyssynergia (anismus, spastic
pelvic floor syndrome, and puborectalis paradox), had been
widely performed by pediatric and adult doctors. The goal of
biofeedback training in gastroenterology is to restore a normal
pattern of defecation. In patients with dyssynergic defecation,
the goal of neuromuscular training is twofold: to correct the
dyssynergia in coordination of the abdominal, rectal, and anal
sphincter muscles to achieve a normal and complete evacua-
tion; and to enhance rectal sensory perception in patients with
impaired rectal sensation. This training consisted of improving
the abdominal push efforts (diaphragmatic muscle training)
together with manometric-guided pelvic floor relaxation fol-
lowed by simulated defecation training.
The symptomatic improvement rate has varied between
44% and 100% in recent randomized controlled trials of
adults with dyssynergic defecation. All of these studies have
concluded, however, that biofeedback therapy is superior to
controlled treatment approaches, such as diet, exercise, and
laxatives, or use of polyethylene glycol, diazepam, or placebo,
balloon defecation therapy, or sham feedback therapy.
Recently published meta-analysis showed that biofeedback
in fecal incontinence was equally effective as nonbiofeed-
back therapy. For constipation, more than threefold superior-
ity of biofeedback to nonbiofeedback, but equal efficacy of
EMG biofeedback to other applications was shown. Other
indications, including chronic proctalgia, irritable bowel syn-
drome, and functional dyspepsia, still need more data with
prospectively designed randomized controlled studies.
Urinary Diseases
Biofeedback is a very specific treatment that can restore bladder
control by teaching patients to modulate the mechanisms of
continence using a pelvic perineal reeducation technique,
which uses a signal that is usually visual, but may be auditory,
which allows the patient with an anorectal or perineal func-
tional anomaly to understand and correct the ttouble. Bio-
feedback has now gained several potential applications for
urologic conditions, having been successfully used for patients
with urologic disorders such as detrusor instability, detrusor
sphincter dyssynergia, and enuresis. In a randomized trial
involving older women with urgent urinary incontinence,
the percent reduction in incontinence episodes was similar
after 8 weeks of behavioral therapy with biofeedback ther-
apy, after 8 weeks of behavioral therapy without biofeedback
therapy, and after self-administered behavioral treatment
according to a self-help booklet.
Enuresis describes intermittent incontinence at night and
refers to any discrete leakage of urine at night. Therefore,
dysfunctional voiding has been used as a broad term. The
reason of dysfunctional voiding in children could include
hyperactive pelvic floor, overactive bladder, incontinence, or
lazy bladder syndrome. For children with hyperactive pelvic
floor, biofeedback therapy focuses on the relaxation and a
return of normal flow. For children with overactive bladder
and incontinence, use of the guarding reflex of the pelvic floor
musculature can maintain continence during uninhibited con-
tractions. In addition, biofeedback can be used for these
children to void optimally and ensure that there is no decom-
pensation of the detrusor muscle that occurs with the lazy
bladder syndrome.
Cardiovascular Diseases
Since the beginning of biofeedback in the 1970s, biofeedback
was used to conttol heart rate and blood pressure in patients
diagnosed with cardiovascular diseases. It remains unclear
whether the biofeedback regarding heart rate variability has
more or less potential than other types of biofeedback in
patients with cardiovascular disease, but these preliminary
observations suggest that it may be useful in improving symp-
toms and the quality of life. Regarding hypertension, blood
pressure can be lowered by 6-10mmHg when biofeedback is
effective, which is less of an effect than what is observed with
most antihypertensive medications. Thus, although biofeed-
back has a potential in hypertension, its efficacy is not well
proven and the systemic trials are lacking.
Neurologic Diseases
Biofeedback methods in the treatment of migraine are based on
the concept that various relaxation skills, including diaphrag-
matic breathing or visualization, were helpful to induce the
relaxation response, which includes relaxation of the sympa-
thetic nervous system and activation of the parasympathetic
nervous system. Recently published meta-analysis showed
that biofeedback can be supported as an efficacious treatment
option for migraine and tension-type headache according to
the Association for Applied Psychophysiology and Biofeedback
and the International Society for Neurofeedback and Research
criteria.
Epileptic patients who were refractory to anticonvulsant
medications have been treated with biofeedback therapy
based on EEG. Several positive outcomes were reported
Biofeedback Therapy 347
especially regarding complex-partial seizures, an extremely
difficult-to-treat seizure type. However, it still needs a rando-
mized, controlled study to prove its efficacy.
Others
Early studies in patients with Raynaud's syndrome demon-
strated that biofeedback alone or in combination with various
procedures reduced the occurrence of vasospastic episodes,
based on the concept that cold hands and cold feet are frequent
concomitants of anxiety and also occur in Raynaud's syn-
drome. Exposure to cold during temperature biofeedback
may enhance the effects of biofeedback alone.
Fibromyalgia syndrome and other myofacial pain had been
treated with biofeedback therapy. Recently randomized con-
trolled study regarding fibromyalgia based on EEG showed the
significant efficacy by a significantly decreased theta/sensory
motor rhythm ratio by EEG waves compared to baseline.
Biofeedback therapy based on EMG was also tried in treat-
ing bronchial asthma. Children who were instructed to reduce
frontalis muscle tension via biofeedback and continue their
strategies in the home setting showed improvement in pulmo-
nary function testing compared with children who were
instructed to maintain frontalis tension.
Further Reading
Ballek NK and McKenna PH (2010) Lower urinary tract dysfunction in childhood.
The Urologic Clinics of North America 37(2): 215-228.
Basmajian JV (1989) Biofeedback: Principles and Practice for Clinicians, 3rd edn.
Baltimore: Williams & Wilkins.
Birbaumer N, Ramos Murgialday A, Weber C, and Montoya P (2009) Neurofeedback
and brain-computer interface clinical applications. International Review of
Neurobiology 86: 107-117.
Chiarioni G and Whitehead WE (2008) The role of biofeedback in the treatment of
gastrointestinal disorders. Nature Clinical Practice. Gastroenterology & Hepatology
5: 371-382.
Enck P, Van Der Voort IR, and Klosterhalfen S (2009) Biofeedback therapy in fecal
incontinence and constipation. Neurogastroenterology and Motility 21:11 33-1 1 41 .
Miller NE (1978) Biofeedback and visceral learning. Annual Review of Psychology
29: 373-404.
Moravec CS (2008) Biofeedback therapy in cardiovascular disease: Rationale and
research overview. Cleveland Clinic Journal of Medicine 75: S35-S38.
Nestoriuc Y, Martin A, Rief W, and Andrasik F (2008) Biofeedback treatment for
headache disorders: A comprehensive efficacy review. Applied Psychophysiology
and Biofeedback 33: 125-140.
Nygaard I (2010) Clinical practice. Idiopathic urgency urinary incontinence.
New England Journal of Medicine 363(12): 1156-1162.
Peters TE and Fritz GK (2010) Psychological considerations of the child with asthma.
Child and Adolescent Psychiatric Clinics of North America 19(2): 319-333.
Rao SC (2008) Dyssynergic defecation and biofeedback therapy. Gastroenterology
Clinics of North America 37: 569-586.
Sterman MB (2010) Biofeedback in the treatment of epilepsy. Cleveland Clinic Journal
of Medicine 77: S60-S67.
Bipolar Disorder
J D Pettigrew, University of Queensland, St Lucia, QLD, Australia
© 2012 Elsevier Inc. All rights reserved.
Glossary
Depression Depression is complementary to mania, with
an extended period of depressed mood and increased sleep
time accompanied by reduced speech, lack of motivation,
and excessive caution. The very high risk of suicide in bipolar
disorder is most apparent at the time when there is a switch
between mania and depression. While there are many
different accounts of the mechanism of the cyclicity and the
opposite polarity of the two phases, the only compelling
theory involves an interhemispheric switch that alternates
between the complementary cognitive styles of the two
hemispheres and becomes stuck on one side during an
episode.
Fractal time A function (after Mandelbrot) of time that
shows self-similarity, such that a change at one temporal
scale affects all temporal scales. One of the first examples
suggesting fractal time in the nervous system came from
work on the neural clocks of Drosophila. A mutant fly with an
altered per gene had a longer circadian period than normal
(25-30 h), but was also found unexpectedly to have a longer
period of the courtship rhythm, which is measured in
seconds. The underlying mechanism of this finding is still
not clear, despite much progress on the understanding of
neural clocks, but there are now many more examples of
fractal time in the nervous system. For example, humans
with a prolonged perceptual rivalry cycle also have
prolongation of other cycles, such as the nasal cycle, which
may be many hours long instead of the normal 1.5 h. The
concept of fractal time could explain why the prolonged
perceptual rivalry cycle of bipolar disorder (8-10s instead of
l-2s) is also associated with much longer lasting alterations
(>days) in mood.
Interhemispheric switch A neural mechanism that switches
activation from one cerebral hemisphere to the opposite
hemisphere. Because of the speed of the switch, it is difficult
to visualize it in scanning studies such as fMRI that have
time constants around 8- 10s. The switch is present in all
vertebrate brains, so its location and physiology can be
determined by comparative studies. Neuropharmacological
studies have further defined the properties of the switch.
Activation of the cerebral hemispheres is achieved by
dopamine, which produces an 'up' physiological state of
activation and readiness for plastic change via the NMDA
channels that are turned on. The rhythmicity of the switch is
determined by serotonin input using serotonin 1A receptors,
whose activation slows the rate of the switch. The most likely
physical correlate of the switch in the human brain is the
dopaminergic ventral tegmental area (VTA) in the ventral
midbrain, with the rhythmic serotonin input coming from
the raphe nucleus.
Mania An extended period of elevated mood and
sleeplessness. Lack of sleep is not accompanied by tiredness,
and perceptual rivalry studies of mania suggest that there is a
primary sleep-wakefulness alteration, rather than one
secondary to the mania, mediated by increased time
spent by the interhemispheric switch in the left
hemisphere at the expense of the right. Manics have
increased verbal fluency and may show great facility at
fitting rhyme, alliteration, and rhythm to words.
They lack caution and may spend unwisely, indulge
readily in sexual adventures, and put themselves in harm's
way. In an effort to reverse mania speedily, psychiatrists
rarely pay any attention to the introspective aspects of
mania. These involve striking alterations in the perception of
space-time with a dramatic alteration in the feeling of
causation that the manic attributes to his/her will for events
happening around them. Manics are also attracted by light,
bodies of water, and the horizon. In keeping with the
interhemipheric switch formulation, there are a number of
pieces of evidence that activation is 'stuck' in the left
hemisphere during mania. These include a large, almost
complete preference for the left hemisphere percept in
perceptual rivalry and continuous blockage of the left nostril
(left hemisphere activation), instead of the usual alternation
between right and left nostril over hours as part of the
interhemispheric nasal cycle.
Mood stabilizer These drugs are the key agents for the
prevention and treatment of the manic and depressive
episodes of bipolar disorder. Mode of action is hard to
define in the absence of an accepted model for the episodes
of bipolar disorder and given the diverse pharmacology of
these agents. One element of their action may be
anticonvulsant, since many mood stabilizers (but not all)
also act to prevent convulsions, such valproate,
carbamazepine, and topiramate. The 'kindling' process, by
which cortical neurons progressively increase their
sensitivity to the same input, is moderated by
anticonvulsants and could explain a role in bipolar disorder
where 'kindling' has been implicated in the triggering of later
episodes and the 'falling in love with one's mania'
phenomenon. Lithium ions are not particularly
anticonvulsant, but do have an effect in development that
might be relevant to the likely role of interhemispheric
switching in bipolar disorder: right-left asymmetrical
structures are eliminated during development by lithium.
Since the ventral midbrain of the adult has many aspects
shared with embryonic brain, such as high concentrations of
retinoic acid not seen in other parts of the adult brain,
lithium may be acting by reducing the right-left asymmetry
(and thereby the magnitude of the interhemispheric swing
from right to left) in this part of the brain. The complexity of
the circuit from midbrain nuclei to cortex and back again
leaves open many possible sites of action for mood
stabilizers.
Nasal cycle This is caused by a slow interhemispheric
switch in the hypothalamus. The modal period of the cycle
348
Bipolar Disorder 349
is 90 min, but in some individuals, the cycle can end even
as fast as 20 min, and such individuals are considered to
be rare. Many more individuals are found in the long tail
of the distribution which stretches to many hours, where
those who suffer from bipolar disorders are also found.
The link between a slow nasal cycle and a slow rivalry cycle
illustrates the principle of 'fractal time' where a slow switcher
in perceptual rivalry (seconds) is also a slow switcher on other
interhemispheric rhythms such as the nasal cycle (many
hours). The alternating, side-to-side patency of the nasal
passages can be readily visualized using MRI, but a
satisfactory, economical method involves subjective
estimation of patency by breathing out through each nostril
alternately while the opposite nostril is occluded. Occlusion
of one nasal passage is brought about by increased blood
supply to that side, which is accompanied by increased blood
supply to the brain on that side. In other words, increased
activity in the Right hemisphere (as in depression) is
accompanied by occlusion of the R nostril and patency of
the L nostril, which can be a diagnostic sign of depression if it
is unrelieved by a switch.
Perceptual rivalry A constant, unvarying stimulus
may nevertheless be accompanied by an oscillatory
perception. By definition, the perceptual oscillation must
arise in the brain rather than the stimulus and has been used
repeatedly as a tool for the study of consciousness. The
oscillation is usually binary, between two alternative
interpretations, in keeping with experimental work
showing that the perceptual alternation corresponds
to a switch between activation of the two hemispheres.
In some cases (e.g., plaid motion rivalry where the
coherent, 'diamonds' alternative is associated with the
left hemisphere and the incoherent, 'sliding' percept with
the right) it is possible to gauge which hemisphere is
active from the choice of the two alternatives, but in
many cases the 'hemispheric assignment' is arbitrary.
A variety of perceptual rivalries can be seen at
www.uq.edu.au/nuq/jack/rivalry.html
Bipolar disorder, formerly manic depression, is a cyclical distur-
bance of mood in which episodes of mania and depression
alternate.
Mechanism
Basic biological principles clearly underpin bipolar disorder
because it is the most heritable of all mental health disorders.
The probability is better than 60% that an identical twin will
develop bipolar disorder if the other twin is affected. An acute
episode of mania certainly has a 'biochemical feel' because
of the sleeplessness, elevated mood, and accelerated verbal
behavior, so it is easy to understand the emphasis on a bio-
chemical basis for the disorder. But it needs to be realized that
biochemical theories provide only a unidimensional gradation
of symptoms and do not deal with the bipolar aspect of the
disorder, in which there is a switch between complementary
aspects, mania, and depression. In contrast to wholly bio-
chemical accounts, there is an interhemispheric mechanism
of bipolar disorder that adequately accounts for the comple-
mentary aspects.
The dynamics of the interhemispheric switch are deter-
mined genetically, as shown by studies of monozygotic twins,
and show significant slowing in bipolar disorder. The disorder
has the highest heritability of all mental disorders, a fact that
may be connected to the high heritability of the interhemi-
spheric switch rate.
Prevention
Post and Weiss have shown that the first episode is always
triggered by an environmental stressor, but that later epi-
sodes may become spontaneous. This finding has the impor-
tant implication that bipolar disorder is preventable. One
needs to only identify a vulnerable individual before the
first episode and then intervene with lifestyle modification
or mood-stabilizer medication. Since the dynamics of the
interhemispheric switch are highly heritable and relatively
stable throughout life, they can be measured noninva-
sively using perceptual rivalry. Interindividual variation in
rivalry switch rate is more than an order of magnitude,
making those susceptible individuals with the slowest switch
rates prominent.
Medications
The key medications for bipolar disorder are 'mood stabilizers.'
These drugs have an obscure mode of action and have com-
monly been used to search for the etiology of bipolar disorder.
Mood Stabilizers
1. Lithium ions: Have longest history of any mood stabilizers,
since Cade showed almost miraculous elimination of mania
using lithium. The long history probably explains the longer
list of side effects compared with more recent, popular, drugs
such as sodium valproate, which have not had time to
accumulate such a list. The dose of lithium and its blood
level are based on its antimanic effect and may be higher
than that required for a maintenance mood-stabilizing
effect, whose longer time course is difficult to titrate.
Lithium has nearly a dozen documented biochemical
effects. Two effects of lithium are relevant to the interhemi-
spheric hypothesis, a reduction in the degree of asymmetry
in brain development and a phase reduction of circadian
rhythm.
2. Sodium valproate: Its anticonvulsant action may relate to
'kindling' in the very plastic hemispheres.
3. Omega-3 fatty acids, DHA, and EPA: These naturally occurring
molecules from seafood have a mild, mood-stabilizing effect
that is best supplemented with more powerful drugs such as
lithium. They may play an important role in prevention.
350 Bipolar Disorder
4. Carbazepine: Powerful mood stabilizer and anticonvulsant
that is complicated by a high incidence of serious side
effects.
5. Lamotrigine: Has the opposite phase reduction to lithium
and is more effective, in consequence, on the depressive
phase of bipolar disorder. Now often combined with lith-
ium to control both manic and depressive phases.
6. Topiramate: Has a powerful effect on the interhemispheric
switch and is used as a mood stabilizer of last resort, in
addition to its role as an anticonvulsant.
7. Dopamine antagonists: Haloperidol and olanzapine (dopa-
mine antagonists) are sometimes used as mood stabilizers
in a maintenance role in addition to their well-accepted
role in acute treatment of mania. They have major side
effects, such as massive weight gain, which limit their long-
term use.
Age of Onset
The disorder can appear first at any age, but peaks in late
adolescence/early adulthood when sexual relations provide
the greatest source of environmental stressors.
Medication and Mood Stabilization
There are two broad classes of medications used in bipolar
disorder: (1) mood stabilizers and (2) antidepressants (neuro-
transmitter (usually 5HT) re-uptake inhibitors).
The first class does not have any obvious subjective effects.
For example, putting aside initial side effects such as nausea,
there is no subjective effect of lithium ions, which could be
placed in a person's breakfast cereal without him or her notic-
ing. Instead of expecting subjective effects, one needs to keep
an extended record of mood changes, which are moderated by
lithium ions even though there may not be a detectable intro-
spective mood shift that is attributable to lithium.
Mood Stabilizers: Mode of Action
Understanding the mode of action of mood stabilizers is lim-
ited by the lack of an underlying model of bipolar disorder. In
fact, many researchers have tried to 'reverse engineer' the basis
of bipolar disorder by starting with the actions of the mood
stabilizer and then working backwards to try to understand
how it might moderate the fundamental cause of bipolar disor-
der. This approach has not been successful. However, if the
evidence that bipolar disorder is based upon a dysrhythmia of
interhemispheric switches is accepted, there is plenty of evi-
dence that illuminates the mode of action of the mood stabi-
lizers. For example, it is known that the hypothalamic,
suprachiasmatic nuclei can act as a daily interhemispheric
switch, whose circadian rhythm is differentially affected by
lithium and lamotrigene. Lithium acts to reduce the action of
the phase advance mechanism, while lamotrigene acts on the
phase retardation mechanism in the circadian cycle. This would
explain the action of these two mood stabilizers, with lithium
being most effective on mania and lamotrigene most effective
on the depressive part of the manic — depressive cycle, since
mania is triggered by a phase advance and depression by a
phase retardation. Related to these effects is lithium's action
to reduce hemispheric asymmetry. Episodes are characterized
by an extreme asymmetry in the interhemispheric switch
('Stuck switch'), with a Left (L) bias in mania and a Right (R)
bias in depression. Lithium has a strong action in eliminating
R — L asymmetry during development and this is likely to apply
also to the interhemispheric switch because it retains many
embryonic features not found elsewhere in the adult brain,
such as a high concentration of retinoic acid.
Antidepressants (Reuptake Inhibitors)
Billions of dollars in profits flow from this class of drugs. The
first of the antidepressants was Prozac, but there are now
dozens with a similar action. In view of the large profits, one
might expect a detailed mode of action to have appeared that
would help one to understand how these drugs bring about
mood change. There is detailed biochemical information
about how they inhibit the reuptake mechanism in 5HT
nerve terminals, but one has to go to animal studies to get a
more global understanding of these drugs. One key fact is that
the hemispheric asymmetry that is no obvious in the motor
and speech system is also present at the level of transmitters
like 5HT, with the 5HT system much more well developed in
the Right (R) hemisphere. So increasing the overall level of
5HT by blocking reuptake will desensitize the R hemisphere,
which has more 5HT transmitter and receptors. In other words,
these drugs will enhance the relative activity of the Left (L)
hemisphere, thus improving mood because of its cognitive
style to choose an alternative and to ignore the rest. The
enhancement of L hemisphere action by these drugs have
been observed in a number of different animal models.
The danger of this mechanism of action is that it can pre-
cipitate mania, also an L hemisphere paroxysm. Mania is a
frequent outcome when a depressed person, treated with one
of these antidepressants, is actually suffering from unrecog-
nized 'bipolar depression'.
Perceptual Rivalry and Interhemispheric Switching
The cerebral hemispheres have complementary cognitive
styles, according to the work of Ramachandran and Pascual-
Leone. It makes logical sense to alternate between the denying,
choosing, hypothesis-forming abilities of the left hemisphere
and the all-accepting, discrepancy-detecting abilities of the
right hemisphere. An oscillatory interhemispheric switch has
been identified using psychophysics, comparative studies, and
neuropharmacology. Caloric vestibular stimulation can be
used to activate one hemisphere for around 30 min and can
be used to induce a manic-like state (left hemisphere activa-
tion), a depressive state (right hemisphere activation), or to
reverse mania by activating the right hemisphere. The temporal
precision of transcranial magnetic stimulation's disruptive
effects can be used to manipulate the switch so that left hemi-
sphere phenomena are interrupted when the left hemisphere,
but not the right hemisphere, is manipulated. And vice versa
for the right hemisphere. The switch is difficult to demonstrate
with functional magnetic resonance imaging (fMRI) because of
Bipolar Disorder 351
the sluggish response of this brain scanning method, but Mao
used a very slow version of perceptual rivalry to show an
associated interhemispheric switch using fMRI.
Perceptual rivalry is an oscillation of perception that switches
between different alternatives in the face of an unchanging, if
ambiguous, stimulus. The source of the oscillation must be
endogenous to the brain, because the sensory input is unchang-
ing. Extensive work has shown that the source of perceptual
rivalry is interhemispheric switching. Caloric stimulation acti-
vates one hemisphere and simultaneously increases the propor-
tion of time spent seeing the percept linked to that hemisphere.
Likewise, transcranial magnetic stimulation applied to one
hemisphere interferes in a highly specific and time-locked fash-
ion with one percept but not the other.
The interhemispheric model predicts that perceptual rivalry
will be linked specifically to mania or depression, and this is
what occurs. The onset of mania is accompanied by increasing
predominance of the left hemisphere percept (e.g., horizontal
in H/V rivalry, or coherent 'diamonds' in plaid motion rivalry
and approaching motion in form from motion sprees, etc.).
Depression is accompanied by increasing predominance of the
right hemisphere percept. The bias for one percept disappears
and returns to the balanced state where the two percepts are
seen for roughly equal amounts of time during recovery.
It is commonly assumed that the interhemispheric switch
mechanism is the corpus callosum, the large bundle of fibers
that interconnects the hemispheres, but this is not true. Inter-
hemispheric switching is universally phylogentical in vertebrates
and both fish and birds switch without a corpus callosum. More-
over, a back-and-forth mechanism between the hemispheres
would be slow and clumsy compared with bilateral projections
from the brainstem that have balanced conduction velocities.
The switch has been dissected pharmacologically by exam-
ining the effects upon perceptual rivalry of catecholamine and
serotonin analogues. The oscillatory mechanism, responsible
for the individual's stable and highly heritable switch rate,
involves 5HT-1A receptors, whose location in the brainstem
help provide confidence about the location of the switch. The
switch itself is probably a collection of cortically projecting
dopaminergic neurons in the ventral midbrain.
Jet Lag
The circadian shifts of jet lag provide one of the most powerful
triggers of an episode. For example, a manic tendency can be
triggered to full mania by flying eastward 7-1 Oh. Similarly
depression can be triggered by westward flight.
Fractal Time
There is strong support from a number of independent studies
for a prolongation of the perceptual rivalry cycle in bipolar
disorder. In turn, this cycle measured in seconds, has been
shown to be an interhemispheric cycle. Interhemispheric
switches can explain all the features of bipolar disorder, but
there is a problem linking the seconds time course of the
interhemispheric switch of perceptual rivalry and the days/
weeks time course of the switches between mania and depres-
sion. This problem is solved by the realization that there are
many interhemispheric switches of different time course, such
as the nasal cycle, whose time course is measured in hours and
the information from the genetics of circadian cycles that a
mutant cycle with a prolonged circadian cycle will have pro-
longed cycles at all scales from seconds to days ('fractal time').
Bipolar I Versus Bipolar II
A diagnosis of Bipolar I requires at least one hospital admission
for mania and has an incidence in the general population
around 1%. Those with this diagnosis have a bias of the inter-
hemispheric switch for the left hemisphere, so that if the switch
gets stuck it is more likely on the left, that is, mania. Bipolar II
is more common, around 3%, and involves recurrent episodes
of depression without frank mania. Normal members of the
population as a whole show a bias toward the right hemi-
sphere (the 'realistic' side) in the interhemispheric switch.
This bias is reflected in a similar rightward bias in the slower
switch of Bipolar II, which is therefore more likely to get stuck
on the right, depressive side.
Famous Bipolars
The genes for bipolar disorder are widespread in all cultures.
The incidence of Bipolar I is around 1% and for Bipolar II
much higher; so it can be inferred that there is a selective
advantage for these genes that prevents them from being elimi-
nated from the population. The 'Bipolar trait' of a prolonged
interhemispheric switch time has consequences that are inde-
pendent of possible episodes. These include a prolonged atten-
tion span that is useful in creative problem solving and
composition. Since time is a proxy for sensitivity when receptor
activation can be prolonged but receptor numbers not
increased, a prolonged switch rate means increased sensitivity.
Virginia Woolf illustrates this extraordinary sensitivity in her
ability to detect fine nuances and injustices in the interactions
of others and to place them in her writings. Woolf was also
acutely aware of the rise of Hitler, whose presence was felt
nearby despite an actual distance of hundreds of miles, and
this sensitivity to the dictator ultimately contributed to her
suicide. In a family of children with one bipolar, its sensitivity
is commonly recognized as the first to detect the onset of an
approaching storm, presumably via barometric pressure
changes, and to recognize subtle nuances in interpersonal
relations that escape other family members.
Winston Churchill (Bipolar II) was voted the man of the
twentieth century, along with Albert Einstein, because of
Churchill's leadership in the Battle of Britain. His success may
be attributable to a number of features of his bipolar brain, such
as the ability to hold the same problem in his mind for much
longer than the usual time, extreme sensitivity to small cues
about the enemy's intentions provided by their movements,
and the lack of normal sleep patterns that enabled a problem
to be faced at any time.
Kay Jamison has shown that the chances of finding a bipo-
lar among a group of poets or artists is 20 x normal and has
further shown the cyclic musical output of Schumann that was
linked to the oscillations of bipolar disorder and eventually
resulted in his suicide.
352 Bipolar Disorder
fMRI
fMRI relies on the magnetic signal that can be induced by a
powerful magnet in hemoglobin molecules that have the Fe 2+
molecule exposed to water by the loss of an oxygen molecule.
In other words, the system detects deoxyhemoglobin in venous
blood that is draining from an area of intense activity. It takes
about 8s for the deoxygenated venous blood to flow after
neural activity begins; so this technique is relatively sluggish
and does not pick up the rapid changes when the hemispheres
switch back and forth. Some studies have seen switching using
fMRI by choosing a form of perceptual rivalry that has a very
slow switch rate.
conditions of this highly heritable disorder may soon be possi-
ble and lead to its prevention, by intervention before the first
episode.
Further Reading
Goodwin FK and Jamison KR (2007) Manic-Depressive Illness, 2nd edn. Oxford: Oxford
University Press.
Miller SM, Liu G-B, Ngo T, Carson R, Riek S, and Pettigrew JD (2000) Interhemipheric
switching mediates perceptual rivalry. Current Biology 10: 383-392.
Pettigrew JD and Miller SM (1998) A ‘sticky’ interhemispheric switch in bipolar
disorder? Proceedings of the Royal Society B 265: 2142-2148.
Post RM and Weiss SR (1989) Sensitization, kindling, and anticonvulsants in mania.
Journal of Clinical Psychiatry 50(supplement): 23—30.
Conclusion
Advances in neuroscience are beginning to clarify the underly-
ing cause of bipolar disorder. Recognition of the predisposing
Birth Order, Effect on Personality, and Behavior
C Salmon, University of Redlands, Redlands, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Big five Personality trait model that proposes five basic
dimensions of personality, conscientiousness, agreeableness,
extraversion, openness to experience, and neuroticism.
Birth order An individual's rank by age within their family
of birth. For example, in a family of three children, there is a
firstborn, secondborn (or middleborn), and a lastborn.
In a family of four children, there is a firstborn, secondborn,
thirdborn, and lastborn. In this case, both the second- and
thirdborn could also be considered middleborn.
Fitness Fitness can be thought of as the relative probability
of survival and reproduction of an individual.
Niche theory Frank Sulloway's theory suggesting that birth
order influences the strategies siblings use to gain resources
and minimize sibling conflict.
Parental investment Any parental expenditure (time,
energy, etc.) that benefits one offspring (through increased
chance of survival and reproduction) at the expense of
investing elsewhere (in other offspring, mating
opportunities, etc.).
Parent-offspring conflict The conflict resulting from
differences in optimal parental investment to an offspring
from the perspectives of parent and offspring. Typically, an
offspring will desire greater levels of investment for a longer
period of time than is optimal from the parental perspective.
Well-studied examples include weaning conflict and
maternal-fetal conflict.
Personality A set of characteristics possessed by an individual
that influence their thoughts, attitudes, and behaviors.
Reproductive value An individual's expected future
reproduction.
Sibling competition Conflict between siblings, often over
parental investment or other resources. It is also referred to
as sibling rivalry.
Introduction
Psychologists have long been fascinated with birth order and
its possible role in shaping personality. Over 75 years ago,
Adler suggested that personality traits are related to an indivi-
dual's ordinal position within the family. He believed that
firstborns would be resentful when they were displaced by
the arrival of a new child and that this would lead to neuroti-
cism and substance abuse; that lastborns would be spoiled and
emotionally immature; and that middle children would be the
most stable as they were never dethroned or spoiled. Adler's
work led to a plethora of studies examining the relationships
between birth order and just about everything from personality
to psychiatric disorders, intelligence, and creativity. Today, there
is reasonable agreement that genetic influences account for a
substantial portion (around 40%) of the variance in personality
but that an almost equal amount of variance (about 35%) is due
to nonshared environment with the remainder due to shared
environment and measurement error. Birth order is one part of
that nonshared environment, for while siblings usually grow up
in the same family, they do not experience that environment in
the same way. Recently, researchers such as Frank Sulloway have
suggested that birth order shapes strategies for dealing with the
family environment that are part and parcel of our personalities.
This article will highlight the current thinking on birth order as a
strategy shaped by differential parental investment and sibling
competition (as originally proposed by Sulloway) and its effects
on personality and behavior.
Parental Investment
Many nonhuman species do not engage in parental care. The
fact that people do invest significantly in their own offspring is
testament to the substantial reproductive benefits that result
from such costly care. Not only do parents provide the basic
physical means for survival (food, shelter, and protection),
they invest greatly in fostering the development of skills that
are required for offspring success across the life span. Theoreti-
cal models of the evolution of parental solicitude suggest that
parents will often treat their offspring differently in relation to
a number of variables including offspring age, sex, birth order,
parental age, and cues of phenotypic quality.
Parental investment in an offspring (as first elucidated by
Robert Trivers) is any investment that a parent makes that
increases the likelihood of that offspring's survival and repro-
duction at the cost of that parent's ability to invest in other
offspring (either current siblings or future ones). In birds, we
conceive of this in terms of feeding and nest protection.
In people, it covers a wide range of activities from food and
shelter to education, piano lessons, or figure skating. Generally,
an offspring's fitness increases with the amount of parental care
received. Extremely low levels of parental investment can result
in the loss of an offspring; a minimum amount is required for
survival. But there is also a point of diminishing returns at very
high levels of investment beyond which offspring are unable to
capitalize on any further investment.
The relevance of birth order to parental investment is
related to several factors, including age of the offspring and
age of the mother. An offspring's expected contribution to
parental fitness rests mainly in their reproductive value. This
value increases with age until puberty, making older immature
offspring more valuable from a parental fitness perspective
than younger ones. It is this assurance of parental favoritism,
as well as an early absence of sibling contenders for a share of
parental attention, that makes firstborn children defenders
of parental values and the status quo while laterborns are
more likely to be rebellious.
353
354 Birth Order, Effect on Personality, and Behavior
However, offspring age is not the whole story. As parents
themselves grow older, the fitness value of any one offspring
increases relative to the parent's residual reproductive value.
As parents grow older, their own chance of reproducing again
drops, thus older parents have been observed to invest more
in offspring, all else being equal, than younger parents. This
can be clearly seen in the dramatic decreases in maternally
perpetrated infanticide as a function of maternal age. As a
result, while firstborns may have an inherent advantage in
terms of parental investment, there is a growing willingness
of older parents to invest highly in their lastborn, their last
chance to invest so to speak. In fact, lastborns are the only birth
order to receive their parental investment without the compet-
ing demands of a younger sibling. This would seem to imply
that middleborns may lose out in terms of parental investment
and attention and there are a number of recent studies that
support this. However, relatively few researchers have focused
on middleborns; the majority have focused on the contrast
between firstborns and laterborns.
Sibling Competition
But parental investment on its own does not tell the whole story.
Sibling competition has a role in how the battle for parental
resources plays out among a group of siblings. Natural selection
shaped mechanisms for sibling competition just as it shaped
mechanisms for discriminative parental solicitude. Strategies for
sibling competition can manifest in several ways including
dominance hierarchies, niche picking, and deidentification.
Dominance Hierarchies
In dominance hierarchies, the type of strategy siblings adopt
in order to deal with resource competition is influenced
by differences in size and strength (usually related to age).
Older siblings are able to physically intimidate their younger
and smaller brothers and sisters. Several aspects of personality,
including dominance and assertiveness on the part of first-
borns, can result from this dynamic. Typically, the younger
sibling is best served by low-power strategies. Rather than
physical force or the threat of it, the baby of the family often
appeals directly to the parent in any sibling dispute. The mid-
dleborn is typically forced to negotiate on their own, with the
parental support given to the youngest.
Niche Picking
Sulloway has argued that children adopt different roles or
niches within the family. Some of this are the result of genetic
differences, some are connected to sex differences, and some
are due to birth order. Specialization of roles within the family,
like specialization of species in the wild (the most famous
example being Darwin's finches), reduces levels of competi-
tion. Specialization also makes children more unique. If your
older brother is an awesome football player, a younger brother
may be better served by a different niche, attempting to excel in
a different sport or a totally different area such as academics or
arts. Eldest siblings also often occupy the role of surrogate
parent with its sense of responsibility and adherence to rules.
For later born children, there is no advantage to trying to
duplicate that role; they have to find their own niche and
their openness to experience with less adherence to rules
and authority facilitates this.
Deidentification
Deidentification is the process by which siblings try to differen-
tiate themselves from one another. In particular, when there is a
very short birth interval and siblings are close in age, their needs
are very similar and so competition for parental resources is
more acute. Diversifying in terms of strategies leads each sibling
to their own niche in the family, reducing direct sibling compe-
tition. This type of process is likely to lead to greater differences
between siblings who are next to each other in birth order (such
as first- and secondborn as opposed to first- and thirdbom).
Personality: The Big Five
Thousands of articles have been published focusing on birth
order and its effects on personality, behavior, and intelligence.
Recently, attention has been called to methodological issues, in
particular the need to control for differences in such variables
as socioeconomic status (SES) and family size. While relatively
little attention has been paid to the possible interaction of
gender and birth order, some evidence suggests that there is
a special role, with costs and benefits, for the firstborn boy,
whether or not he has an older sister. There have been some
researchers who have suggested that birth order differences are
nonexistent, while others have documented through individ-
ual studies and meta-analysis that there are a number of con-
sistent patterns to be found. One useful way to think about
personality in this setting is as a strategy. Birth order differences
in personality reflect the different strategies siblings adopt in
their attempts to acquire the resources they need, whether
they are acquired from family or friends. Here, I will concen-
trate on the Big five classification of personality dimensions
which focuses on the dimensions of conscientiousness, agree-
ableness, extraversion, openness to experience, and neuroti-
cism. This is not the only model of personality, but it is the
one most often focused on (in whole or in part) in studies of
birth order and personality. While not all studies of birth order
and these factors are consistent, they are quite frequently in
accordance with evolutionarily informed predictions.
Conscientiousness
Characteristics associated with conscientiousness include:
organized, neat, perfectionist, responsible, practical, and depend-
able. People who score high on conscientiousness are those
that tend to get ahead in life, the industrious go-getters. The
result is success in a variety of domains including elementary/
high school and college as well as high job performance and
satisfaction. High conscientious translates into playing by the
rules which usually results not only in job success but also
success at maintaining relationships, those that are sexual or
romantic as well as family ties.
As one would predict from niche theory, firstborns gener-
ally score higher on conscientiousness than their later born
Birth Order, Effect on Personality, and Behavior 355
siblings. Family and peers rate them as more disciplined and
thorough than their younger siblings. This should not be
surprising for several reasons. One is that firstborns often
adopt a surrogate parent role within the family, looking after
and perhaps also bossing around their younger brothers and
sisters. Responsibilities and organizational skills develop along
with this role. Two, because of their privileged status as the first
child, it is in their best interests to be the supporters of the
status quo which generally means being parent-like, reinfor-
cing the parental view of them as not only dependable but
obedient. They identify with parents and typically take the
resulting traditional approach. This family role is not typically
available to laterborns. They must find another parental invest-
ment route, exploring and developing their abilities in other
areas. The lastborn rarely gets, given the responsibilities that
foster the development of conscientiousness, though middle-
borns sometimes do.
Agreeableness
Those traits most highly associated with the factor of agreeable-
ness are: kind, understanding, helpful, cooperative, trusting,
and softhearted. People who score high in agreeableness are
highly invested in getting along well with others. They favor
strategies of negotiation rather than those of power and are
likely to withdraw from confrontation, finding such discord
distressing. Harmony and cooperation are important to them
in all their relationships, especially with family and friends. As
a result, they tend to be well liked by others and are often best
served by finding careers where this trait can pay off for them,
in particular, those where establishing a friendly relationship
with the customer or client is essential.
Generally, laterborns score higher than firstborns here.
While firstborns have the physical size advantage with which
to dominate their younger siblings, laterborns need to find
other strategies to get their way. Rather than power strategies,
laterborns use such tactics as negotiation, appealing to their
sibling's sense of fairness (not usually the most effective), and
puppy-like appeals to parents for help. Middle children, stuck
between the dominant firstborns and babies of the family, are
often rated most highly on agreeableness. These traits allow
them to get what they need, whether at home or from friends.
Their negotiating and mediating skills are forged by their posi-
tion in the middle. They are neither the favored first nor the
indulged last. Overall, laterborns tend to score high in social
popularity, a reflection of how easy others find their company.
Extraversion
Extraversion includes traits such as talkative, energetic, assertive,
and outgoing. Social interaction is the key here. Extraverts often
take on positions of leadership; first to offer their opinion and
suggestions. They are often quick to approach others, especially
on the dating scene. They often report greater levels of happi-
ness. Many people consider extraversion a very positive trait but
it can be problematic as they are often easily distracted from a
task by their social interactions which can be hazardous under
some circumstances (such as driving down the LA freeways!).
Birth order effects on extraversion have often been confus-
ing because some of the specific components, like assertiveness
and dominance, are ones that firstborns score highest on.
On the other hand, components such as sociability are
ones on which laterborns score higher than firstborns. As a
result, many studies do not show clear birth order effects on
extraversion unless they break it down into those aspects asso-
ciated with dominance and those associated with sociability.
It should not be surprising that firstborns score high in domi-
nance-related aspects of extraversion considering their often
pseudo parental (and sometimes physical) authority in the
family and the expectations parents typically have for them
and their future success. Laterborns follow a different strategy,
one of being outgoing, willing to try new things, meet new
people; they are extraverted in a highly sociable way, finding
pleasure and excitement in the company of others.
Openness to Experience
Traits associated with openness to experience include being
imaginative and creative, inventive, open to unusual ideas,
adventure, and nonconformity. Those scoring high in this
dimension are independent-minded and willing to tolerate
more ambiguity (or less certainty). As a result, they are often
the first to entertain new concepts and beliefs (which are some-
times later confirmed and other times refuted). Some studies
have indicated a relationship between openness and IQ.
Birth order effects on this factor are also somewhat mixed
with firstborns scoring higher on aspects that focus more
on intellect and intelligence, while laterborns score higher on
measures that center on the desire for novelty, experimenta-
tion, and a taste for the unconventional. This interest in being
unconventional is perhaps behind the tendency for laterborns
to identify themselves (as well as for others to identify them) as
rebellious, or the rebel of the family. Some studies have indi-
cated that middleborns may be the most rebellious birth order
as they are the ones most disadvantaged in the distribution of
parental investment. This personality trait facilitates their abil-
ity to find their own unique niche by making them more
willing than others to embrace the new and untried.
Neuroticism
The factor represented by neuroticism is also sometimes
referred to as emotional instability. Individuals who score
high on neuroticism are emotionally reactive and tend to
experience persistent negative emotional states such as anger
and anxiety. Life is full of challenges and emotional instability
reflects the way people deal with such stressful events. Those
that are emotionally stable weather the storms of life; they react
but not disproportionally. Those that score high on neuroti-
cism are rocked by challenges; they disrupt their lives and
emotional balance so that they are often perceived as moody
or unpredictable. They are more likely to have high levels of
anxiety and are more likely to develop posttraumatic stress
disorder in response to highly emotional negative events.
As you might imagine, this can put a strain on their personal
and professional relationships.
The relationship between birth order and this factor seems
quite small. However, one related birth order effect that shows
up relatively consistently is that middleborns score lower in
self-esteem than their siblings. This may be a result of the lower
levels of parental investment they receive in combination with
their unclear role in terms of family structure and parental
356 Birth Order, Effect on Personality, and Behavior
expectations. They do not know quite where they fit in and this
anxiety may be contributing to their issues of self-esteem. There
are also a few studies that suggest higher levels of neuroticism
in firstborns but the results are typically trends rather than
significant relationships. If the connection is there, it is more
likely related to the higher levels of anxiety and concern over
success (and meeting parental expectations) that sometimes
play a role in firstborns' drive to succeed.
Birth Order and Behavior
The impact of birth order on behavior has been demonstrated
in a number of domains including interactions with family,
friends, sexual partners, the workplace, religion and politics,
consumer behavior, risk-taking, and trust and cooperation.
While a majority of people recognize that birth order plays a
role in the family setting, the magnitude of birth order effects
outside the family is still open to debate. This section will
illustrate some of the better-documented connections between
birth order and behavior, emphasizing that birth order effects
are often strongest when the context or situation activates those
strategies that were effective in the family setting. We will
return to this issue of context in the section on methodology.
Family Interactions
While a significant proportion of personality can be attributed
to genetics, birth order differences in personality are forged by
the family environment. Birth order is not a genetic trait; it is a
proxy for the differences that exist between siblings such as
size, power, experience/skills, etc. As such, we would expect
birth order to have a greater impact on family relationships
and family-related behavior than in any other domain. This
has been demonstrated in a wide range of studies looking at
everything from emotional closeness and gift-giving to fre-
quency of contact with family members. Firstborns tend to
have more contact with siblings and parents than laterborns.
First- and lastborns report family identity to be more important
to their sense of self as well as feeling closer to parents than
middleborns. Middleborns are also less likely to be family
genealogists. While firstborns spend more money on gifts for
family, middleborns spend more on friends. Middleborns gen-
erally have less contact (once they are no longer living at
home) with parents than first or lastborns. However, middle-
born children do tend to have solid sibling ties, reporting being
more satisfied with sibling support than other birth orders. In
the absence of strong relationships with parents, they tend to
be more sibling- and peer-oriented, seeking advice and support
from peers as opposed to parents. Such birth order differences,
particularly those of middleborns, are a reflection of the degree
of perceived parental investment as well as their sense of their
natal family as a source of support. They are more likely to
report friends as their greatest social support and they appear to
invest more in them, as discussed in the next section.
Friends and Lovers
People often remark on how friends can be like family and
many individuals share intensely close friendships. Does birth
order influence the quality or quantity of friendship? What
about romantic relationships? Studies in the past several dec-
ades have suggested that laterborns, in some cases particularly
middleborns, are more popular with their peers. They certainly
score highly on the prosocial aspects of extraversion as well as
agreeableness, something you would expect from those who
seek out friends and get a good response. Middleborns also
hold more positive views on friends and friendship generally,
seeing them as confidants, reliable sources of help, support.
In turn, they are perceived as being good friends, empathetic,
reliable, and trustworthy. And middleborns do seem to care
about being seen this way, wanting to be perceived as good
friends.
When it comes to sexual relations, there are some interest-
ing birth order differences. Firstborns tend to wait until a later
age for their first sexual experience. They tend to be slightly
younger, however, when they first have a child and seem to
have slightly more children overall. Waiting a bit in terms
of having sex fits with their more conscientious and cautious
nature, as opposed to the move of adventurous lastborns who
have often also heard about the sexual exploits of their older
siblings. When it comes to marriage, reports from research and
marriage counselors seem to suggest that the most happy mar-
riages are those between a first- and a lastborn or anyone with a
middleborn. This is another place where the ability to negoti-
ate and mediate conflict that middleborns learn from being
sandwiched between siblings comes in handy. Two firstborns
seem to produce the most unstable marriages. And some data
suggests that there are differences in the likelihood of cheating
when in a monogamous relationship with middleborns being
the least likely to cheat, followed by firstborns, while lastborns
were the most likely to stray.
Workplace
Workplaces are highly variable in nature. People may interact
with the same small group of people all the time, or with many
different people most of the time. There may be a clear chain of
command or a more horizontal structure. But for most people,
coworkers are not family. And yet, work can be one place
where birth order can have a significant impact not only in
terms of such things as occupational choice but also personal
dynamics. There are indications that career choice can be
influenced by birth order. Firstborns tend to be drawn to the
business field as well as accounting and finance. Middle chil-
dren are often found in greater numbers in service fields, or
ones focusing on justice or interpersonal ties between clients or
coworkers. Lastborns tend to often gravitate to creative envir-
onments, teachers, and performers.
Business leaders and CEOs may also conduct their business
differently as a result of their own birth order and resulting
personality. They are often assumed to be firstborns, and many
are, as many firstborn-associated traits are likely to lead to
success at the top of the business world, such as dominance,
tough-mindedness, and competitiveness. All part of the per-
ception of the traits required for the drive to the top. But there
are also leaders who are laterborns and they have a very differ-
ent business style producing working environments more char-
acterized by a personable style of interaction with employees
and a willingness to take risks with strategies and products,
Birth Order, Effect on Personality, and Behavior 357
shaped by the inclination of laterborns to believe in the new
and untested. It is not so much that one birth order is more
successful here than another but that the style of their success is
different. Firstborns tend to find their success in traditional
established environments. Laterborns often find a more crea-
tive nontraditional style, characterized by flexible and original
approaches.
Religion and Politics
Religion and politics are two topics often to be avoided in
social situations. Both can be influenced by birth order. As
Frank Sulloway thoroughly documented in his 1996 book,
Born to Rebel, birth order has had a significant impact on who
supports radical political and scientific change with laterborns
leading the charge. Laterborns, like middleborn Darwin him-
self, provided much of the initial support for his theory of
evolution by natural selection, a concept incredibly revolution-
ary for the time. A majority of those most resistant to radical
change have been firstborns and when firstborns are leaders of
revolutions, they typically have less ideological implications
(Einstein and relativity, for example). Laterborns' openness
to experience makes them more willing to adopt new and
radical ideas perhaps especially when there are strong ideolog-
ical implications (such as Copernicus' theory that the earth
revolved around the sun as opposed to the traditional sun
rotating around the earth viewpoint).
When it comes to being a politician, firstborns are overrep-
resented among holders of political office. Perhaps that is one
reason that they are predisposed to use political rhetoric in
their political speeches. My own study on the use of kin termi-
nology in political speech indicated that both firstborns and
lastborns were somewhat susceptible to the use of such lan-
guage. Appeals to 'my brothers and sisters' were more influen-
tial with these birth orders. Interestingly, middleborns were
least responsive to such overtures. From an evolutionary per-
spective, this is not surprising. If middleborns receive less
attention and investment from parents, they are less likely to
see family as a valuable resource with shared best interests.
Instead, friends may be a better source of resources and shared
interests and, in fact, middleborns were more responsive to the
use of terms indicating friendship in political speech than they
were to those of kinship.
There have been relatively few examinations of birth order
and religion. Some indicate middleborns score lowest on the
importance of religion in their lives, perhaps a reflection of
their tendency to score highest on rebelliousness. An indivi-
dual's sense of religiosity is highly influenced by parental reli-
giosity and the nature of the parent-offspring relationship.
Adhering to parental religious views reflects a willingness
to conform to parental desires and expectations (i.e., more
commonly seen in firstborns). Middleborns, in the unconven-
tional and rebellious way, are less likely to accept the family
religious tradition. They do, however, often score high in terms
of spirituality.
Consumer Behavior
This is an area you might not immediately assume is
influenced by birth order. But there have been a number of
intriguing findings that suggest birth order can shape aspects of
consumer behavior. Consider, for example, the middleborn
son. In the face of firstborn parental favoritism, he has striven
to be the best he can be, not always the top dog but in the
top echelon. What kind of marketing campaign might get him
to buy your product? Perhaps one that signals that he is not
second best, that he is a winner. A study by Reid Claxton
suggested that indeed, in an ad campaign that focused on
exactly that train of thought, a secondborn son who was sec-
ond in his class etc. and finally purchased a first-class pen,
middleborns were more likely to identify with the character
in the ad copy (with the assumption that this would be likely
to lead to a future product purchase of this brand of pen).
One might also predict that laterborns would be more open to
innovative or new products and perhaps less attached to brand
names. And indeed it does appear that laterborns are more
open to product innovation while firstborns seem to be more
susceptible to the influence of others (normative influences
and status markers) when it comes to product choices.
Risk-Taking
Considering the impact of birth order on openness to experi-
ence, it should not be surprising that studies have noted birth
order effects on risk-taking. They are a variety of ways to look at
risk-taking, from survey instruments to actual behaviors.
Greater participation in high-risk sports such as skydiving,
automobile, and motorcycle racing, which are highly corre-
lated with sensation seeking and related to aspects of openness
to experience, has been widely noted on the part of laterborns
in comparison to firstborns. In fact, a recent study by Sulloway
and Zweigenhaft on the risk-taking behaviors of brothers in
major league baseball demonstrated that younger brothers
were more likely to attempt the high-risk strategy of base
stealing and were also more likely to succeed at it than their
older siblings. Older siblings were more successful in terms of
batting success. While both birth orders were successful, they
went after their success in different ways, excelling at different
aspects of the game. Other work has indicated that laterborns
are at greater risk of indulging in binge drinking, smoking, pot,
and more likely to engage in acts of civil disobedience. Sullo-
way in his 1996 book also documented that laterborns were
more likely than firstborns to travel to remote parts of the
world, risking death due to accident or disease.
Trust and Cooperation
There are a number of reasons to expect birth order to have an
impact on trust and cooperative behaviors. For one thing, birth
order influences scores on such personality aspects as altruism
and tendermindedness (components of agreeableness) with
laterborns scoring higher than firstborns. Being altruistic,
agreeable, and tender minded would tend to increase the will-
ingness to give others the benefit of the doubt and engage in
activities that would be to everybody's benefit. Some studies
have indicated that laterborns are perceived by others as more
trusting and straightforward than firstborns. Firstborns even
describe themselves as more self-centered and less empathetic
than others. Firstborns are also less trustful. There is also some
evidence that middleborns may be the most trustful of the
358 Birth Order, Effect on Personality, and Behavior
three birth orders toward nonkin and the most concerned
with maintaining the balance of reciprocity from their side
(in other words, they want to be seen as good reciprocators
or cooperators).
Some Methodological Issues
There are several methodological issues relevant to any discus-
sion of the impact of birth order on personality or behavior.
The first concerns surveys and personality inventories. Much
personality research relies on self-report personality inven-
tories, such as the Revised Neo Personality Inventory (NEO
PI-R) which measures the five-factor model. Are people being
honest in their responses and are they the best judge of their
own traits? Research has compared self-report with peer or
parental reports (even sibling pair reports), with a number
describing similar findings, some only finding effects for one
of the report types. For those who argue that birth order only has
an impact within the family domain, if peer reports agree with
self or parental, they have a problem to explain as this provides
support for the claim that the influence of birth order extends
outside the family domain. Another approach is to look at
actual behavior, rather than self-report or inventories. Measures
of actual behavior rather than surveys may provide more clear
and consistent results. An additional problem is the number of
participants needed to get reliable birth order effects. Some of
the clearest evidence is in the largest sample size studies which
make some study designs a challenge for many psychologists.
A related issue also has to do with study design. Most of
the early work, and much current research still, involves
studies that compare birth orders across, as opposed to within,
families. Obviously, there are significant benefits to a within-
family design as it controls for between family differences in
SES, ethnicity, and family size. Siblings are in direct conflict
with their own brothers and sisters, not siblings from other
families (for parental investment anyway) so that comparisons
between first- and lastborns, for example, are best done with
siblings from the same family, such as in Paulhus, Trapnell,
and Chen's study of personality and achievement and Sullo-
way and Zweigenhaft's recent study of birth order and baseball
performance of brothers in the major league.
The question of the relevance of birth order outside the
family setting is a good one that needs to be further addressed.
Some behaviors are very much situation-specific. Perhaps
some of the best approaches to the impact of birth order on
behavior are ones that attend to this and look for birth order
effects in relevant domains (ones where the context is similar
to that of the family or contains family- type cues) in the real
world (like major league baseball) or by recreating appropriate
situations in the laboratory. For example, I conducted a study,
mentioned previously, that examined the impact of kin terms
in political speech. This is a context where we might expect
susceptibility to kin cues. The speaker is usually trying to
convince his audience of their shared agenda and best interests,
reinforcing the sense of the people and the orator as a group
with common goals. It is a situation ripe for the manipulation
of kin cues and there are many documented examples of
the use of such terminology in speeches on public record.
I expected such techniques with kin language would be less
effective for middleborns who are perhaps less inclined to see
their family as always having their best interests at heart (as
opposed to their siblings) . Similarly, my work with Martin Daly
on genealogists and those people who chronicle their family
histories was a real world behavioral test of birth order's influ-
ence on such a behavior. Again we found that middleborns were
underrepresented among family historians. More work needs
to be done with attention to the behavioral context and how it
may or may not trigger family sentiments and psychology.
One final methodological consideration is that of the
specific birth orders. Historically, the majority of birth order
research has lumped middleborns and lastborns together and
compared them to firstborns or simply compared first and
last. Based on a Darwinian or niche model, this is simply not
the best approach. We would expect substantial differences
between middleborns and both firstborns and lastborns not
only on personality dimensions but also in terms of behavior.
A number of those who have separated the middleborns out
have indeed found such differences. Hopefully more research-
ers will continue to attend to the relevance of all three birth
positions and not merely consider firstborn versus all others.
Middleborns are in fact unique in their particular parental
investment situation (eldest and youngest do tend to have
periods of parental monopolization that the middleborn
never has). This has consequences for middleborn traits and
behaviors beyond simply not being the first.
Summary
Birth order is one of many environmental factors that play a
role in the development of personality and a variety of beha-
viors. A Darwinian family niche perspective makes clear pre-
dictions about the ways in which birth order effects will play
out and many of them have been tested in numerous research
programs over the past 10 years. In general, higher scores on
conscientiousness are found in firstborns while laterborns
score higher on agreeableness, and openness to experience.
The dominance aspects of extraversion are high-scoring for
firstborns while the sociability ones are high-scoring for last-
borns. Neuroticism has much smaller and less clear-cut results.
These distributions of personality traits serve individuals of
different birth orders in their attempts to secure the best resource
acquisition outcome possible. In turn, they also shape their
behavioral tendencies as children and adults.
See also: Big Five Model and Personality Disorders; Competition;
Family Systems; Parenting; Parent-Offspring Conflict.
Further Reading
Claxton RP (1 995) Birth order as a market segmentation variable. Journal of Consumer
Marketing 12: 22-38.
Courtiol A, Raymond M, and Faurie C (2009) Birth order affects behavior in the
investment game: Firstborns are less trustful and reciprocate less. Animal Behaviour
78: 1405-1411.
Ernst C and Angst J (1983) Birth Order: Its influence on Personality. New York:
Springer-Verlag.
Birth Order, Effect on Personality, and Behavior 359
Freese J, Powell B, and Steelman LC (1999) Rebel without a cause of effect: Birth order
and social attitudes. American Sociological Review 64: 207-231 .
Healey MD and Ellis BJ (2007) Birth order, conscientiousness, and openness to
experience: Tests of the family-niche model of personality using a within-family
methodology. Evolution and Human Behavior 28: 55-59.
Paulhus DL, Trapnell PD, and Chen D (1999) Birth order effects on personality and
achievement within families. Psychological Science 10: 482-488.
Saad G, Gill T, and Nataraajan R (2005) Are laterborns more innovative and
nonconforming consumers than firstborns? A Darwinian perspective. Journal of
Business Research 58: 902-909.
Salmon CA (1998) The evocative nature of kin terminology in political rhetoric. Politics
and the Life Sciences 17: 51-57.
Salmon CA (2003) Birth order and relationships: Family, friends and sexual partners.
Human Nature 14: 73-88.
Salmon CA and Daly M (1998) Birth order and familial sentiment: Middleborns are
different. Evolution and Human Behavior 1 9: 299-312.
Saraglou V and Fiasse L (2003) Birth order, personality, and religion: A study
among young adults from a three-sibling family. Personality and Individual
Differences 35: 19-29.
Sulloway FJ (1996) Bom to Rebel : Birth Order, Family Dynamics and Creative Lives.
New York: Pantheon.
Sulloway FJ (2008) Birth order. In: Salmon CA and Shackelford TK (eds.) Family
Relationships: An Evolutionary Perspective, pp. 162-182. New York: Oxford
University Press.
Sulloway FJ and Zweigenhaft RL (2010) Birth order and risk taking in athletics:
A meta-analysis and study of major league baseball. Personality and Social
Psychology Review. April 30. Doi: 1 0 . 1 1 77/1 08886831 0361 241 . First published
online.
Body Dysmorphic Disorder: A Review
J E Grant and B L Odlaug, University of Minnesota Medical Center, Minneapolis, MN, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cognitive behavior therapy A type of therapy used in either
individual or group settings in which the therapist attempts
to change distorted thoughts and problem behaviors
through the identification and modification of these
thoughts and changing the rewards for behaviors.
Compulsion Repetitive behaviors (e.g., mirror-checking,
camouflaging with makeup or hats, etc.) or mental acts (e.g.,
comparing with others) undertaken to prevent or reduce
anxiety or stress and not to provide pleasure or gratification.
Double-blind A research method in which the identity of
the treatment and control groups is concealed (or 'blinded')
from the researchers administering the treatment condition
and research subjects involved in the study to eliminate
biased results.
Impulsivity A predisposition toward rapid, unplanned
reactions to either internal or external stimuli with
diminished regard for negative consequences.
Obsession Persistent ideas, thoughts, impulses, or images
which are experienced as intrusive and inappropriate and
which cause marked anxiety and/or distress.
Serotonin-reuptake inhibitor A class of drug that acts as a
neurotransmitter uptake inhibitor on serotonin receptors
within the brain.
Introduction
Most people are dissatisfied with some aspect of their appear-
ance. In fact, more than half of all women and nearly half
of all men in the United States are dissatisfied with the way
they look. For some people, however, the preoccupation with
perceived defects in physical appearance is excessive. These
patients suffer from body dysmorphic disorder (BDD), a dis-
order characterized by significant distress, impaired function-
ing, social withdrawal, and repeated attempts to hide or correct
a perceived defect in appearance.
Despite having been described around the world for more
than a century, BDD was not included in diagnostic systems
until DSM-III and ICD-10. DSM-III included BDD as an exam-
ple of an atypical somatoform disorder and DSM-III-R provided
the diagnostic criteria. Although DSM-IV allows people with
delusional BDD to receive a diagnosis of both delusional disor-
der and BDD, ICD-10 classifies delusional BDD as a variant of
delusional syndrome.
BDD is a relatively common disorder. Although studies
within the general population have found prevalence rates
ranging from 0.7% to 2.4%, studies of clinical samples suggest
higher rates: 7% among cosmetic surgery patients; 8.0% in
patients with major depression; 8.8% among dermatology
patients; 1 1% in patients with social phobia; 8-12% in patients
with obsessive-compulsive disorder (OCD); 13.1% among psy-
chiatric inpatients; and 39% in patients with anorexia nervosa.
BDD affects both children and adults. BDD usually starts in
adolescence and appears to be about equally common in men
and women or slightly more common in women.
Clinical Characteristics
Individuals with BDD are preoccupied with the idea that some
aspect, or aspects, of their appearance look abnormal. They
may describe these body areas as being unattractive, deformed,
disfigured, ugly, hideous, or 'not right.' Although any body part
can be the focus of concern, and most individuals with BDD
are preoccupied with numerous body areas, the face or head is
commonly the body area that troubles BDD patients, with a
focus on skin flaws, defects, blemishes, wrinkles, scars, or
supposed acne. Typical patients are preoccupied with their
appearance for several hours each day. The majority also have
ideas of reference, thinking that others notice their imagined
defect and react to it with dislike or disgust. As a result, many
BDD patients limit their social interactions and may become
quite isolated. In response to their obsessive thoughts, patients
may spend hours checking their skin in the mirror and
camouflaging their skin with excessive and unneeded makeup.
Nearly all persons with BDD perform repetitive, time-
consuming behaviors. These behaviors focus on examining,
improving, being reassured about, or hiding the perceived
defect. These behaviors are often described as 'compulsive' in
the sense that the urge to perform them is strong and difficult
to resist. They are also sometimes referred to as 'safety beha-
viors,' meaning that they are performed to prevent a feared
catastrophe (e.g., camouflaging 'ghost-like' skin with bronzer
to prevent feared scrutiny by other people).
The most common behaviors are camouflaging (e.g., with
hair, a wig, makeup, body position, sunglasses, a hat, or other
clothing), comparing one's appearance with that of other peo-
ple, excessively checking the perceived flaw in mirrors or other
reflecting surfaces (e.g., windows), excessively grooming (e.g.,
applying makeup or tweezing, styling, or cutting hair), and
seeking reassurance that the body area looks normal or accept-
able to others. The behaviors are unlimited and varied, how-
ever, and may consist of dieting, excessive exercising or
weightlifting, touching or measuring the body part, tanning,
buying excessive amounts of beauty products and compulsive
shopping, (which can lead to financial indebtedness), repeated
clothes changing, seeking surgery or medical treatment, and
using potentially dangerous anabolic steroids to bulk up.
Nearly all patients with BDD experience impairment in
social and occupational/academic functioning as a result of
their appearance concerns. People with BDD also have
360
Body Dysmorphic Disorder: A Review 361
markedly poor quality of life, and approximately one quarter
of patients with BDD are so distressed that they attempt sui-
cide. However, the severity of BDD varies, with some people
appearing to lead relatively normal lives despite the suffering
and interference they experience.
Gender Differences
The prevalence of BDD appears to have a fairly equal gender
distribution. A study examining gender disparities in 200 indi-
viduals with BDD reported more similarities than differences.
While men were more likely to be older and single, women
were significantly more likely to have a lifetime diagnosis of
an eating disorder (42.3%), pick at their skin (52.6%), and
use camouflaging techniques such as wearing concealing cloth-
ing or using makeup. Women were also more likely to be
concerned with their buttocks (28.5%) and thighs (28.5%)
while men were more concerned with body building (36.5%
vs. 9.5%). Men are also more likely to report an alcohol use
disorder (55.6% vs. 36.5%) and engage in weight-lifting
behavior (25.4%), however, both men and women equally
engage in behaviors such as dieting (37.5%) and excessive
exercise (21.5%).
Diagnosing BDD
Numerous studies have found that BDD usually goes unrecog-
nized in clinical settings, probably because clinicians fail to
screen for the disorder and patients are too ashamed of their
symptoms to spontaneously reveal them. Diagnosing BDD,
however, is usually straightforward and can be accomplished
by asking patients if they are unhappy with how they look or if
they worry excessively about their appearance. An affirmative
answer can be followed up with questions determining the
degree of distress that these concerns cause and whether they
cause functional impairment. DSM-IV criteria are met when
the patient reports being preoccupied with a nonexistent or
minimal appearance flaw that causes clinically significant dis-
tress or impairs functioning. It must also be ascertained that
the appearance concerns are not better accounted for by an
eating disorder. Simple self-report and clinician-administered
screening and diagnostic measures are available. It is especially
important to screen for BDD in patients who present with
some of the disorder's clues - for example, the behaviors
described above, histories of seeking cosmetic surgeries, ideas
or delusions of reference, or being housebound.
Comorbidity
Most individuals with BDD have other co-occurring psychiatric
disorders. Lifetime major depression (84%), substance use
(49%), social phobia (40%), and OCD (33%) are the most
common. Personality disorders, particularly avoidant (38%),
paranoid (38%), and obsessive-compulsive (28%), are also
common.
In a recent study of 176 BDD subjects, 49% had a lifetime
substance use disorder and a majority of subjects (68%)
reported that BDD symptoms contributed to their substance
use. The lifetime substance use disorder rate in subjects with
BDD in this study is notably higher than reported rates in the
general population (26.6-29.0%). In addition, those with a
lifetime substance use disorder had a significantly higher rate
of suicide attempts. The high lifetime rate of substance use
disorders and the very high rate of suicide attempts in subjects
with a substance use disorder suggest that clinicians should
carefully screen BDD patients for a substance use disorder, as
the presence of a lifetime substance use disorder may have
treatment implications.
Relationship to Other Mental Illnesses
Obsessive-Compulsive Disorder
Although BDD is classified as a somatoform disorder in DSM-
IV, it has many similarities to OCD - most notably, the presence
of prominent obsessions and repetitive compulsive behaviors.
There also appear to be differences between these disorders; for
example, BDD patients are less likely to be married, possibly
because of their avoidance of dating due to an intense fear of
evaluation by others. BDD subjects also have poorer insight into
their illness and are more likely to think about suicide or make
a suicide attempt because of their disorder. They also have higher
lifetime rates of major depression and social phobia than
individuals with OCD. In addition, BDD appears more often
characterized by shame, embarrassment, low self-esteem, and
rejection sensitivity - features that it shares with social phobia.
Major Depressive Disorder
BDD and depression often co-occur and both disorders are
characterized by low self-esteem, rejection sensitivity, and
feelings of unworthiness. Unlike depression, however, a core
feature of BDD is prominent obsessional preoccupations
and repetitive compulsive behaviors. Many depressed patients
focus less on their appearance, even neglecting it, rather than
overfocusing on it. In addition, onset of BDD usually precedes
that of major depression, suggesting that BDD is not simply
a symptom of depression. Perhaps most important, BDD
appears to respond to serotonin-reuptake inhibitors (SRIs)
but not non-SRI antidepressants, which is an important differ-
ence between BDD and depression that has clinical implica-
tions. Another apparent difference is that BDD does not appear
to respond to ECT, although data on this topic are very limited.
Anorexia Nervosa
BDD and anorexia nervosa share disturbed body image and a
preoccupation with perceived appearance flaws. Many patients
with anorexia are preoccupied with nonweight aspects of
appearance, such as the size of the stomach or thighs, or even
body areas such as the skin or nose. Conversely, some BDD
patients are preoccupied with body weight and body shape.
Both disorders involve intrusive thoughts about appearance,
dissatisfaction with appearance, and an overemphasis on the
importance of appearance to the evaluation of self-worth.
Both BDD and anorexia may also involve appearance-
related ritualistic or repetitive behaviors (e.g., mirror checking
362 Body Dysmorphic Disorder: A Review
and body measuring), and some BDD patients diet or exces-
sively exercise. Furthermore, people with anorexia or BDD may
avoid places, activities, and ways of dressing that provoke self-
consciousness about how they look. In addition, in both dis-
orders a quest to improve appearance drives attempts to
change the body's appearance - for example, through weight
control in anorexia and excessive grooming or cosmetic surgery
in BDD.
While anorexia and BDD have similarities (as noted above),
as well as a hazy area of overlap, they also appear to have some
important differences. For example, while 90% of patients with
anorexia are women, this appears to be the case for only
50-60% of patients with BDD. The two disorders also have
somewhat different comorbidity patterns. In a controlled fam-
ily study of OCD, BDD was more common in first-degree
relatives of OCD probands than control probands, whereas
the eating disorders were not; this finding suggests that BDD
and not eating disorders be considered part of a familial OCD
spectrum. In addition, a study examining 45 patients with
anorexia or bulimia nervosa, 51 patients with BDD, and 50
nonclinical controls, found that the eating disorder and BDD
groups had equally severe body image disturbance and nega-
tive self-esteem. Flowever, the eating disorder subjects had
more widespread psychopathology, whereas the BDD subjects
reported more negative self-evaluation due to appearance and
more avoidance of activities due to self-consciousness about
appearance. Importantly, BDD and anorexia seem to respond
differently to treatment. Unlike patients with anorexia, a
majority of patients with BDD improve with SRIs, some very
robustly. And while preliminary data indicate that BDD often
responds well to cognitive behavioral therapy (CBT), CBT has
generally been less effective for anorexia nervosa.
Although BDD and anorexia nervosa have various simila-
rities and differences, it is also important to recognize that the
disorders may also co-occur. Recognizing BDD in patients with
anorexia nervosa is important because the co-occurrence of the
two disorders may have clinical implications. In the only study
of this issue, 16 patients with both disorders were compared to
25 women with anorexia nervosa alone. Those who had
anorexia nervosa plus BDD had significantly poorer function-
ing, had been psychiatrically hospitalized more often (6.3
times vs. 3.8), and had three times the rate of suicide attempts
(63% vs. 20%).
Social Phobia
Previous research has indicated high rates of lifetime co-occur-
ring social phobia in patients with BDD (31.1-37%). The fear
of negative evaluation by others and social reclusiveness are
hallmarks of both disorders and, in some cultures, BDD is
actually regarded as a form of social phobia. Treatment regi-
mens for both disorders have indicated the efficaciousness of
certain serotonergic medications, including fluoxetine and clo-
mipramine, indicating at least some shared neurobiological
processes between BDD and social phobia. Research specifi-
cally examining those with social phobia alone or those with
BDD alone, however, has yet to be obtained. Given the signifi-
cant overlap between the two disorders, screening patients with
either social phobia or BDD for the co-occurrence of the other
disorder is imperative.
Pathologic Skin Picking
Some research suggests that BDD and pathologic skin picking
may share common phenomenological links. In fact, a previ-
ous study of patients with BDD found that 81.8% picked at
their skin to improve appearance. Therefore picking behaviour
may be viewed as a symptom of BDD in many individuals.
A recent study of 53 individuals diagnosed with pathological
skin picking (excluding those individuals who picked primar-
ily because of appearance), however, found that no individual
had co-occurring BDD. This study highlights the heterogeneity
of picking. Picking may be a symptom of a disorder such as
BDD as well as an independent compulsive behavior. Clini-
cians should therefore be aware of the possible heterogeneity
of picking as treatment options may differ. If seen as an inde-
pendent disorder, picking may respond to glutamate agents
but may respond preferentially to serotonergic medications
when it is a symptom of BDD.
Neurobiology
Although the etiology of BDD is most likely multifactorial
(biological, psychological, sociocultural), serotonin dysregula-
tion has been associated with BDD. BDD symptoms have been
exacerbated after tryptophan depletion and have had their
onset after abuse of ciproheptadine, a serotonin antagonist.
Neuropyschological and brain imaging studies have also
suggested that there may be impairment of the frontal-striatal
and temporo-parietal-occipital circuits which process facial
images and emotional information. In previous neuroimaging
studies, subjects with BDD demonstrated greater total white
matter compared with controls and a leftward shift in caudate
asymmetry, which might be suggestive of a striatal pathophys-
iology similar to that of OCD.
A small functional imaging study of six BDD patients, using
single photon emission computed tomography (SPECT),
showed variable, discrepant findings including relative perfu-
sion deficits in bilateral anterior-medial temporal and occipi-
tal regions and asymmetric perfusion in parietal lobes.
Clinical observation and neuropsychological testing indicate
that BDD patients hone in on details of certain appearance
features, usually their face, at the expense of global or configural
aspects. BDD patients may process faces in a piecemeal manner,
whereas healthy controls' perception of faces may be more
configural and holistic. The left inferior frontal gyms and right
amygdala in individuals with BDD demonstrate functional
hyperactivity for processing faces and their size correlates with
symptom severity. These regions may be involved in pathologi-
cal face processing that underlies the core symptoms of BDD.
Two recent functional magnetic resonance imaging studies
in BDD that examined visual processing of faces found that
individuals with BDD demonstrated abnormal activation pat-
terns that included greater left hemisphere activity in regions
including the inferior frontal gyms, as well as abnormal amyg-
dala activation compared to controls. Using voxel-based mor-
phometry, a second study found significant positive correlations
between severity of BDD and volumes of the left inferior frontal
gyms and the right amygdala.
The stmctural and functional studies to date suggest a visual
and emotional processing deficit involving the inferior frontal
Body Dysmorphic Disorder: A Review 363
gyrus and amygdala. Caudon, however, is required when
attempting to interpret data from preliminary studies con-
ducted in small samples of BDD patients in whom comorbid-
ity is common. Although important early insights have been
gleaned from research to date, our understanding of the neu-
robiology of BDD remains lacking.
Treatment
Many people suffering from BDD seek and receive surgical and
nonpsychiatric medical treatments (e.g., dermatologic) for
their perceived appearance flaws. Available data suggest that
such treatments are usually ineffective and that appearance
concerns usually persist unchanged. Many patients do not
report their symptoms to their physician. A study of 50 patients
with BDD reported that only 38% had told their general prac-
titioner of their body concerns. For patients who do report
their symptoms, 92% report feeling dissatisfied with their
treatment. This makes sense, since BDD involves an obses-
sional body image disturbance that is not based on reality;
changing a physical feature would not be expected to diminish
the disorder's core pathology.
Nonpsychiatric Treatments
Individuals with BDD often seek and receive treatment from
nonpsychiatric sources. Approximately 63% of BDD patients
have sought treatment from dermatologists and almost 75%
of those requesting dermatological treatment received it.
Antibiotics were most commonly prescribed, as well as isotret-
inoin or dermabrasion for perceived acne. Only 9.8% of all
dermatological treatments led to an improvement in BDD. In a
study of 11 patients with BDD, 7 (63.6%) sought treatment
from a dermatologist and four of those patients actually
received treatment. In the cases in which treatment was
received, no improvements in BDD were observed.
Surgical treatments are also common. A study assessing
cosmetic surgery patients found that 7% met BDD criteria.
Studies on BDD patients have illustrated that 23.2-26% have
undergone one or more surgical interventions on their per-
ceived defect. However, unlike most cosmetic surgeries which
result in fairly high levels of satisfaction (78-90%), patients
with BDD report dissatisfaction with surgical inventions more
often than not. In fact, a survey of 178 cosmetic surgeons who
had operated on patients with BDD found that in 43% of cases
the patient was more preoccupied with the perceived defect
after the surgery than before it. Furthermore, despite the high
number of surgeries performed, only 1% reported satisfaction
and free of BDD symptoms following surgery.
Pharmacotherapy
Case series and open-label studies suggest that selective seroto-
nin reuptake inhibitors (SSRIs), such citalopram, escitalopram,
and fluvoxamine have shown benefit in treating BDD. In addi-
tion, the antiepileptic medication levetiracetam has recently
demonstrated promise in reducing BDD symptoms in an
open-label study of 17 subjects treated for 12 weeks (mean
dose of 2044.1 ± 1065.2 mg day" 1 ).
Two placebo-controlled trials of serotonergic medications
have been conducted. In the first, 29 subjects with BDD were
randomized to a 16-week cross-over study of clomipramine
(mean dose of 138 mg day" 1 ) and desipramine (mean dose
147 mg day” 1 ). Clomipramine was superior to desipramine in
the acute treatment of BDD symptoms.
In the second placebo-controlled study, 67 subjects with
BDD were randomized to either 12 weeks of fluoxetine (mean
dose of 78 mg day” 1 ) or placebo. Fluoxetine was significantly
more effective than placebo for BDD beginning at week 8.
Eighteen (53%) of 34 assigned to fluoxetine and 6 (18%) of
33 assigned to placebo were responders.
Although some treatments, including serotonergic medica-
tions such as fluoxetine and clomipramine, are encouraging
treatments for BDD, there is insufficient evidence at this time
to definitively recommend these as first-line treatments. Con-
sequently, larger controlled studies are necessary to properly
identify efficacious treatments (Table 1).
Psychotherapy
CBT is another promising approach for BDD. Studies have found
that BDD often significantly improves when the following
approaches are used: cognitive restructuring (e.g., developing
more accurate and helpful beliefs about appearance), exposure
(e.g., exposing the perceived defect in social situations and pre-
venting avoidance behaviors), response prevention (stopping
compulsive behaviors, such as mirror checking), and behavioral
experiments (empirically testing hypotheses - i.e., dysfunctional
thoughts and beliefs). Additional components, such as mirror
retraining and mindfulness, may also be used. CBT has demon-
strated effectiveness in treating BDD in both individual and
group formats.
There have been three controlled psychotherapy studies in
BDD. In the first two, 12 weeks of CBT was compared to wait
list. In one study, 22 of 27 subjects with BDD no longer met
diagnostic criteria for BDD after treatment with eight 2-h ses-
sions of CBT. In the second study, those subjects receiving CBT
reported significant changes on specific measures of BDD and
depressed mood compared to those assigned to wait list.
In the only other controlled study, ten subjects with BDD
participated in a 6-week course of behavioral therapy followed
by 6 months of maintenance treatment. Subjects received a
standard behavior therapy protocol which consisted of expo-
sure in vivo and in imagery, with response prevention. Follow-
ing treatment, a 6-month maintenance program was instituted
for five subjects, with the other five serving as controls. Subjects
improved when using exposure with response prevention.
Although all subjects remained symptom free at follow-up,
those in the maintenance program continued to improve.
Longer follow-up demonstrated that at 12-, 18-, and 24-month
follow-up assessments, subjects participating in the mainte-
nance program were more efficient in managing limited
symptom return and had significantly lower anxiety and
depression. It is unclear, however, whether participants in
the maintenance program were more efficient in managing
limited symptom return etc., relative to their pretreatment
assessment or relative to the control group.
Similar to conclusions reached for pharmacotherapeutic
interventions for BDD, a dearth of information currently exists
on psychotherapeutic treatment efficacy. CBT appears very
promising, however, larger studies are needed to decipher
364 Body Dysmorphic Disorder: A Review
Table 1 Medication trials in body dysmorphic disorder
Medication
Mean dose
(mg day- 1 )
Study design
Subjects
Duration
Results 3
Clomipramine;
Desipramine
138.0 ±87.0
Double-blind
cross-over
design
29 randomized;
18 completed
16 weeks
Clomipramine significantly more effective than
desipramine; 65% of clomipramine and 35% of
desipramine subjects were responders
Fluoxetine
77.7 ±8.0
Double-blind
67 randomized
(fluoxetine = 34;
placebo = 33);
59 completed
(fluoxetine = 31;
placebo = 28)
1 week placebo
lead-in;
12 weeks of
treatment
Fluoxetine significantly more effective at weeks 8-
12; 53% of active and 18% of placebo group were
responders
Pimozide
augmentation
of fluoxetine
1.7 ± 1 .0
Double-blind
28 randomized
(pimozide =11;
placebo = 17);
19 completed
(pimozide = 6;
placebo = 13)
8 weeks
Pimozide augmentation of fluoxetine was not more
effective than placebo; 18.2% of active and 17.6%
of placebo group were responders
Fluvoxamine
238.3 ±85.8
Open-label
30 enrolled;
18 completed
16 weeks
63.3% responders
Fluvoxamine
208.3 ±63.4
Open-label
15 enrolled;
12 completed
10 weeks
66.6% responders
Citalopram
51 .3 ±16.9
Open-label
15 enrolled;
11 completed
12 weeks
73.3% responders
Levetiracetam
2044.1 ±1065.2
Open-label
17 enrolled;
11 completed
12 weeks
60% responders
a Results are presented for intent-to-treat analyses, except for the clomipramine study, which used a minimum treatment analysis.
whether CBT or another form of therapeutic intervention
should be utilized for BDD.
Conclusion
BDD causes significant distress and impairment in func-
tioning, and it is associated with an unusually poor quality
of life. Because this disorder appears relatively common, it
is important to screen for BDD in clinical settings. Research
on the treatment of BDD is in an early stage, so our
understanding of the most effective treatments is still
incomplete. SRIs, however, appear efficacious, particularly
when used at relatively high doses. In addition, CBT, using
a combination of cognitive and behavioral techniques, may
also be beneficial.
Future research needs to address clinical strategies to mini-
mize the underrecognition of BDD and to improve proper
treatment. Because of the high rates of comorbidity in patients
with BDD, treatment studies need to include subjects with
co-occurring disorders and thereby attempt to understand
how these other disorders affect treatment response. Finally,
more research is needed to address the pathophysiology of
BDD (e.g., by incorporating imaging and genetics). As with
any disorder, understanding etiology should allow for more
effective treatments.
See also: Body Image; Cognitive Behavior Therapy; Obsessive-
Compulsive Disorder.
Further Reading
Carey P, Seedat S, Warwick J, van Heerden B, and Stein DJ (2004) SPECT imaging of
body dysmorphic disorder. Journal of Neuropsychiatry and Clinical Neurosciences
16: 357-359.
Grant JE, Kim SW, and Eckert ED (2002) Body dysmorphic disorder in patients with
anorexia nervosa: Prevalence, clinical features, and delusionality of body image.
International Journal of Eating Disorders 32: 291-300.
Grant JE, Menard W, Pagano M, Fay C, and Phillips KA (2005) Substance use disorders in
individuals with body dysmorphic disorder. Journal of Clinical Psychiatry 66: 309-316.
Hanes KR (1998) Neuropsychological performance in body dysmorphic disorder.
Journal of the International Neuropsychological Society 4: 1 67-1 71 .
Hollander E, Neville D, Frenkel M, Josephson S, and Liebowitz MR (1992) Body dysmorphic
disorder. Diagnostic issues and related disorders. Psychosomatics 33: 156-165.
Ishigooka J, Iwao M, Suzuki M, et al. (1998) Demographic features of patients seeking
cosmetic surgery. Psychiatry and Clinical Neurosciences 52: 283-287.
Neziroglu F and Khemlani-Patel S (2002) A review of cognitive and behavioral treatment
for body dysmorphic disorder. CNS Spectrums 7: 464-471.
Phillips KA (1991) Body dysmorphic disorder: The distress of imagined ugliness.
American Journal of Psychiatry 1 48: 1 1 38-1 1 49.
Phillips KA (1996) The Broken Mirror: Understanding and Treating Body Dysmorphic
Disorder. New York: Oxford University Press.
Phillips KA, Albertini RS, and Rasmussen SA (2002) A randomized placebo-controlled
trial of fluoxetine in body dysmorphic disorder. Archives of General Psychiatry
59(4): 381-388.
Phillips KA, Grant JE, Siniscalchi J, and Albertini RS (2001) Surgical and
nonpsychiatric medical treatment of patients with body dysmorphic disorder.
Psychosomatics 42: 504-510.
Sarwer DB, Wadden TA, Pertschuk MJ, and Whitaker LA (1998) Body image
dissatisfaction and body dysmorphic disorder in 100 cosmetic surgery patients.
Plastic and Reconstructive Surgery 1 01 : 1 644-1 649.
Veale D, Boocock A, Gourney K, et al. (1996) Body dysmorphic disorder: A survey of
fifty cases. British Journal of Psychiatry 169: 196-201 .
Relevant Website
http://www.ocfoundation.org - International 0CD Foundation.
Body Image
N L Burke, L M Schaefer, and J K Thompson, University of South Florida, Tampa, FL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Body image A generic term used to describe a variety of
specific phenomena that relate to an individual's perception
of his or her body.
Body image disturbance Any form of affective, cognitive,
behavioral, or perceptual disturbance that is directly
concerned with an aspect of physical appearance.
Body mass index (BMI) This is the most widely used index
of weight status. It is a ratio of weight (kg) per height
squared (m). Cutoffs for weight status in adults are:
underweight (19.99 or less), average weight (20-24.99),
overweight (25-29.99), obese (30 and above).
Incidence The number of new cases of a disease or disorder
in a defined population during a specific period of time.
Meta-analysis A method of combining effect sizes across
studies with similar independent variables and outcomes.
Prevalence The number of cases of a disease or disorder that
exists in a defined population at a specified time.
Introduction
Body image is a subjective, self-evaluation of appearance. It can
have a profound effect on how one perceives his or her world
and, among other things, can affect an individual's self-esteem,
mood, eating behaviors, and social interactions. Negative body
image is often described as body image dissatisfaction or body
image disturbance. The two terms are often used interchange-
ably; however, some experts highlight a very important distinc-
tion. Body image dissatisfaction is just one component of body
image disturbance. The latter is a combination of body-related
dissatisfaction, significant emotional distress as a result of the
dissatisfaction, and impairment in psychosocial functioning.
An extreme form of body image disturbance is body dysmor-
phic disorder.
Body image disturbance used to be considered a phenome-
non that occurred in primarily Western societies, but this trend
has shifted in recent decades with the export of Westernized
culture and fashion trends via television, magazines, and
the Internet. Unfortunately, this has likely contributed to the
increased prevalence of body image dissatisfaction and body
image disturbance throughout the world. A recent global study
reconfirmed these results. The study consisted of females aged
15-64 from 10 countries: Brazil, Canada, China, Germany, Italy,
Japan, Mexico, Saudi Arabia, the United Kingdom, and the
United States. This investigation found that appearance-related
body image disturbance is indeed a global phenomenon: 67% of
females avoid doing basic activities (e.g., going to work/school,
voicing their opinion, engaging in physical activity, etc.) because
they feel badly about the way they look and 90% of females want
to change the way they look with the greatest dissatisfaction
stemming from body weight and shape followed by height.
Weight, shape, and height are the primary areas of body
image dissatisfaction for females worldwide, but there is varia-
tion among what aspects of appearance they most want to
change after that. The variation is partially dependent on tradi-
tional, culturally accepted ideals of beauty and partially depen-
dent on more contemporary, cross-cultural comparisons of
beauty. Depending on country, females have varying levels of
desire to change their hair, skin complexion, eye shape, eye
color, and the overall appearance of their face.
Although much of the research in this area has focused on
adult women, recent investigations of younger women, boys,
and men indicate that there is also a high level of disturbance
in these groups. For men and boys, the particular dimension
of dissatisfaction appears to primarily focus on muscularity
concerns, rather than level of body fat or overall size. In particu-
lar, over the past lOyears, a relatively new subtype of body
dysmorphic disorder, muscle dysmorphia, has been described
in males who have extreme concerns regarding their level of
muscularity.
History
Body image has a rich history in the fields of philosophy,
medicine, and psychology. The term body image originated
during the late 1800s. Neurologists, with the help of Gestalt
psychologists, were trying to understand phenomena that alter
body perception as a result of traumatic brain insult or injury.
They concluded that there may be an internal mental represen-
tation of body information, which they called body image. The
concept of body schema is credited to Head, one of the foremost
researchers at that time. The concept arose from the discovery
that sensory changes accompanied brain lesions.
Although Gestalt psychologists were part of the early discov-
eries in body image, neuropathology dominated the investiga-
tion of body experience. Psychology and psychological variables
were rarely considered until Schilder, a neurologist, reconcep-
tualized body image as a multidisciplinary field of study.
Schilder argued that body image should be studied from neuro-
logical, sociocultural, and psychological viewpoints. Since that
time, psychology has embraced the study of body image, and it
has been studied from the social, developmental, and clinical
perspectives.
Within clinical psychology, body image has been studied
from several schools of thought. For decades, psychodynamic
theory was the predominant lens through which body image
was conceptualized and researched. One of the leading psy-
chodynamic researchers, Fisher, pioneered the theory of body
image boundaries. According to the theory, the body is a
boundary between the self and the external world. Projective
365
366 Body Image
techniques were used to test the concepts of barriers and pene-
tration, which were considered measures of the strength and
permeability of the body image boundary. Shontz was a key
figure in shifting the focus of body image from a psychodynamic
viewpoint to one that focused more on the physical aspects of the
body. He suggested body image be studied from the viewpoints
of cognitive theory, field theory, and Gestalt psychology. Since
Shontz's time, the study of body image has almost exclusively
remained in the realm of physical aspects of the body.
The majority of current research focuses on physical appear-
ance, body weight, and body shape concerns. There are several
theories that seek to explain body image and body image dis-
turbance within these contexts. The following section offers a
brief review of contemporary theories of body image distur-
bance, as well as emerging evidence in support of each theory.
Theories of Body Image Disturbance
Sociocultural Approach
The sociocultural perspective is an approach to understanding
human behavior that focuses on how the values of society
affect the values and behaviors of individuals. A sociocultural
theoretical model of body image and body image disturbance
posits that current societal standards of beauty emphasize the
importance of thinness for women and muscularity for men.
While this ideal is omnipresent and regularly reinforced
through media, parents, and peers, it is often very difficult to
achieve, if not completely unattainable.
Social comparison theory is a subcomponent of a socio-
cultural approach that proposes that humans engage in con-
tinuous comparisons of themselves with others in their social
environment as a process of collecting information about
one's standing on some dimension. Within the context of
body image, comparison of one's weight, shape, and overall
appearance to that of others provides information about
that individual's relative appearance. Research has shown that
while women frequently make body-focused comparisons,
individuals vary in the degree to which they engage in social
comparisons of appearance and in their tendency toward
upward or downward comparisons. As would be anticipated,
research has suggested that individuals with greater body dissat-
isfaction tend to engage in more appearance-focused compar-
isons. When upward comparisons are made, they are associated
with an increase in negative affect, body dissatisfaction, and
thoughts of exercise, especially among women who report
high preexisting levels of body dissatisfaction. A recent meta-
analysis of 170 studies revealed a significant relationship
between social comparisons and increased body dissatisfaction
(d= 0.77), which was strongest among young women, com-
pared to men and older women. This indicates a strong
and reliable link between appearance-based comparisons and
feelings of dissatisfaction with one's body.
Media exposure studies, in which participants view depic-
tions of the thin or muscular ideal, have also provided support
to the sociocultural perspective of body dissatisfaction. A meta-
analysis on the role of media in body image concerns among
women brought together the findings of experimental and
correlational research to examine the effect of media exposure
on internalization of the thin ideal and body dissatisfaction.
Figure 1 Tripartite influence model of body dissatisfaction.
Reproduced from Thompson JK, Heinberg LJ, Altabe M, and
Tantleff-Dunn S (1999) Exacting Beauty. Washington, DC: American
Psychological Association, with permission from American
Psychological Association.
Using a sample of 77 studies (141 effect sizes), the authors
found strong support for the negative impact of exposure to
media depictions of the thin ideal on each of these domains.
The tripartite model of body image is one well-supported
socioculturally based approach for understanding how par-
ticular formative influences operate to produce body image
problems. In this model, three primary influences (peers, par-
ents, and media) are thought to directly and indirectly contrib-
ute to body image problems (see Figure 1). The indirect, or
mediated pathways, include an influence through appearance
comparison and internalization of societal values regarding
appearance norms.
Interpersonal Approach
An interpersonal approach to understanding body image
disturbance focuses on the influence of relationships and
person-to-person interactions on the development of body dis-
satisfaction. While sociocultural theory tends to look at macro-
level influences (e.g., society, the media), an interpersonal
perspective takes a more micro-level approach to examine
more proximal subcultural influences on body dissatisfaction.
Research has investigated the impact of numerous sources
of interpersonal influence, including parents, peers, romantic
partners, and personal acquaintances or strangers.
A wealth of research has focused on the influence of
appearance-related commentary and has consistently demon-
strated the detrimental effect of receiving negative comments
about one's appearance. Moreover, actual verbalization of the
negative appearance evaluation is not necessary to produce
a negative impact. Research has demonstrated that subtle ges-
tures or negative looks (i.e., nonverbal communication) can
convey this negative appearance evaluation as well and may
have a negative effect on the recipient of the evaluation.
Finally, strong evidence is mounting in support of the role
of appearance and weight-related teasing during childhood
and adolescence in the development of body dissatisfaction.
Numerous studies have investigated both the frequency and
effect of such teasing, as well as the differential impact
of various sources of teasing (e.g., family, peers). A new
meta-analysis compiled the findings of 58 studies, revealing a
reliable moderate association between weight-based teasing
Body Image 367
[d= 0.39) or appearance-based teasing (d = 0.32) and current
levels of body dissatisfaction.
Cognitive-Behavioral Approach
As the name implies, the cognitive-behavioral perspective
on body image draws from both cognitive and behavioral
paradigms to create a more integrated and comprehensive
understanding of body image and its related processes. The
cognitive-behavioral model first distinguishes between histor-
ical or developmental influences on body image and more
proximal events and process. Historical influences refer to
personal attributes or past experiences that lay the foundation
for the way a person thinks, feels, or acts in relation to his
or her body. These historical influences include cultural social-
ization, interpersonal experiences (e.g., being teased about
weight), physical characteristics, and personality attributes
(e.g., perfectionism). Each of these factors contributes to
one's body image schema (i.e., the cognitive organization of
the information about one's physical self), which includes
both body image investment (the perceived importance of
the physical self to overall self-evaluations) and body image
evaluations (satisfaction or dissatisfaction with one's body).
Proximal events and process refer to current life events and
cognitive processes that activate or maintain appearance
schemas. This may include precipitating events such as standing
in front of a mirror, receiving an appearance-related comment,
or feeling that one's body is being exposed or evaluated. Such
events then activate internal dialogues regarding one's appear-
ance, which may include emotion-laden thoughts, inferences,
or interpretations about one's appearance. Regarding indi-
viduals with body image disturbance, these dialogues are often
automatic, inaccurate, and upsetting. Notably, the cognitive-
behavioral perspective is the most widely accepted explanation
for the formation of individuals' body image. Moreover, there is
strong empirical support for the efficacy of cognitive -behavioral
therapy (CBT) in the treatment of body image disturbance
among community and clinical samples.
Feminist Approach
The feminist perspective of body image disturbance has, in
part, grown out of the historical gender disparities in the rates
of body dissatisfaction and eating disorders. Feminist theory
notes that body image disturbance is significantly more preva-
lent among women than among men and suggests that these
differential rates are born out of differences in the way that
society views women and their bodies. While significant diver-
sity in feminist approaches to understanding body image dis-
turbances exists, at its core feminist theory contends that
Western culture places an inordinate amount of value and
emphasis on women's bodies and their appearance. Moreover,
this culture teaches women to equate physical attractiveness
with self-worth. Within this appearance-focused environment,
women over-identify with their bodies, and their self-esteem
becomes contingent upon conforming to the prevailing stan-
dards for thinness and attractiveness.
One notable and increasingly studied contribution to
the understanding of the etiology of body dissatisfaction is
objectification theory. This theory asserts that women are
consistently sexually objectified through multiple modalities
including sexual harassment, leering glances, and exposure
to both insidious and explicit sexualized media portrayals of
women. As a result of this constant sexual objectification from
the social environment, girls learn to view themselves from a
third-person perspective and subsequently internalize that out-
sider's perspective as their own primary view of themselves
(self-objectification). This leads to habitual body monitoring,
which in turn, leads to increased body shame, body anxiety,
lowered bodily awareness, and finally, to body dissatisfac-
tion. Indeed, the role of self-objectification in body image
disturbance and women's disordered eating has recently
received empirical support, although further research in this
area is warranted.
Risk Factors for Negative Body Image
There are several risk factors that influence the development
of body image dissatisfaction and body image disturbance.
This section will focus on four of those factors: biology, family,
peers, and media.
Biology
Biological factors may affect body image dissatisfaction
in three general ways. First, certain neurological disorders
can lead to misperceptions of body image. One of the most
intriguing examples of this is the phantom limb phenomenon.
With phantom limb phenomenon, a person who has lost a
body part (e.g., by way of accident or surgical removal) still
feels - mentally and physically - that the body part is still there.
Second, certain naturally occurring body parts or features
may contribute to disturbances with body image. Examples of
this are disproportionately large breasts, a pregnant body,
or certain physical disabilities. Third, biological characteristics
that are outside of social norms may affect body image, for
example, physical deformity.
Family
In general, parents are thought to influence their child's body
image in two ways. The first way is that parents' dissatisfaction
with their own bodies may indirectly affect the child. For
instance, parents model certain weight and body behaviors
(e.g., negative comments on their own appearance, dieting
behaviors) that the child then internalizes. The second way
that parents may influence their child's behavior is by making
direct comments to the child about his or her weight or appear-
ance. When parents make critical or derogatory comments
about the child's body or encourage the child to lose weight,
he or she is more likely to have a poorer body image.
Most of the work regarding family and body image has
focused on the parent-child relationship. Far less research
has been conducted on sibling relationships. This is unfortu-
nate as it stands to reason that siblings would play an impor-
tant role in the development of body image in similar ways as
parents, that is, indirectly by modeling or directly by targeted
critical (or positive) comments. The few studies that include
368 Body Image
siblings focus on the frequency and effect of weight- and
appearance-based teasing by family members. (The majority
of those studies include only female participants.) Researchers
have found that the greatest amount of teasing within families
is perpetrated by siblings. Approximately 20-30% of siblings
engage in weight- or appearance-based teasing.
Peers
Peers can play a significant role in body image development
and maintenance. Childhood and adolescence are particularly
vulnerable times for influence by peer groups. One of the
reasons is the amount of peer-related teasing. Estimates of
weight- and appearance-based teasing by peers are as high as
60%. This is problematic as some researchers have established
a causal link between weight and shape-related teasing and
body image dissatisfaction.
Although it is not clear if people seek groups with similar
views on body image or converge on similar body image views
over time, it is known that peer groups tend to share ideas
on body image and similar levels of body dissatisfaction.
In addition, peers are frequently the targets of body compari-
son. Individuals often compare themselves to the weights
and shapes of their peers and use these comparisons to rate
their place in the subjective appearance hierarchy.
Media
Media plays a significant role in the development of body
image disturbance for certain individuals. In advertisements
and fashion magazines, women are extremely thin and tall;
men are increasingly overly muscular. The images of both are
often further enhanced by graphic alteration of images (i.e.,
digitally enhanced or reshaped). Even young children are
exposed to images of exaggerated, unobtainable body shapes
from some of their toys, dolls, and action figures. These idea-
lized images are portrayed as beautiful and successful, but
the actual body shapes and sizes are not attainable for the
majority of individuals. Studies show that women exposed to
thin media images report significantly more negative body
image compared to women not exposed to these images in a
laboratory setting. The duration of the effect is not known, but
some theorize a dose-response effect; more prolonged expo-
sure may result in greater body image disturbance. While some
have found support for this, others have found it only holds
true with certain types of media. Still other researchers propose
that the association between the media and body image is
complex and may be moderated by several variables including
age, engagement in social comparison, and initial body image
dissatisfaction. Specifically, effects of the media are higher for
those who are younger, tend to engage in social comparisons,
and have higher levels of initial body image dissatisfaction.
Gender Differences
Both males and females become aware of body image at a
very young age, typically between the ages of 4 and 6. During
this time, there do not appear to be any gender differences in
appraisals of body image. Differences emerge in preadoles-
cence with a larger percentage of girls reporting body image
dissatisfaction than boys. Notably, recent studies have con-
cluded that this differential may be changing on at least one
aspect of appearance - the desire for thinness. There are now
similar proportions of preadolescent boys and girls desiring to
be thinner, which may be an artifact of the growing problem of
childhood overweight and obesity.
Puberty is considered a pivotal time period for the develop-
ment of body image disturbance. During this time, females
have marked body changes like weight gain and widening
hips. These changes are in contrast to societal pressures to
obtain the thin ideal and result in a discrepancy between real
and ideal images of self. This discrepancy is considered to be
one of the greatest precursors to body image disturbance in
females. For both sexes, being overweight or obese is associated
with higher body dissatisfaction, but as adolescent males reach
puberty, being underweight and lacking muscularity are often
catalysts for body image dissatisfaction. In this way, in contrast
to females, weight gain in adolescent males is not necessarily
as taboo. Males are as equally likely to want to be thinner as
they are to be larger and more muscular. Overall, body image
dissatisfaction is significantly greater in females than males
during this time.
In adulthood, body image dissatisfaction and body image
disturbance remain considerably higher for females compared
to males. Of note, however, the prevalence of negative body
image has increased for both genders over the past few decades.
Nevertheless, perhaps one of the single greatest differences
between the genders in the field lies within research method-
ology. The overwhelming amount of research in the field of
body image has been conducted with females. The amount of
research conducted with males is increasing, but is still regret-
tably far behind that of their female counterparts.
A relatively new area of research on gender differences is
on body image and sexual orientation. According to the earliest
research in the area, there seems to be an interaction between
gender and sexual orientation. As a group, lesbians are less
dissatisfied with their bodies than heterosexual females. Con-
versely, gay males are more dissatisfied with their bodies than
heterosexual males. More recent research in the area supports
the difference between gay males and heterosexual males, but
has not supported a difference in body image dissatisfaction
between lesbians and heterosexual females. Clearly, more
research is needed to clarify the relationships between sexual
orientation, gender, and body image dissatisfaction.
Assessment of Body Image and Body Image
Disturbance
In recent years, the field of body image has undergone
remarkable growth, which has been partially manifested in
the expansion of assessment techniques and measures aimed
at evaluating various dimensions of body image. While percep-
tual measures of body image were once widely used, there is
currently considerable debate regarding whether these mea-
sures represent a distinct aspect of body image and, over
time, newer techniques have largely eclipsed the use of percep-
tual measures. Current measures generally focus on the
Body Image 369
attitudinal component of body image (as opposed to the per-
ceptual component) and can be classified into three dimen-
sions: subjective and affective, cognitive, and behavioral. While
it is impossible to cover the full breadth of available measures
of body image, the following offers a description of each of
these categories as well as a brief review of relevant measures.
Subjective and Affective Measures
Subjective and affective measures essentially assess satisfaction
with one's body. Such measures may range from a global rating
of overall body satisfaction, to slightly more specific assessment
of weight or shape satisfaction, and on to measures of very site-
specific satisfaction (e.g., satisfaction with one's thighs, hips,
chest). Subjective and affective measures can be further subdi-
vided into figural rating methods and questionnaire measures.
The former involves presenting an individual with line drawings
of human silhouettes ranging in size and shape (e.g., very thin to
very overweight). The individual is instructed to indicate the
figure that represents their ideal body as well as the figure that
represents their current size. Discrepancy scores (i.e., the differ-
ence between these two figures) provide an index of body
dissatisfaction. The Contour Drawing Rating Scale is a com-
monly used figural rating scale with both male and female
figures that range from underweight to overweight. The soma-
tomorphic matrix is a computerized measure that presents a
more lifelike two-dimensional figure and allows for assessment
of dissatisfaction regarding overall body fat and muscularity for
both men and women.
Questionnaire measures vary widely in their level of
specificity regarding body image concerns. Some provide an
estimate of global body dissatisfaction, while others measure
satisfaction with specific areas of the body. The Appearance
Evaluation Subscale of the Multidimensional Body Self-
Relations Questionnaire is a widely used 7-item measure of
overall body dissatisfaction that has been used with both men
and women of various ethnicities. An example statement from
this scale is 'I like my looks just the way they are.' The Body
Shape Questionnaire is a 34-item global assessment measure,
which taps overall concerns with one's shape and weight.
Finally, the Body Esteem Scale for Adolescents and Adults
and the Body Image subscale of the Self-Image Questionnaire
for Young Adolescents offer a global measure of body satisfac-
tion for adolescents.
As discussed, other measures provide a more site-specific
assessment of body dissatisfaction. The Body Areas Satisfac-
tion Scale (BASS), a subscale of the Multidimensional Body
Self-Relations Questionnaire, asks individuals to rate their
degree of satisfaction with their overall appearance and with
eight specific areas/attributes (e.g., face, hair, lower torso,
mid-torso, upper torso, muscle tone, height, and weight).
Additionally, the Body Uneasiness Scale, which has been
used for adolescent and adult men and women, contains 34
items that assess overall body dissatisfaction, as well as 37
items that assess more specific concerns about individual
body parts. Finally, the Physical Appearance State and Trait
Anxiety Scale addresses the affective component of body
image and assesses both trait and state anxiety regarding 16
body sites. A sample item from the state version of this scale is
'Right now I feel anxious, tense, or nervous about my thighs.'
Cognitive Measures
Cognitive measures of body image attempt to address the
thoughts, beliefs, attributions, and attitudes related to an
individual's appearance. For example, the Appearance Schemas
Inventory-Revised is a 20-item scale that assesses beliefs
and assumptions about both the importance (Self-Evaluative
subscale) and influence (Motivational Salience subscale) of
appearance in one's life. The Beliefs About Appearance Scale
(BAAS) contains 20-items that assess an individual's beliefs
regarding consequences of their appearance on relationships,
achievement, self-perception, and emotions. Sample items
from the BAAS include: 'How I feel about myself is largely
based on my appearance,' 'My relationships would improve
if I looked the way I wished,' and 'The opportunities that are
available to me depend upon how I look.' The Drive for Mus-
cularity Scale measures men's desire for a muscular appearance
and strategies to obtain this physique. Finally, the Internaliza-
tion subscale of the Sociocultural Attitudes Toward Appearance
Scale-3 assesses the degree to which an individual has inter-
nalized or accepted the thin ideal. A sample item from this
subscale of the SATAQ-3 is 'I would like my body to look like
the models who appear in magazines.'
Behavioral Measures
Behavioral measures of body image have received less atten-
tion than subjective, affective, and cognitive measures pri-
marily because of the largely intrapsychic nature of the
disturbance. In other words, while body image disturbance
can lead to behaviors aimed at avoiding, changing, or con-
stantly monitoring one's appearance, aspects of the distur-
bance are more readily tapped using self-report assessments
of the individual's attitudes and experiences. True objective
behavioral measures would ideally require an outside observer
to rate the person's body image-related behaviors such as
avoidance of mirrors or body-monitoring behaviors. Given the
often difficult and time-intensive nature of such assessments,
current behavioral assessments of body image rely almost exclu-
sively on subjective self-reports from individuals regarding their
frequency of various behavioral manifestations of body distur-
bance. For example, the Body Image Avoidance Questionnaire
assesses the frequency of body image avoidance behaviors
including avoiding being weighed or wearing very loose-fitting
clothing to avoid body exposure. Example items include:
"I don't wear 'revealing' clothes (e.g., bathing suits, tank tops,
or shorts)," "I do not go out socially if it involves eating," and
"I avoid physical intimacy." The Body Checking Questionnaire
is a more recent 23-item measure that assesses frequency of
body-checking behaviors related to overall appearance and spe-
cific body parts in adult women, while the newly developed
Male Body Checking Questionnaire (MBCQ) assesses these
behaviors in men. An example item from the MBCQ asks the
individual to rate how often he 'checks muscles in most reflec-
tive surfaces.'
Recent Advances in Assessment
Assessment measures have received wide utilization in a variety
of contexts, including basic and applied research in the field of
370 Body Image
eating disorders and body image disturbance. More recently,
there has been an application to relatively new topics, such as
the inclusion of body image assessment in the evaluation of
the effectiveness of pharmacological agents as a treatment for
eating disorders and body image problems. Additionally, there
has been a tremendous increase in the last decade in the use
of cosmetic and plastic surgery options for the modification of
the body, and researchers have begun to use selected measures
of body image to evaluate the role of these approaches as
strategies to modify body satisfaction.
Treatment and Prevention
Body image disturbance has often been treated using CBT. CBT
has been found to be effective in improving perceptual distor-
tions, body image evaluations, and body image dysphoria.
It improves other areas of psychosocial functioning such as
self-esteem, social anxiety, depression, and eating behaviors.
An empirically validated, manualized CBT treatment for body
image was developed by Cash and colleagues. The treatment
adheres to the general tenets of CBT, but is specifically tailored
to the challenges of those with body image disturbance.
Research on body image interventions with obese popula-
tions is relatively recent but extremely relevant given both the
scope of the obesity problem and the research that indicates
obese people report higher body image dissatisfaction and
distress than those of average weight. Treatment for body
image within obese populations occurs mostly within weight
loss interventions. Improvements in body image occur for
many, but not all, who lose weight. Weight loss is often not
sustained, however, and improvements in body image may be
lost. However, there is some evidence that body image may
be improved, even in the absence of weight loss. Some
researchers propose adding a body image treatment post
weight loss to aid in long-term weight maintenance. There is
also growing empirical support for CBT in the treatment of
body dysmorphic disorder. Evidence from randomized con-
trolled trials reveals significant reductions in body dysmorphic
disorder symptoms.
Although treatment of body image disturbance is impor-
tant, prevention is the way to decrease overall prevalence by
decreasing rates of incidence. Considering the early onset of
body image disturbance, most prevention programs have been
developed for and targeted to children and adolescents. As
with treatment, outcomes on the prevention of body image
disturbance are often part of larger studies on eating disorders.
These prevention programs are universal, selective, or targeted.
The goal of both universal and selective prevention programs is
to reduce the incidence of the disorder. Selective prevention is
aimed at asymptomatic, but at-risk individuals, whereas uni-
versal prevention programs are aimed at the population at
large including those not considered at risk for the disorder.
Targeted prevention programs are for subclinical populations
that have been identified as high risk because of clear precur-
sors or warning signs. Targeted programs are important for the
reduction of symptoms, but do not reduce incidence.
Researchers have found that universal and selective pro-
grams show small, but significant effects in the reduction of
body image dissatisfaction and internalization of the thin
beauty ideal, but targeted prevention programs are the most
effective at reducing body image dissatisfaction and thin-ideal
internalization. Targeted prevention programs for children
younger than high school age have yet to be developed even
though those are critical years in the development of body
image disturbance.
Stice and colleagues developed a targeted prevention pro-
gram for adolescent girls 14 years and older with elevated levels
of body image dissatisfaction. The cognitive dissonance pro-
gram consists of verbal, written, and behavioral activities
where the thin ideal is critiqued and challenged. The program
has been empirically validated and has been shown to reduce
internalization of the thin ideal, body image dissatisfaction,
negative affect, and psychosocial impairment. Comparative
outcome studies and follow-up studies confirm the superiority
of the cognitive dissonance program over other active treat-
ments, active controls, and no-treatment controls. Stice and
colleagues also developed another viable treatment alternative
called the healthy weight program which focuses on making
gradual and healthy diet, exercise, and lifestyle changes. The
program leads to reductions in body image dissatisfaction and
eating disorder symptomatology as well.
Conclusion
The field of body image has expanded greatly since its humble
beginnings in neurology, but there is still much to learn. In the
future, researchers should further investigate understudied
populations including elderly persons, young children, and
males. There should be a push not only to understand the
intricacies of various risk factors of negative body image, but
to understand protective factors and what leads to the devel-
opment of a healthy body image across the life span. Research-
ers should also continue to explore the role of culture and
country to appreciate the subtle or stark differences that will
aid in prevention and treatment efforts.
See also: Body Dysmorphic Disorder: A Review; Cognitive Behavior
Therapy; Interpersonal Psychotherapy.
Further Reading
Calogero R, Dunn S, and Thompson JK (2011 ) Objectification in Women. Washington,
DC: American Psychological Association.
Cash TF (2007) The Body Image Workbook: An 8-Step Program tor Learning to Like
Your Looks. Oakland, CA: New Harbinger.
Cash TF and Pruzinsky T (1990) Body Image Development, Deviance, and Change.
New York: Guilford.
Cash TF and Pruzinsky T (2002) Body Image: A Handbook of Theory, Research, and
Clinical Practice. New York: Guilford.
Fingeret MC, Warren CS, Cepeda-Benito A, and Gleaves DH (2006) Eating disorder
prevention research: A meta-analysis. Eating Disorders: The Journal of Treatment
and Prevention 14: 191-213.
Fisher S (1986) Development and Structure of the Body image. Hillsdale, NJ: Erlbaum.
Menzel JE, Schaefer LM, Burke NL, Mayhew LL, Brannick MT, and Thompson JK
(2010) Appearance-related teasing, body dissatisfaction, and disordered eating:
A meta-analysis. Body Image 7: 261-270.
Shroff HP, Calogero RM, and Thompson JK (2009) Assessment of body image. In:
Allison DB and Baskin ML (eds.) Handbook of Assessment Methods for Eating
Body Image 371
Behaviors and Weight-Belated Problems: Measures, Theory, and Research,
2nd edn., pp. 115-136. Los Angeles: Sage.
Smolak L and Thompson JK (2009) Body Image, Eating Disorders, and Obesity in
Youth: Assessment, Prevention, and Treatment, 2nd edn. Washington, DC:
American Psychological Association.
Stice E, Marti CN, Spoor S, Presnell K, and Shaw H (2008) Dissonance and healthy
weight eating disorder prevention programs: Long-term effects from a randomized
efficacy trial. Journal of Consulting and Clinical Psychology 76: 329-340.
Stice E, Shaw H, and Marti CN (2007) A meta-analytic review of eating disorder
prevention programs: Encouraging findings. Annual Review of Clinical Psychology
3: 207-231.
Thompson JK (2004) Handbook of Eating Disorders and Obesity. Hoboken, NJ: Wiley.
Thompson JK and Cafri G (2007) The Muscular Ideal. Washington, DC: American
Psychological Association.
Thompson JK, Heinberg LJ, Altabe MN, and Tantleff-Dunn S (1999) Exacting Beauty:
Theory, Assessment, and Treatment of Body Image Disturbance. Washington, DC:
American Psychological Association.
Relevant Websites
www.bodyimagedisturbance.org - Body Image Research Group.
www.womenshealth.gov/bodyimage - Body image: Loving your body inside and out.
Borderline Personality Disorder
J Paris, McGill University, Montreal, QC, Canada; Institute of Community and Family Psychiatry, Montreal, QC, Canada
© 2012 Elsevier Inc. All rights reserved.
Glossary
Affective instability (emotion dysregulation) A tendency
for mood to change rapidly in relation to environmental
cues.
Borderline personality disorder A disorder that affects
mood, impulse control, and interpersonal relationships over
many years.
Impulsivity A tendency to act on feelings rather than taking
the time to consider them.
Personality disorder Maladaptive personality traits
beginning early in life that have consistent and serious
effects on long-term functioning.
Clinical Features and Diagnosis of Borderline
Personality Disorder
Personality disorder (PD) describes one of the major forms of
psychopathology. In the DSM-IV-TR system (American Psychi-
atric Association, 2000), one first makes an overall PD diagno-
sis, and then determines whether a patient can fit into any of the
ten listed categories. PDs as a group are defined by maladaptive
personality traits, beginning early in life, that have consistent
and serious effects on long-term functioning. To make a diag-
nosis, one must establish that personality problems affecting
cognition, mood, interpersonal functioning, and impulse con-
trol account for dysfunction.
Borderline PD (BPD) differs somewhat from other PDs in
that it is associated with serious and distressing symptoms. And
since BPD is frequently seen in practice and is problematic for
management, it has been the subject of a very large body of
research.
The use of the term 'borderline' has led to some confusion.
It was originally based on a theory that psychopathology can
lie on a border between psychosis and neurosis. While that
idea is out of date, nobody has found a better label; so the term
has been retained.
BPD is a complex syndrome with dysfunction in mood,
impulse control, and interpersonal relationships. The com-
mon pattern for all BPD symptoms is instability. Mood is
characterized by intense and painful affective states that shift
rapidly - often within hours - leading to emotional turmoil
and enraged outbursts, with a loss of control. These affective
symptoms are associated with chronic suicidality, in which
patients can think about killing themselves on a daily basis
for years on end. BPD patients are also impulsive, and most
make repetitive suicide attempts, primarily by overdose.
About half of all patients seen in emergency rooms with
repeat suicide attempts can be diagnosed with BPD. These
incidents often follow disappointments in interpersonal
relationships.
BPD patients can show a wide range of other impulsive
behaviors, such as substance abuse, bulimia nervosa, sexual
promiscuity, and shoplifting. One of the most common fea-
tures is self-cutting - usually on the wrists or arms. But that is
not suicidal behavior. BPD patients rarely slash deeply; rather,
they cut themselves to relieve tension.
Interpersonal relations in BPD are typically stormy and
unstable. These patients have trouble being alone, and all too
quickly get involved with people, many of whom take advan-
tage of their neediness. Intimate relations are marked by seri-
ous quarrels, as patients require constant reassurance that they
will not be abandoned. It is, therefore, not surprising that
many of their relationships end in sorrow.
One of the least understood aspects of BPD is that about
half of patients show micropsychotic symptoms (transient
hallucinations, paranoid trends, or depersonalization). That
is the most 'borderline' aspect of the disorder. However, full-
blown psychotic episodes are uncommon.
To make a formal diagnosis of BPD, DSM-IV-TR lists nine
criteria but requires only five of them to be present. This 'poly-
thetic' structure makes for a heterogeneous diagnostic group.
Research definitions have been developed to narrow down
the scope of BPD, requiring that patients score high in all patho-
logical domains (affective instability (AI), impulsivity, cognitive
symptoms, and interpersonal relationships). DSM-5, scheduled
for 2013, will define BPD somewhat differently, based on a
rating as to whether cases have prototypical symptoms, supple-
mented by a trait profile.
PD can, in general, be understood as pathological amplifi-
cations of normal personality trait dimensions. One of the
most important traits underlying BPD is AI, also known as
emotion dysregulation (ED). AI is a construct that describes
ultra-rapid mood shifts, in which emotional states are highly
responsive to changes in the interpersonal environment. These
phenomena must be distinguished from bipolar disorder, in
which mood changes remain stable for days at a time and are
much less responsive to environmental triggers. When AI is
monitored with standardized instruments, emotions are found
to be intense, but reactive to circumstances, with more of a
tendency to anger than to grandiosity.
AI is a basic aspect of BPD that appears early in develop-
ment and which can still be found in patients who have recov-
ered from all other symptoms of the disorder. It appears to
have a heritable basis in temperament. However, the neurobi-
ological nature of AI remains largely unknown. Understanding
may benefit from the large body of recent research on the
psychology of emotions, as well as from experience sampling
methods that can measure emotional shifts in the course of
daily life.
372
Borderline Personality Disorder 373
The second fundamental trait behind BPD is impulsivity.
This personality dimension has also been the subject of a large
body of research. The impulsivity seen in BPD has been some-
times been thought to be secondary to AI. However, patients
who are only emotionally dysregulated (and have other mental
disorders) do not necessarily overdose, cut, or feel chronically
suicidal. Evidence from imaging and from neuropsychological
methods suggests that impulsivity is associated with defective
executive functioning (a failure of the prefrontal cortex to
modulate input from the limbic system), while neurochemical
studies have suggested that trait impulsivity is associated with
dysfunction of the central serotonergic system.
What is most characteristic of BPD is the combination of
AI with impulsivity. These trait dimensions account for most
aspects of its clinical presentation, including suicidality and
instability of interpersonal relations.
There is a very wide range of symptoms seen in BPD, many
ofwhich overlap other diagnoses (mood and anxiety disorders,
substance abuse, and eating disorders). This has led to a ten-
dency among clinicians to focus on 'comorbid' diagnoses, and
to withhold a diagnosis of PD. However, there are important
differences in how any of these symptoms develop when BPD
is also present.
The most common comorbidity in BPD patients is with
depression or dysthymia. However in BPD patients, depressive
symptoms are not episodic but continuous, that is, not neces-
sarily associated with extended and continuous periods of
low mood. Unlike classical melancholia, emotions can shift
rapidly with changes in environmental circumstances.
In the same way, mood swings in BPD differ greatly from
those in bipolar disorder, since patients do not show continu-
ously elevated mood (hypomania), but a pattern of rapid shifts
in affect related to life events, with 'high' periods that last only
for hours (rather than for days or weeks). Unfortunately, recent
enthusiasm for extension of the bipolar diagnosis has led some
clinicians to diagnose many BPD patients with mood swings as
falling within a bipolar spectrum.
BPD need not be mistaken for schizophrenia. Instead of
long-term psychotic symptoms, one sees 'micropsychotic' phe-
nomena of short duration, in which patients hear voices or
experience paranoid thoughts under stress, followed by long
periods in which these symptoms completely disappear.
BPD is a diagnosis with important implications for clinical
management. But it must be recognized that this construct, like
almost all other mental disorders, depends entirely on observ-
able signs and symptoms. Unlike general medicine, psychiatry
lacks specific markers associated with specific genes, neuro-
chemical changes, or brain imaging. It is possible that BPD is
not a single disorder but reflects multiple endophenotypes
leading to a common final pathway marked by a cluster of
psychological symptoms.
Epidemiology and Cross-Cultural Prevalence
Systematic epidemiological studies of mental disorders have
been carried out for several decades. However, data on the com-
munity prevalence of PDs has only emerged in the last 1 0 years.
Most of these community surveys indicate that the preva-
lence of BPD is close to 1%, a rate similar to that of
schizophrenia. While about 80% of patients receiving treat-
ment are women, gender differences have not been found in
most community samples. The explanation is that men with
BPD are less likely to seek treatment. This supposition is sup-
ported by psychological autopsy studies that have found BPD
to be common in young males who commit suicide; most of
these patients were never seen in the mental health system. The
other notable demographic is that, as is the case for PDs in
general, BPD is more common in patients with a lower socio-
economic level and lower levels of education.
There have been no systematic studies of BPD across cul-
tures. The disorder has been described by psychiatrists in large
urban centers in countries such as India and China, but its
community prevalence in these locations is not known. How-
ever, indirect evidence suggests that BPD is more common in
modern societies undergoing rapid social change. There has
been an increase over the last few decades in developed
countries in many of its associated symptoms (youth suicide,
self-harm, suicide attempts).
Even if the vulnerability to BPD is rooted in universal trait
dimensions, prevalence could vary across cultures. The expla-
nation is that mental disorders present different symptoms in
different societies. Similar illness processes need not lead to
identical clinical presentations, since specific pictures can be
determined by mechanisms of social contagion. Many psycho-
logical problems express themselves as physical symptoms,
and patients may not report abnormal mental states. More-
over, impulsive ('externalizing') symptoms are less common in
traditional societies than are 'internalizing' symptoms. Thus, it
is not really surprising that behaviors such as cutting and
repetitive overdosing do not seem to have been common his-
torically, but have become more frequent in modern urban
societies.
Etiology: Biological, Psychological, and Social
We are only beginning to understand the causes of BPD. As is
the case for most mental disorders, no single factor can explain
its development. Instead, multiple risk factors (biological, psy-
chological, and social), in interaction with each other, seem to
play a role.
The biological vulnerability for PDs depends on tempera-
mental (inborn and/or heritable) characteristics that can be
assessed in adulthood as personality traits, that is, stable pat-
terns of thought, affect, and behavior that characterize indivi-
duals, and that persist over time. Behavioral genetic research
shows that nearly half the variability in personality traits can be
accounted for by heritable influences.
Both AI and impulsivity have a similar genetic component.
Moreover, most twin studies have found that BPD itself shows
a level of heritable influence of about 50%. While it is unusual
for BPD itself to be found in first-degree relatives of probands,
genetic influences are confirmed by family history studies
showing that impulsive disorders such as antisocial personality
and substance abuse, as well as mood disorders such as depres-
sion, are common.
A heritable vulnerability to BPD would be expected, in
principle, to be associated with biological markers. But neu-
roscience is at an early stage of development, and most
374 Borderline Personality Disorder
pathological behavioral patterns are not consistently associated
with any measurable changes in brain function. No markers
specific to the diagnosis of BPD as a whole have been identified.
The most robust findings thus far have come from studies of
central neurotransmitter activity, which have shown that impul-
sive traits in these patients tend to be associated with deficits in
central serotonergic functioning. While some initial research
on AI has been conducted, its biological correlates are largely
unknown.
Age of onset can provide clues about the development of
the disorder. BPD usually begins around puberty, but first
presents clinically at a mean age of 18. (Most patients experi-
ence symptoms for a number of years without seeking help.)
An adolescent onset suggests that biological changes during
puberty (hormonal shifts and synaptic pruning) might convert
a temperamental vulnerability into an overt disorder. But it
might also suggest that when a diathesis exists, the develop-
mental stressors associated with adolescence can lead to the
emergence of overt psychopathology.
There has been a great deal of research on the psychological
factors leading to BPD. The results have been striking but not
really consistent. Many patients describe experiences reflecting
family dysfunction and psychological trauma during child-
hood. Tfowever, other patients state that they felt normal
before puberty.
Statistically, one finds a high frequency of reported trau-
matic events (sexual or physical abuse, parental neglect, family
violence) during childhood in patients who develop BPD.
There is also support for this relationship from prospective
community studies that have examined the precursors of PD
symptoms. Even so, one should not conclude that all patients
with BPD have had a traumatic childhood, since many do not
report such experiences.
Nor should one conclude that children who undergo
trauma are necessarily at risk for BPD. Longitudinal studies of
the outcome of adversities during childhood show that such
early traumatic experiences are risk factors, but not predictable
causes, of mental disorders. Moreover, community studies
document extensive resilience following trauma, particularly
following less severe adversities.
The relationship between childhood abuse and BPD has
aroused great interest. However, the most careful studies show
that only about a quarter of patients describe sexual abuse over
time from a caretaker (community studies show this to be the
most damaging type of trauma). In contrast, most incidents of
sexual abuse reported by BPD patients have occurred on one
occasion, and involve strangers, and about a third of patients
do not report any abuse at all. The data on physical abuse are
less consistent, although one prospective study of child abuse
cases that ended up in court found BPD to be more likely as
an outcome.
One crucial issue is that child abuse is a risk factor for mental
disorder in general rather than being specific to BPD. But since
trauma is a marker for a wide variety of other adversities, and is
strongly associated with family dysfunction, the effects may be
due more to overall upbringing than to specific events.
Another crucial point concerns interactions between
genetic vulnerability and adverse life experiences. By and
large, childhood trauma is most likely to produce pathological
sequelae in those who are temperamentally vulnerable.
These gene-environment interactions are crucial, and support
a stress-diathesis model of BPD.
Finally, social factors play a role in the development of
BPD. It is possible that modern societies, which require
young people to find their own social roles, may lower the
threshold for a disorder in which mood becomes unstable,
impulsive actions common, and identity uncertain. It is also
possible that the propensity to BPD takes other forms in other
societies, and that the specific picture seen in developed
countries is a cultural product.
In summary, none of the risk factors for BPD are necessary
conditions for the development of the disorder. The sufficient
conditions require a combination of biological, psychological,
and social vulnerability.
Course and Prognosis
Most patients with BPD improve with time. Follow-up studies
show that by age 35-40, about 75% function at close to nor-
mal levels. By the age of 50, 90% have recovered from the
disorder. Prospective studies have shown that many patients
stop meeting diagnostic criteria after a few years, when they
stop overdosing and cutting. Social functioning does not
improve that rapidly, and full remission usually takes many
years. Even so, the fact that most patients improve - and never
commit suicide in spite of having threatened to end their lives
on multiple occasions - suggests that BPD has a more positive
prognosis than previously thought. It is not necessarily a life-
long illness, but most often a disorder of youth.
Nonetheless, about one out of ten BPD patients will even-
tually commit suicide. The problem is that this outcome is
notoriously difficult to predict. But by and large, the patients
who die by suicide are those who fail to recover. The peak for
completion takes place after age 35, usually after a series of
unsuccessful treatments. On the other hand, younger BPD
patients are not at so high a risk, so that clinicians who feel
alarmed by their dramatic suicidal threats and actions should
feel reassured.
The mechanism of improvement in BPD over time is not
fully understood. Impulsivity generally decreases with age in
everyone. Moreover, many BPD patients learn over time how
to avoid the situations that give them the most trouble (intense
intimate relationships). Most eventually find stable social and
occupational niches that provide them with the structure they
need. Once they stop acting impulsively, they no longer meet
diagnostic criteria. The minority of patients who fail to find
stable employment may remain in the mental health system.
Finally, even in those who function well, some traces of AI can
be observed.
Nonetheless, earlier perceptions that BPD is an incurable
condition have been disproven, and patients should be in-
formed that they are likely to improve. They can also be told
that treatment may make recovery proceed more rapidly.
Treatment
BPD is a therapeutic challenge. However, recent research has
shown that treatment specifically designed for the disorder
Borderline Personality Disorder 375
can be successful. While this is a disorder that can remit with
time, treatment effects that can be documented over shorter
periods, and which are superior to control conditions, are
likely to be truly therapeutic.
In modern psychiatry, most patients tend to be managed
with drugs. However, no pharmacological agents specific for
the symptoms of BPD have been developed. Thus, patients are
generally prescribed agents that were originally developed for
other purposes, such as antidepressants, mood stabilizers, and
antipsychotics.
Quite a few randomized controlled trials of pharmaco-
therapy for BPD have been published. But this research suffers
from a number of defects: small sample size, attrition, as well
as durations too short to assess treatment of a chronic disorder.
Thus most of these studies have been conducted on less than
20 subjects observed for only a few months. Finally, outcomes
in these studies are generally based on self-report, and one
cannot tell whether the clinical picture has shown full
remission.
In general, clinical trials document only a mild degree of
symptomatic improvement with drug therapy. There have not
been enough studies to merit a meta-analysis, and neither
the British National Institute of Clinical Excellence (NICE)
guidelines nor the Cochrane report has found enough data to
recommend any specific agent. However, while drugs are over-
prescribed for BPD patients, they are sometimes useful to
control symptoms. A number of agents, including low-dose
atypical neuroleptics, specific serotonin reuptake inhibitors,
and mood stabilizers, can reduce impulsive behaviors. However,
antidepressants are much less effective for mood symptoms
than in patients who have no PD. Again, pharmacological
agents only 'take the edge off' symptoms, and do not produce
remission of BPD.
The most evidence-based method of treatment for BPD is
psychotherapy. However, ordinary methods (sometimes called
'treatment as usual') are unsuitable for these patients. In gen-
eral, the best results are associated with treatments specifically
designed for this group, rather than general methods that can
be useful for other clients.
The evidence-based psychotherapies for BPD are highly
structured and have specific elements designed to manage
symptoms. 'Dialectical behavior therapy' (DBT) is a form of
cognitive-behavioral therapy targeting AI and impulsivity,
using group and individual sessions to teach patients how to
regulate their emotions. In eight clinical trials comparing the
method to various types of standard treatment, DBT was more
effective in bringing suicidal behaviors under control within a
year. However, it is not clear whether effects are entirely specific
to the method, or whether they reflect the advantage of using a
highly structured treatment. One study that compared DBT
with a manualized form of psychiatric management found no
difference in outcome.
While DBT has the most evidence supporting it, two clinical
trials point to the efficacy of 'mentalization-based treatment'
(MBT), a therapy that combines cognitive and psychoana-
lytic techniques. MBT, like DBT, teaches patients how to
self-observe interpersonal situations so as to reduce levels of
conflict. These studies showed an advantage for MBT over
treatment as usual, both in a day hospital setting, and in an
outpatient clinic.
While there have been a few successful trials of other psy-
chotherapy methods for BPD, again suggesting that structured
therapies work better than treatment as usual, the main con-
tenders are DBT and MBT. Although there are insufficient data
for meta-analysis, the evidence was sufficient for a recommen-
dation for some form of cognitive therapy in both the NICE
and Cochrane guidelines.
Unfortunately specialized forms of psychotherapy for BPD
are expensive, since they require highly trained personnel and a
dedicated team. In most settings, these specific therapies are
not readily available.
Nonetheless, psychotherapists conducting a more general
practice can learn from evidence-based approaches. It is not
enough to sympathize with distress or 'put out fires' every time
a crisis arises. It is also not helpful to focus too much on past
events. BPD requires a practical and structured approach that
teaches patients how to manage emotional dysregulation and
impulsivity.
Chronic Suicidality in BPD
One of the main qualifications for therapists treating BPD
patients is a thick skin. Chronic suicidality is a troubling prob-
lem. But many of the methods that have been used to handle
suicidal threats and actions were developed for acute suicidal-
ity associated with episodes of depression. In BPD, these
approaches are not effective, and they can even make matters
worse.
Clinicians are sometimes advised to hospitalize patients
who threaten to commit suicide. In BPD, this often leads to
unproductive and regressive admissions. These patients need
to learn life skills, which is not possible under a suicide watch
in a hospital ward. There has been little research on the effec-
tiveness of hospitalization, no evidence that admission pre-
vents completion of suicide, and no evidence that it is an
effective treatment modality that can change the course of the
disorder.
Suicidality in BPD peaks when patients are in their early
20s. But as we have seen, completions are most common after
age 30, in patients who fail to recover. Thus, impulsive over-
doses, most often seen in younger patients, do not carry a high
short-term risk. These behaviors need to be understood as
communicating distress. Self-harming behaviors such as
chronic cutting are not even truly 'suicidal.' Given that these
behaviors are an attempt to regulate dysphoric emotional
states, it is the task of therapists to teach patients better ways
of doing so.
In summary, therapists should not be paralyzed by fear of
chronic suicidality. Instead, they should respect the need that
BPD patients have to contemplate an 'exit' when their life is
going badly, and focus on modifying the emotions and beha-
viors that feed hopelessness.
Conclusions
We need more research to understand the causes and treat-
ment of BPD. If we understood the etiology and pathogenesis
of this disorder better, we would be in a position to develop
376 Borderline Personality Disorder
more evidence-based approaches to therapy. However, an
important start has been made, and we can feel hopeful
about most cases seen in practice. Management of these
patients has improved much over the last few decades. While
not all will recover completely, we can help most of them.
See also: Big Five Model and Personality Disorders; Psychotherapy;
Suicide.
Further Reading
Bateman A and Fonagy P (2006) Mentalization Based Treatment: A Practical Guide.
New York: John Wiley.
Bateman A and Fonagy P (2009) Randomized controlled trial of out-patient
mentalization based treatment versus structured clinical management for borderline
personality disorder. American Journal of Psychiatry 166: 1355-1364.
Binks CA, Fenton M, McCarthy L, Lee T, Adams CE, and Duggan C (2006)
Pharmacological interventions for people with borderline personality disorder.
Cochrane Database of Systematic Reviews, Issue 1. Art No. CD005653.
D0I: 10.1 002/1 4651 858.C D005653.
Binks CA, Fenton M, McCarthy L, Lee T, Adams CE, and Duggan C (2006)
Psychological therapies for people with borderline personality disorder. Cochrane
Database of Systematic Reviews, Issue 1. Art No. CD005652. D01: 10.1002/
14651 858.CD005652.
Brown MZ, Comtois KA, and Linehan MM (2002) Reasons for suicide attempts and
nonsuicidal self-injury in women with borderline personality disorder. Journal of
Abnormal Psychology 111:1 98-202.
Forman EM, Berk MS, Henrigues GR, Brown GK, and Beck AT (2004) History of
multiple suicide attempts as a behavioral marker of severe psychopathology.
American Journal of Psychiatry 161: 437-443.
Gunderson JG and Links P (2008) Borderline Personality Disorder: A Clinical Guide,
2nd edn. Washington, DC: American Psychiatric Press.
Linehan MM (1993) Dialectical Behavioral therapy of Borderline Personality Disorder.
New York: Guilford.
Linehan MM, Armstrong HE, Suarez A, Allmon D, and Heard H (1991) Cognitive
behavioral treatment of chronically parasuicidal borderline patients. Archives of
General Psychiatry 48: 1060-1064.
McMain SF, Links PS, Gnam WH, et al. (2009) A randomized trial of dialectical behavior
therapy. American Journal of Psychiatry 1 66: 1 365-1 374.
National Institute of Clinical Excellence (2009) Borderline Personality Disorder:
Treatment and Management. http://guidance.nice.org.uk/CG78.
Paris J (2008) Clinical trials in personality disorders. Psychiatric Clinics of North
America 31: 517-526.
Paris J (2008) Treatment of Borderline Personality Disorder: A Guide to Evidence-Based
Practice. New York: Guilford.
Skodol AE, Gunderson JG, Pfohl B, Widiger TA, Livesley WJ, and Siever LJ (2002) The
borderline diagnosis I: Psychopathology, comorbidity, and personality structure.
Biological Psychiatry 51 : 936-950.
White CN, Gunderson JG, Zanarini MC, and Hudson Jl (2003) Family studies
of borderline personality disorder: A review. Harvard Review of Psychiatry
11: 8-19.
Zanarini MC, Frankenburg FR, Khera GS, and Bleichmar J (2001) Treatment histories of
borderline inpatients. Comprehensive Psychiatry 42: 144-145.
The Brain
R G Mair, University of New Hampshire, Durham, NH, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Amygdala It is a complex of nuclei in anterior midline areas
of the temporal lobe implicated in the perception and
expression of emotions and in emotional memory.
Association cortex It includes areas that lack direct sensory
or motor connections important for higher aspects of
sensory perception, sensorimotor integration, and cognitive
processes including language, memory, and executive
functions.
Basal ganglia These are a group of nuclei at the base of the
forebrain that are strongly connected to cerebral cortex. They
play a critical role in selection and initiation of motor
responses, procedural learning, and executive functions of
prefrontal cortex.
Hypothalamus It consists of a number of nuclei located
below thalamus and just in front of the midbrain that
regulate behavior state and functions related to homeostasis:
feeding, drinking, thermoregulation, and reproduction.
Motor cortex This includes areas that play a critical role
in voluntary control of movements, particularly
involving the hands. These include primary motor
cortex, which executes voluntary movements, and
premotor and supplementary motor areas, which plan
and program sequential movements.
Sensory cortex It includes areas of cortex that receive
and process information from the senses. These include
primary sensory areas that receive direct input from
sensory nuclei in thalamus (or olfactory bulb for primary
olfactory cortex) and sensory association areas that
process sensory input into coherent sensory
perceptions.
In many ways, the brain is the ultimate product of evolution.
Vertebrate and invertebrate brains share many features that
allow organisms to survive and reproduce in ever-changing
and sometimes unpredictable environments. The human brain
consists of 100- 150 billion neurons, organized into systems that
allow individuals to sense important information and to per-
ceive the external world, to act adaptively, to maintain internal
homeostasis, to think and plan, to leam and remember, and to
communicate with other individuals. The human brain evolved
from common ancestors and thus shares important cellular and
molecular properties with simpler nervous systems found in
invertebrate and vertebrate species. The anatomical organization
of the human brain bears a striking resemblance to the anatomi-
cal organization of the brains of other vertebrates, particularly
mammals.
Anatomical Organization of the Brain
Cells
The brain contains two main types of cells: neurons, electrically
active cells that are specialized to process and transmit infor-
mation, and glial cells, cells that serve a number of functions
that support the activity of neurons. Neurons typically have
four main divisions (Figure 1). Dendrites are appendages that
serve as input zones, conducting electrical activity toward the
cell body. Inputs to neurons are transmitted across synapses
(points of contact with other neurons). Neurons often have
multiple dendrites that tend to branch extensively in the area of
the cell body allowing neurons to summate inputs from many
other neurons. The cell body (or soma) contains the nucleus
and can act as both an input zone and an integrative zone,
summing inputs from multiple dendrites. Axons transmit activ-
ity away from the cell body, often in the form of brief electrical
impulses called action potentials. Neurons tend to have one
axon that can extend up to a meter or more in the peripheral
nervous system. Axons are of constant diameter throughout
their length. Thicker axons conduct signals more rapidly but
are more expensive to maintain metabolically. Axon terminals
are the output zones. In most neurons, these contain synaptic
vesicles that hold molecules of neurotransmitters that are
released from specialized areas of presynaptic membranes in
response to action potentials transmitted along the axon.
Neurons come in many sizes and shapes. The structure of a
neuron can provide important clues about its function. Inter-
neurons (or local circuit neurons) connect between afferent
and efferent neurons in neural pathways and tend to be
involved in local processing of information. Projection neu-
rons have long axons that communicate between areas of the
brain. The organization of dendrites determines what inputs
drive a neuron and how they are integrated. Large neurons with
thick axons tend to conduct high priority information rapidly
(such as visual signals from rapidly moving objects or signals
to fast acting muscles). The speed of axonal conduction is also
increased by myelin sheaths. These are processes of glial cells
(oligodendrocytes in the central nervous system, Schwann cells
in the peripheral nervous system) that wrap around axons.
Myelin increases the speed of conduction by increasing electri-
cal resistance of axonal membranes and causing action poten-
tials to jump between small gaps in the myelin (nodes of
Ranvier).
Gross Anatomy
Nervous systems in simple invertebrates, like the sea anemone,
consist of nets of interconnected neurons. More complex inver-
tebrates like the fly have nervous systems that include periph-
eral masses of neurons (ganglia) that are connected to central
377
378 The Brain
bodies receive inputs from thousands of synapses with other neurons.
Dendrites conduct electrical activity toward the cell body, while axons
conduct activity away from cell bodies. Axon terminals transmit activity
through synapses with postsynaptic neurons. In most mammalian
neurons, this involves the release of neurotransmitter molecules that act
on receptors on the postsynaptic membrane. This figure is reproduced
from Breedlove SM, Watson NV, and Rosenzweig MR (2010) Biological
Psychology. Sunderland, MA: Sinauer with permission from Sinauer
Associates, Inc.
lobes that function as a primitive brain. Many invertebrates
have populations of neurons that mediate basic drives, sleep
and wakefulness, and species-specific reproductive behaviors.
Invertebrates can also demonstrate simple forms of learning
and memory. Vertebrate brains share the distinction between
central and peripheral nervous systems and the evolutionary
trend toward central control of peripheral ganglia. Vertebrates
tend to have larger numbers of neurons, increased reliance on
myelin as a means to speed neural conduction, and central
nervous systems encased in a bony skull and spinal column.
The vertebrate peripheral nervous system consists of cranial nerves
( 1 2 in humans) that connect directly to the brain, spinal nerves
(31 in humans) that connect at segments of the spinal cord,
and the autonomic nervous system, which regulates internal
organs. Spinal nerves consist of a dorsal (toward the back)
root that consists of sensory projections from the body and a
ventral (toward the front) root that consists of motor projec-
tions to the muscles. The autonomic nervous system has two
divisions (sympathetic and parasympathetic) that have oppo-
site effects on the organs they innervate. The sympathetic system
mobilizes the body's resources in response to stress, producing
the 'flight or fight' response. The parasympathetic system pro-
motes vegetative and restorative functions, producing the 'rest
and digest' response.
The vertebrate central nervous system has four main divisions:
spinal cord, hindbrain, midbrain, and forebrain (Figure 2).
The spinal cord extends caudal (toward the tail) from the
medulla connecting the brain to segments of the body through
paired spinal nerves. Spinal nerves provide sensory input to the
brain from the body, innervate skeletal muscles to produce
movement, and control visceral functions through the auto-
nomic nervous system.
The hindbrain is immediately in front of the spinal cord and
consists of the medulla, pons, and cerebellum. The medulla
relays signals between the brain and the spinal cord and con-
tains nuclei that control vital functions related to respiration,
cardiac and vasomotor function, and reflex control of swallow-
ing, vomiting, coughing, and sneezing. The pons (Latin for
'bridge') is the point at which many fiber tracks connecting
the brain to the spinal cord cross (decussate) so that the right
side of the brain controls the left side of the body and the left
side of the brain the right side of the body. The ascending
reticular formation runs through deep areas of the pons
extending caudally into medulla and rostrally into the mid-
brain. This is a poorly differentiated area that plays an impor-
tant role in arousal and the regulation of sleep-waking cycles.
The medulla and pons contain nuclei of cranial nerves that
mediate sensory input from the head, taste, auditory (hearing),
and vestibular (body orientation) sensations, sensory input
from and motor output to bodily organs, and control of chew-
ing, facial expression, salivary and tear glands, and movements
of the neck and tongue. The pons is also the point at which the
cerebellum connects to the brain. The cerebellum is a specialized
enlargement at the back of the brain that regulates movement. It
receives input from sensory and association areas of cortex,
peripheral proprioception (relative position of limbs and other
body parts), and the vestibular system (bodily orientation and
acceleration) and provides outputs to higher and lower motor
control systems. Cerebellum does not directly activate move-
ment, but plays a critical role in the planning and sensory
guidance of movements. Ataxia (gross incoordination of muscle
movements) is the cardinal sign of cerebellar damage.
The midbrain contains the tectum along its ceiling and
tegmentum along its floor. The tectum consists of two paired
structures: the inferior and superior colliculi. The superior
colliculi receive input from visual and other sensory systems
and play a critical role directing behavioral responses in ego-
centric (body centered) space. The inferior colliculi are an
important auditory center that receives inputs from brainstem
auditory nuclei. The tegmentum contains the periacqueductal
gray, a site of action of opiates that plays an important role
regulating the response to pain. It is also the main source of
dopamine in the brain (substantia nigra pars compacta, ventral
tegmental area). Dopamine is a neurotransmitter that plays an
important regulatory role in voluntary movement, motivation,
reward, and executive functions of prefrontal cortex.
The forebrain has two main divisions: the diencephalon,
which forms an inner segment attached to the midbrain, and
the cerebral hemispheres, outer segments extending from the
diencephalon. The diencephalon consists of two main areas:
hypothalamus and thalamus. Hypothalamus is located below
thalamus and just in front of the midbrain. It consists of a
number of nuclei (clusters of neurons) that regulate behavior
state and functions related to homeostasis: feeding, drinking,
thermoregulation, and reproduction. The pituitary is attached
to the base of hypothalamus by a thin stalk (the infundibulum).
The pituitary acts as the master gland releasing hormones that
The Brain 379
Lateral view
Postcentral
gyrus
Precentral
gyrus
(a)
Fornix
Pineal gland
Hypothalamus,
Cingulate.
gyrus
.Superior
colliculus
Corpus-
callosum
Inferior
Pituitary'
Midbrain'
Cerebellum
Medulla
Brainstem'
Basal view
Olfactory
bulb
chiasm
Pituitary
Pons
Mammillary
body
Medulla
Spinal
cord
Figure 2 The organization of the human brain as viewed from (a) lateral, (b) midsagittal (midline), and (c) inferior (beneath) perspectives. The major
lobes and important divisions are color coded and labeled. This figure is reproduced from Breedlove SM, Watson NV, and Rosenzweig MR (2010)
Biological Psychology. Sunderland, MA: Sinauer with permission from Sinauer Associates, Inc.
regulate responses related to stress, metabolism, sexual differ-
entiation, reproduction, growth, fluid balance, and blood pres-
sure. The anterior lobe of the pituitary is under humoral
control, secreting tropic hormones in response to releasing
hormones that originate in hypothalamus and reach the pitui-
tary gland through the portal vein. The posterior lobe is directly
controlled by neurons with cell bodies in hypothalamus and
axons terminating on secretory cells in the pituitary. The thala-
mus is the gateway to cerebral cortex. Thalamus provides the
main source of sensory input to cortex leading to conscious
perception of visual, auditory, somatosensory, and taste sensa-
tions as well as inputs from the basal ganglia and cerebellum
that give rise to intentional responses. Thalamus is also a major
target of descending projections from cortex that reciprocate
thalamocortical projections and provide information to other
areas of cortex. Beyond pathways carrying specific information
to distinct areas of cortex, thalamus contains 'nonspecific' mid-
line and intralaminar nuclei. These nuclei have common
inputs from arousal- and visceral-related systems in the brain-
stem and influence the activity of extended neural networks
that involve widespread areas of cerebral cortex, the basal gang-
lia, and the limbic system.
The cerebral hemispheres consist of an outer rind, cerebral
cortex, and subcortical structures that include amygdala and
the basal ganglia. Cerebral cortex includes olfactory cortex,
hippocampus, and neocortex. The amygdala is a complex of
nuclei in anterior midline areas of the temporal lobe that have
been implicated in the formation of emotional memories and
the perception and production of emotions, particularly fear.
Amygdala is a prominent component of the limbic system, a
series of interconnected structures that also includes hippo-
campus and portions of neocortex that are important for
learning, memory, and emotions. Hippocampus is a primitive
cortical area located behind amygdala in medial temporal
lobe. Hippocampus has been implicated with adjacent areas
of entorhinal and perirhinal cortex as a site of pathology caus-
ing profound amnesia. The basal ganglia constitute a subcorti-
cal system intimately connected with frontal cortex that plays a
critical role in cognitive and motor function. Striatum, the
input side of the basal ganglia, is divided into two main areas
in humans: caudate and putamen. The caudate receives input
from prefrontal cortex and related cortical areas and is primar-
ily concerned with cognitive functions. The putamen receives
inputs from premotor and motor cortex and related cortical
areas and is primarily concerned with the selection and initia-
tion of motor responses. Pallidum, the output side of the basal
ganglia consists of several interconnected structures: globus
pallidus, substantial nigra pars reticulata, and the subthalamic
nucleus. Globus pallidus and substantia nigra project through
thalamus to prefrontal and premotor areas of cortex. The basal
ganglia are organized in a series of parallel loop pathways that
originate in cortex and project through discrete areas of stria-
tum, pallidum, and then thalamus, that project back to the
cortical area of their origin. Each of these parallel cortico-
striato-pallido-thalamo-cortical pathways is thought to medi-
ate a distinct aspect of cognitive or motor function. Diseases of
the basal ganglia, such as Parkinson's disease and Huntington's
chorea can affect the ability to select responses, inhibit
380 The Brain
unwanted responses, initiate voluntary movements, cognitive
aspects of motor function, executive functions mediated by
prefrontal cortex, and the capacity for learning and memory,
particularly the incremental acquisition of habits or stimulus-
response associations.
Neocortex (or isocortex) consists of six distinct layers of
neuropil that form the outer shell of cortex (Latin for 'bark')
of the cerebral hemispheres. The expansion of cortex in
humans caused cortex to fold into a series of sulci (fissures)
and gyri (ridges). Human cortex is divided into four main lobes
(Figure 2). Frontal cortex is in front of the central sulcus. Parietal
cortex is located behind the central sulcus (in front of occipital
and above temporal cortices). Occipital cortex is at the back of
the brain behind the occipito-parietal sulcus. Temporal cortex is
located below the Sylvian fissure (lateral sulcus or fissure), in
front of the occipital lobe and below the parietal and frontal
cortices. There are extensive connections between homologous
areas of cortex in the right and left hemispheres that travel
through two white matter tracts. The corpus callosum is the
main source of interhemispheric cortical communication.
The anterior commissure carries fibers connecting portions of
temporal cortex. Although all areas of neocortex have a com-
parable six-layer organization, there are systematic differences
across different areas of cortex in the extent to which the six
layers of cortex are developed. In motor cortex, for instance,
layer V (which contains large neurons projecting beyond local
areas of cortex) is well developed. By contrast, layer IV (which
receives specific inputs from thalamus) is well-developed pri-
mary sensory cortices. Classic neuroanatomists mapped these
differences in detail long before neuropsychologists described
the localization of function in cortex. The results were maps
based purely on stmctural anatomical differences. One of the
great discoveries of twentieth century neuroscience was the
recognition that maps based on subtle anatomical differences
correspond to the localization of function in cortex. The most
widely used map was developed by the German anatomist
Brodmann in 1909 and still serves as a standard system to
refer to different areas of cerebral cortex.
How do human brains differ ? Humans have unique cogni-
tive capabilities that include the abilities to communicate
with language and to use tools flexibly in complex ways.
Neuroscientists have long been interested in understanding
how the human brain evolved to give rise to these abilities.
It seems unlikely that human cognitive function reflects
a fundamental change in the organization of the brain. The
human brain follows a general blueprint shared with brains
of other vertebrate species. Behavioral neuroscience studies
have revealed similar functions in homologous brain struc-
tures in humans and in other mammalian species. Humans
have relatively large brains for their body mass compared to
other species. The fossil record reveals evidence of a tremen-
dous increase that nearly tripled the size of the human brain
within the last 2 million years. This increase must have con-
ferred substantial evolutionary advantages as it came with a
steep metabolic cost - brain activity consumes about 20%
of our calories. Careful analyses have revealed important
changes in cerebral cortex that may account for our unique
cognitive abilities. The human brain has relatively large
amounts of neocortex and subcortical structures that commu-
nicate with neocortex, including the lateral cerebellar
hemispheres, basal ganglia, and thalamus. Human cortex
has relatively large numbers of cells/ unit area of cortex,
particularly within later developing, more superficial layers
thought to play a critical role in processing information.
Human cortex is more clearly differentiated with more dis-
tinct signs of structural specialization from area to area.
Human cortex has relatively less area dedicated to primary
motor and sensory function and thus more association cortex
available for integrative functions.
Functional Organization of the Brain
Sensation and Perception
Sensory systems consist of peripheral receptor cells specialized
to detect and encode sensory signals, pathways by which this
information is processed and transmitted to primary sensory
cortex, and sensory association cortices that further process this
information to give rise to perception. There are many poten-
tial sensory signals in the environment. Organisms sample and
process a small fraction of the available signals, focusing on
sensory information that fits species-specific needs.
Vertebrate vision begins with photoreceptor cells in the
retina, a thin band of tissue at the back of the eye. Photorecep-
tor cells contain photopigments that absorb visible wave-
lengths of light and release neurotransmitters that affect the
activity of second order visual neurons (bipolar cells). These
interact with interneurons within the retina (horizontal and
amacrine cells) to control the activity of ganglion cells, the
output neurons of the eye. The wavelengths of electromagnetic
radiation detected by vertebrate photoreceptors are in a narrow
range that is ideal for detecting the objects in the external world
including color information useful for finding food and detect-
ing predators. Shorter wavelengths (X-rays) can penetrate
the surface of objects. Longer wavelengths (radio waves) are
better suited for detecting large distant objects (like planets).
The physiological properties of visual neurons are likewise
specialized to encode motions consistent with the speed of
moving animals (rather than say a speeding bullet or a moving
glacier). Axons originating in the eye terminate in the lateral
geniculate nucleus of thalamus as well as in subcortical nuclei
specialized to direct attention to objects of interest (superior
colliculus) and to entrain the brain to daily rhythms of light
and dark (suprachiasmatic nucleus). The projections respon-
sible for visual perception run from the lateral geniculate
nucleus to primary visual (striate) cortex in the occipital lobe
(Figure 2). From here, the visual system diverges to areas of
visual association cortex, extending from the occipital lobe ven-
trally into inferior temporal cortex and dorsally into posterior
parietal cortex, responsible for processing specific aspects of
vision, such as color, motion, and spatial relationships. Rare
cases of brain disease will sometimes affect these discrete areas
of cortex and leave individuals without the ability to perceive
color (achromatopsia), motion (akinetopsia), or spatial rela-
tionships ('Balint's syndrome'). Lesions damaging visual asso-
ciation areas can sometimes interfere with the ability to
recognize objects visually without affecting more elementary
visual processes (visual agnosia).
The auditory system converts vibrations (pressure waves) to
sound. Species differ in the range of both loudness (amplitude)
The Brain 381
and pitch (frequency) of vibrations that they hear. The range of
sounds heard by different species reflect use of sound for
communication, detection of prey and avoidance of predators,
and the media through which auditory signals travel (air,
water, or the ground). In humans, pressure waves are collected
by the outer ear and mechanically amplified during transmis-
sion through the eardrum and the three bones of the middle ear
(the ossicles) to the oval window of the cochlea. The cochlea is a
spiral shaped bony structure. Pressure waves travel from the
oval window through fluid filled cavities of cochlea where they
cause vibrations of the basilar membrane. Vibrations of the
basilar membrane displace ciliary appendages of hair cells
(stereocilia), which then activate terminals of the auditory
nerve. The frequency (or pitch) of a sound is encoded by the
point that the basilar membrane vibrates maximally (place cod-
ing) and by the timing of action potentials produced by the
vibrations (volley coding) . The pattern of activation that emerges
from the cochlea is represented tonotopically as auditory infor-
mation and is transmitted through the cochlear nucleus to the
inferior colliculus, the medial geniculate nucleus in thalamus, and
then primary auditory cortex ( Heschl's gyrus ) in superior tempo-
ral cortex extending into the Sylvian fissure. In tonotopic coding,
different frequencies of sound are represented topologically as a
spatial map. Conscious awareness of sound depends on activa-
tion of primary auditory cortex. Auditory association cortices adja-
cent to primary cortex play critical roles in the ability to perceive
complex patterns of sound as music, language, or elements of a
complex and changing environment.
Somatosensory information (touch) is transduced by differ-
ent types of receptor end organs distributed across the skin.
These include end organs specialized to detect different aspects
of touch (pressure, vibration, stretch, tickle, itch) and temper-
ature (warm, cool). Pain is detected by receptors with higher
thresholds for activation (nociceptors) and by high levels of
activation of somatosensory nerve endings. Somatosensory
information travels through the dorsal columns of the spinal
cord to the medulla where they synapse with neurons that send
axons to the ventral basal complex in thalamus. From here,
somatosensory information travels to primary somatosensory
cortex in the postcentral gyrus and to somatosensory association
areas in adjacent regions of parietal cortex (Figure 2).
Chemosensory information is detected by the olfactory
(smell), gustatory (taste), vomeronasal, and trigeminal systems.
What we commonly experience as flavor of foods or strong
smells are often produced by stimulation of multiple chemosen-
sory systems. Free nerve endings of the trigeminal nerve line the
nasal cavities and respond to irritating concentrations of odor-
ants (which can happen at low concentrations for particularly
noxious chemicals). For the untrained observer, the sting asso-
ciated with trigeminal stimulation is often perceived as an inte-
gral component of a strong smell. Flavor is a complex perception
based on a combination of taste, olfaction (often through retro-
nasal stimulation), and texture (somatosensory input from the
mouth and tongue). There are five basic tastes: salty, sour, sweet,
bitter, and umami. These are detected by receptors on taste buds
in papillae on the tongue and at the back of the throat. Taste
information travels over three different cranial nerves (facial,
glossopharnygeal, and vagus) through axons that terminate in
the nucleus of the solitary tract. From here, taste pathways are
closely aligned with somatosensory areas in thalamus and
cortex. Axons from second order gustatory neurons terminate
in the ventral posterior medial thalamic nucleus, which project in
turn to the cortical taste area in somatosensory cortex.
Smells are detected by receptor molecules located on cilia
projecting from receptor neurons into the mucus layer in the
upper portions of the nasal cavity. Vertebrates have about 1000
different types of olfactory receptors, about 350 of which are
functional in humans. The ability of humans to perceive novel
synthesized molecules that are not found in nature demon-
strates the ability of at least some of these receptors to respond
to a wide range of novel stimulus substances. Axons from
receptor neurons cross the cribriform plate and synapse with
dendrites from mitral cells in the glomeruli of the olfactory bulb.
Each glomerulus receives inputs from receptor cells that
express the same receptor molecule. Mitral cell axons travel
into basal areas of the forebrain where they terminate in primi-
tive cortical areas collectively called primary olfactory cortex. The
ability to discriminate diverse smells results from a diversity of
receptor molecules that are organized to produce distinct topo-
graphic patterns of neural activity across the receptor epithe-
lium. This topographic patterning is preserved within olfactory
bulb and primary olfactory cortex. The vomeronasal organ is
located in the base of the nasal cavity, just above the palate,
in many animals. It is specialized to detect pheremones (chem-
ical signals that trigger social responses). Although the vomer-
onasal organ has been identified in humans, recent evidence
suggest that humans lack functional genes for vomeronasal
receptor molecules and thus that the vomeronasal organ is
vestigial in humans.
Acting on the External World
We act on the world by contracting muscles as Sir Charles
Sherrington noted "whether in whispering a syllable or in fell-
ing a forest." Our ability to act purposively is organized hier-
archically. In 1889, John Hughlings Jackson characterized
three levels of organization that continue to guide modern
thinking about motor control. The lowest level consists of the
spinal cord and brainstem motor nuclei, the middle level of
sensorimotor cortex, and the highest level of premotor and
prefrontal cortex. Jackson emphasized the role of the brain in
organizing movement and the ultimate control of lower motor
centers by higher motor centers.
Spinal mechanisms: Skeletal muscles contract when extrafusal
muscle fibers are excited by axons of a -motoneurons that are
located in the ventral horn of the spinal cord. As muscles can
only exert force in one direction (by contraction), they are
generally organized in groups of agonists (or synergists) pro-
ducing movement in one direction and antagonists that pro-
duce movement in the other direction. In limbs, these pairs
consist of extensors (muscles that resist the force of gravity or
straighten limbs) and flexors (muscles that bend limbs, moving
them in the opposite direction as extensors). The spinal cord
receives inputs from proprioceptors that provide information
about muscle length and tension. Muscle spindles consist of
intrafusal muscle fibers aligned in parallel with extrafusal mus-
cles. Sensory endings in muscle spindles signal information
about the length of muscles and provide the sensory compo-
nent of the stretch reflex. Golgi tendon organs are located at the
insertion of muscle fibers into tendons - in series with skeletal
382 The Brain
muscles. They sense the force exerted on muscles and provide
the sensory component of the autogenic inhibition reflex.
Spinal reflexes provide control over the activation of skeletal
muscles by a-motoneurons. In the stretch reflex, activation of
muscle spindles directly excites a-motoneurons producing
contractions of agonist muscles to compensate for an increase
in muscle length. At the same time, the spindles activate inter-
neurons to reciprocally inhibit antagonist muscles that would
otherwise resist the compensatory movement. The stretch
reflex is what allows a waiter to hold a tray steady while it
grows lighter as food is served or heavier as dishes are placed
on it. The autogenic inhibition reflex occurs when Golgi tendon
organs are activated by muscle force and excite spinal cord
interneurons to inhibit agonist a-motoneurons. This was
once thought to function as protective response, preventing
muscles from being damaged by strong forces. It is now recog-
nized that Golgi tendon organs are activated by much weaker
forces, where the inhibitory reflex is thought to spread work
out evenly across a population of muscle fibers. The spinal
cord also contains circuits that mediate protective reflexes.
The flexion reflex causes a limb to withdraw in response to a
noxious stimulus (like stepping on a tack). Here, flexors are
excited and extensors inhibited for the limb involved, while
extensors are excited for the opposite limb (crossed extension)
to maintain sufficient extensor activity to resist the force of
gravity (and remain upright).
Central pattern generators are neural networks that activate
motoneurons in rhythmic patterns to produce stereotyped
movements. They spare higher motor centers the cost of pro-
gramming individual muscles to produce routine, repetitive
sequences of contractions. Some central pattern generators,
like those that control breathing, are active at rest. Others,
such as those that control locomotion or chewing, are activated
by descending projections from higher motor centers. Central
pattern generators are influenced by sensory feedback that can
modify the duration of phases of the activity cycle to adapt
responding to environmental conditions - for instance adjust-
ing one's stride when stepping on a rock while hiking. Central
pattern generators for locomotion are localized in spinal cord,
while those controlling breathing, chewing, swallowing, and
eye movements are located in brainstem motor nuclei.
Cortical control of movement is mediated by projections
from primary motor cortex in the precentral gyms (Figure 2) to
the spinal cord (corticospinal tract) and to brain stem nuclei
that project to spinal motor centers (such as the rubospinal,
vestibulospinal, and reticulospinal tracts). The corticospinal
tract also contains projections from supplementary and premotor
cortices, located just in front of primary motor cortex, and
somatosensory cortex, in the postcentral gyms. The supple-
mentary and premotor areas project heavily to primary motor
cortex and play important roles in planning movements, par-
ticularly complex sequential movements. While lesions of pri-
mary motor cortex have profound effects on muscular strength
(particularly for the hands), lesions in premotor and supple-
mentary motor areas impair the ability to develop sequential
strategies for movement but have little effect on strength.
Primary motor cortex contains populations of neurons that
respond to the activation of specific muscles and others that
code direction of movement regardless of the muscles involved.
Neurons in primary motor cortex are coarsely tuned, exhibiting
responses coding movements in a large range of directions. The
ability to make precise movements is thought to depend on
the ability to sum responses across a large population of neu-
rons to determine a precise average vector. The motor map is
not rigidly fixed, but can be changed by motor learning. Larger
areas of motor cortex are activated when well-trained move-
ments are performed. This presumably reflects an increased
level of precision for well-practiced motor skills like playing a
musical instrument. Neurons in premotor and supplementary
motor areas exhibit responses that seem consistent with a role
in motor planning. They tend to respond earlier before the
onset of movement, in response to stimuli that indicate the
direction of an impending movement. Studies of brain-injured
patients and results of functional imaging studies suggest that
higher order motor skills such as skilled tool use or pantomime
of tool use depend on large-scale distributed neural networks
that connect the premotor and supplementary motor areas to
association areas in prefrontal and parietal areas of cortex in
the left (dominant) hemisphere of the brain.
The cerebellum and basal ganglia have important effects on
motor function at multiple levels of the central nervous
system. Virtually all areas of cerebral cortex project to striatum,
the input area of the basal ganglia. Striatum projects back
to much of frontal cortex through topographically organized
pathways that pass through pallidum and thalamus. These
cortico-striato-pallido-thalamo-cortical projections provide
parallel pathways that link functionally related areas of cortex,
basal ganglia, and thalamus. Current theories emphasize the
role of the basal ganglia in action selection - deciding which of
several possible actions to execute at a given point in time. Basal
ganglia areas associated with motor cortex are thought to medi-
ate sensory-guided responding, while areas associated with pre-
frontal cortex are thought to guide responding on the basis of
information represented within the brain as working memory.
The cerebellum guides responding on the basis of sensory input
from sensory areas of cortex, peripheral proprioceptors, and the
vestibular system. Cerebellum is divided into regions having
distinct phylogenetic origins. The vestibulocerebellum (floccuo-
modular lobe) targets motor nuclei controlling axial muscles
and primarily affects balance. It is well developed in fish and
reptiles and is regarded as the oldest part of cerebellum in
evolutionary terms. The spinocerebellum consists of the medial
and intermediate zones. It projects to brainstem and midbrain
motor nuclei and is primarily concerned with the modulation
and execution of ongoing movements. The cerebrocerebellum
consists of the lateral zone of the cerebellar hemispheres.
These are particularly well developed in humans and project
to thalamus to provide input to premotor cortex. These are
regarded as the most recently evolved part of cerebellum and
play an important role in the initiation, planning, and timing of
movements.
Adapting to the Environment
Humans, like other mammals, have evolved mechanisms that
allow us to adapt to changing conditions in external and
internal environments. These include systems specialized to
maintain homeostasis of internal conditions, to regulate
behavioral states to maintain beneficial levels of wakefulness
and arousal, to attend to critical tasks and information needed
The Brain 383
to optimize cognitive and perceptual function, and to express
and perceive emotions to facilitate social interactions.
Homeostasis
In the middle of the nineteenth century, the French physiologist
Claude Bernard called attention to the importance of having a
constant internal environment in the body. He emphasized the
organic composition of blood as the nutritional medium of
cells. In the 1920s, the American physiologist Walter Cannon
drew attention to the role of coordinated physiological
mechanisms in regulating internal state conditions. He coined
the term homeostasis to characterize these functions. It is now
well established that the hypothalamus interacts with brainstem
nuclei to control homeostasis through actions of the peripheral
nervous system and hormones released by the pituitary gland.
Thermoregulation maintains body temperature within a narrow
range for mammals and other homeotherms. Body temperature
is sensed by cutaneous thermoreceptors and by temperature
sensitive cells in hypothalamus. Regulator circuits in spinal
cord, brainstem, and hypothalamus sense deviations from the
setpoint (normally 98.6°F for humans) and trigger both com-
pensatory behaviors (heat seeking/ avoiding) and reflex actions
(vasoconstriction/dilation, sweating, respiration, brown fat
metabolism, thyroid hormone secretion) to restore homeosta-
sis. Fluid balance is monitored by cardiac and kidney barore-
ceptors that signal hypovolemia (low blood volume) and
hypothalamic osmoreceptors that signal increased concentra-
tions of solutes in extracellular fluids. These signals are pro-
cessed in hypothalamus to give rise to behavioral adaptations
(motivated drinking, salt appetite) and reflex responses (water
and salt retention by the kidneys, vasoconstriction to maintain
blood pressure) that maintain homeostasis. Energy balance is a
more complex process. Appetite is influenced by short-term
factors signaling hunger, satiation, and the palatability of avail-
able food and by long-term factors related to the amount of
energy stored in adipose tissue. The arcuate nucleus in hypo-
thalamus acts as an appetite controller, responding to hor-
mones released in the gut reflecting both short-term (insulin,
ghrelin, obestatin peptide YY) and long-term (leptin) factors.
Feeding behavior is regulated by nuclei in hypothalamus and
hindbrain based on input from the arcuate nucleus, neural
signals from the gut, and nutrient sensing neurons in the brain.
Behavioral state control
Almost all living species exhibit circadian rhythms, daily pat-
terns of biochemical, physiological, or behavioral processes
that allow organisms to adapt to regular changes in the external
environment. Circadian rhythms result from internal circadian
clocks that typically run slightly longer than 24 h and zeitgeber
('time giver') stimuli (most commonly light) that reset the
clock in phase with the daily cycles of the external environ-
ment. The primary circadian clock in mammals is located in
the suprachiasmatic nuclei. These are small nuclei located just
above the optic chiasm (Figure 2). They contain neurons that
exhibit endogenous circadian rhythms and receive direct input
from specialized cells in the retina that allow light to serve as
a zeitgeber. Circadian rhythms of the suprachiasmatic nuclei
regulate daily patterns of sleep and wakefulness that are char-
acteristic of most mammalian species. Mammalian brains
exhibit different rhythms of neural activity that correspond to
states of vigilance (or alertness). In most general terms, these
consist of the wake state, associated with a low amplitude,
desynchronized EEG activity; slow-wave sleep, divided into
four stages associated with slow-wave EEG activity; and rapid
eye movement (REM) sleep, associated with small-amplitude,
fast EEG waves, lack of muscle tone, and rapid eye movements.
Wakefulness is regulated by neurons in the lateral hypotha-
lamic area and posterior hypothalamus that release the neuro-
peptides orexin A and orexin B (also known as hypocretinl
and 2) in neurochemically specific wake-promoting nuclei that
are located in the brainstem. These wake-promoting nuclei
include noradrenergic neurons in locus coeruleus, serotonergic
neurons in the raphe nuclei, cholinergic neurons in peduncu-
lopontine tegmental and laterodorsal tegmental nuclei, gluta-
matergic neurons in the midbrain, dopaminergic neurons
in tegmentum, and histaminergic neurons in the tuberomam-
millary nucleus. They produce arousal through direct effects
on neurons throughout thalamus and neocortex and through
actions on neuromodulatory systems in the basal forebrain.
Slow-wave sleep results from decreased activity of wake-
promoting systems by substances like adenosine that build
up during prolonged periods of wakefulness and by descend-
ing inhibitory projections that originate in the preoptic area of
hypothalamus. Rapid eye movement sleep is triggered by cholin-
ergic neurons in the pons and maintained by cell groups in the
brainstem that activate cortex and produce muscular atonia.
Attention
William James famously described attention in The Principles of
Psychology (1890): "Everyone knows what attention is. It is the
taking possession by the mind, in clear and vivid form, of one
out of what seem several simultaneously possible objects of
trains of thought. Focalization, concentration, of conscious-
ness are of its essence. It implies withdrawal from some things
in order to deal effectively with others, and it is a condition
which has a real opposite in the confused, dazed, scatter-
brained state ..." Modern theories have unpacked James'
description into a number of components. Top-down attention
(endogenous or voluntary attention) is the ability to deploy
attention according to internal behavioral goals. Bottom-up
attention (exogenous, involuntary attention) is the tendency
for highly salient stimuli like the ring of a cell phone or a
flashing light to attract attention automatically. Stimulus
saliency depends on both the nature of the stimulus and the
environment in which it is experienced. Working memory is
a temporary store of information used to carry out cognitive
functions like decision-making or planning complex beha-
viors. Working memory can hold only a limited amount of
information at a given moment in time. Both top-down and
bottom-up attention determine which of multiple types or
sources of information is held in working memory (and is
thus available for decision-making or planning). At the same
time, working memory plays an important role in directing
top-down control of attention. Convergent evidence supports
a critical role for prefrontal cortex in working memory and
prefrontal and posterior parietal cortices in both top-down
and bottom-up attention. The capacity for attention depends
critically on the level of alertness and arousal and thus many
of the neurochemical systems that control wakefulness (see
above). The relationship between arousal and attention is
384 The Brain
characterized by an inverted-ll (Yerkes-Dodson) function: too
little arousal results in drowsy inattentiveness and too much
arousal produces the frazzled, 'dazed, scatterbrained' state
described by James.
Emotion
In 1872 Charles Darwin published The Expression of the Emo-
tions in Man and Animals in which he characterized emotions as
instinctive reflex actions that serve as an important means of
communication and an efficient way to adapt to changing
environmental demands or opportunities. Modern theories
largely accept the idea that there are fundamental types of
emotion that are expressed through stereotyped motor re-
sponses (primarily facial expressions in humans) and physio-
logical changes. According to cognitive theory, emotions are
activated by cortical mechanisms that perceive and interpret
sensory stimuli within a particular context and separately trig-
ger both emotions and autonomic nervous system responses.
The autonomic responses are thought to influence how the
emotion is experienced and this overall experience of emotion
then provides feedback signals that affect subsequent interpre-
tation of sensory inputs. Neurological theories emphasize the
role of amygdala in processing socially and emotionally rele-
vant stimuli. The amygdala receives sensory information from
thalamus and cortex and contextual information from hippo-
campus required for this function. The central nucleus of the
amygdala sends projections to brainstem areas that trigger vis-
ceral and somatic responses and to hypothalamic nuclei that
control autonomic and hormonal responses. The role of amyg-
dala in perception and expression of emotions has been con-
firmed by convergent evidence from studies of lesion effects on
experimental animals and human patients and analyses of
neural activity in animals and functional imaging in human
subjects. The amygdala is interconnected with two regions of
frontal cortex, the orbitofrontal and anterior cingulate areas,
with which it interacts to produce normal patterns of emotion
and social behavior.
Higher cortical functions in homosapiens
Localization of function in cerebral cortex: Nineteenth century
neurology established beyond doubt that cerebral cortex is
necessary for cognitive function. Controversy remains however
over the extent to which cognitive functions are localized in
discrete areas of cortex. Cortex is broadly divided into motor
areas that give rise to voluntary movements, sensory areas that
receive and process information in the senses, and association
areas that lack direct sensory or motor connections and medi-
ate cognitive functions like perception, attention, language,
and memory. Apart from simple functions of primary motor
or sensory cortex, like flexion of a finger or perception of visual
stimuli in a discrete part of the retina, evidence of localized
cortical function must be treated with caution. Association
areas participate in large-scale distributed neural circuits that
involve other areas of cortex and subcortical areas like thala-
mus, amygdala, and the basal ganglia. It may be misleading to
identify discrete cortical locations with functions like reading,
tool use, or working memory that depend on interactions
between multiple brain areas. A second concern is that cogni-
tive functions, as currently understood, may not accurately
characterize what a particular brain area does. The human
brain is a biological system that evolved gradually over many
millions of years. Cognitive functions are psychological con-
structs that do not necessarily reflect the physiological func-
tions performed by a particular brain area. These concerns
aside, there are strong arguments that support localization of
function. Cerebral cortex is not a homogeneous mass. Human
cerebral cortex is anatomically differentiated by cytoarchitec-
ture (cellular composition), anatomical connections, gene
expression, and neurochemistry. There are many compelling
examples of structure-function relationships that suggest
specialized functions of localized areas of cortex. Similarly,
there are many examples of well-defined cognitive deficits
that result from lesions damaging discrete areas of cortex and
of local areas of cortex exhibiting alterations in neural activity,
metabolic function, and early or intermediate gene expression
when individuals perform specific cognitive tasks. The bottom
line is that while there is abundant evidence that cerebral
cortex is topographically organized into discrete areas that
serve specialized functions, there are also good reasons to be
cautious in identifying specific cognitive functions with dis-
crete areas of cortex.
Language: Much of what we know about the organization of
language in the brain has come from the study of aphasia, the
loss of language ability as a result of brain damage. The mod-
ern study of aphasia began in the early 1860s when Paul Broca
described several patients who lost the ability to speak ('aphe-
mia') with lesions that damaged the third frontal convolution
in the left hemisphere of cerebral cortex (now known as Broca's
area). Broca noted that this inability to speak did not result
from an inability to move muscles needed for speech, to
understand spoken speech, or to communicate nonverbally.
In 1868, Hughlings Jackson argued that the emissive symp-
toms described by Broca reflect a higher-level inability to
express a proposition in words. He based this on observations
that expressive aphasia commonly spares the ability to utter
emotional oaths or overlearned phrases or to produce words in
the context of automatic speech (saying 'no' to indicate some-
thing not to be done, while being unable to say 'no' as a
voluntary answer to a question). He also noted that speaking
and writing are often similarly impaired. In 1874, Carl
Wernicke, a 26-year-old medical student, published a short
monograph based on ten cases of aphasia (only four autopsied)
that described a new type of aphasia and proposed a framework
for understanding aphasia that remains relevant today.
Wernicke described sensory aphasia, a syndrome characterized
by diminished comprehension of speech (but not deafness),
fluent and nonsensical speech, and frequent paraphasias
(errors in speech production). Wernicke attributed sensory
aphasia to lesions damaging the posterior portion of the supe-
rior temporal gyrus in the left hemisphere (now known as
Wernicke's area). Wernicke predicted a third type of aphasia
(conduction aphasia) primarily affecting the ability to repeat
spoken language when lesions disconnect the anterior (Brocas)
and posterior (Wernicke's) language centers. The current view
of aphasia accepts Wernicke's view of functionally distinct ante-
rior and posterior language areas located in the left hemisphere
of most individuals (virtually all right-handed individuals).
Current analyses emphasize the involvement of broader areas
of cortex surrounding the classic Broca's and Wernicke's areas,
particularly for cases in which severe symptoms persist. Current
The Brain 385
descriptions of aphasic syndromes have also changed. Broca's
aphasia is characterized by slow and effortful speech produc-
tion, word-finding difficulty, and grossly intact comprehension
of speech. Aphasiologists emphasize a fundamental deficit in
grammar (agrammaticism) affecting comprehension of gram-
matically complex sentences, the lack of grammatical endings
on words and words carrying grammatical meaning in sponta-
neous speech, and the difficulty of producing more than one
or two word utterances (telegraphic speech). Wernicke's apha-
sia is characterized by a fundamental inability to use language
to express ideas. This is reflected in fluent speech rife with
paraphasic errors, naming errors (often bizarre misnamings),
and impaired comprehension and repetition of speech.
Memory is defined by what we remember, that is the influ-
ence on ongoing thought processes or behaviors of informa-
tion that is no longer present in the environment (and thus
must be represented in the brain). Failure to retrieve is gener-
ally taken as evidence that a memory no longer exists. Memory
is necessary for any cognitive process that is on the basis of
more than information immediately available to the senses.
For the most part, remembering is an automatic, effortless
process. At a cellular level synaptic plasticity, the ability of
synapses between neurons to change in strength, is thought
to underlie learning and memory. Newly learned information
exists in a labile state of altered synaptic strength for a limited
period of time. With the passage of time information becomes
consolidated in a more permanent form. When information is
retrieved, it is reactivated into a labile form for a brief time
before it is reconsolidated back into a more stable form. While
synaptic plasticity seems to be a general feature of many neu-
rons, neurons in memory circuits are organized to represent
information in neural networks that allow for information to
be retrieved and to influence systems that guide responding.
Most information stored in memory is represented in cerebral
cortex, particularly within association areas of cortex. Percep-
tual memories are stored in sensory association cortex while
motor memories such as skilled tool use are stored in premotor
and posterior parietal areas. Working memory, the ability to
hold and manipulate information in consciousness, is encoded
by the activity of neurons in prefrontal cortex. Cortical mem-
ories are accessed by different neural systems that mediate
distinct types of memory. The cerebellum and basal ganglia
mediate procedural memories, including habit learning, skill
learning, and sensorimotor adaptation. The lateral cerebellum
(cerebrocerebellum) receives its main inputs from parietal and
premotor areas of cortex. The basal ganglia receive convergent
corticostriatal projections from widespread areas of cortex.
The amygdala mediates implicit (subconscious) emotional
memories. The amygdala has prominent reciprocal connections
with limbic-related areas of cortex (frontal, cingulate, and insu-
lar areas) and projections that control autonomic and hor-
monal aspects of emotional responses. Hippocampus and
adjacent areas of parahippocampal cortex mediate declarative
(explicit) memories. Parahippocampal cortex has extensive re-
ciprocal connections with association areas of cortex and is
connected in turn with hippocampus. The hippocampal system
is thought to facilitate the formation of associations between
memories represented in different areas of association cortex.
Declarative memories are of information like facts or events
that can be consciously recalled. These include both episodic
memories of specific personal experiences and semantic mem-
ory of facts that are not necessarily tied to a particular context.
Executive functions are core activities of prefrontal cortex that
optimize behavioral performance, particularly in situations
that place demands on cognition. Executive function is critical
for cognitive flexibility, the ability to adapt ongoing behavior
on the basis of the status of external events and internal inten-
tions. Executive functions include working memory, supervi-
sory (top-down) control of attention, and cognitive control.
Working memory is active, short-term memory that has been
related to activation of neurons in dorsolateral prefrontal cor-
tex. Working memory holds information about external sti-
muli, intended actions, and internal goals in consciousness
so that it can be used to guide both overt actions and covert
thinking. Supervisory (top-down) control of attention promotes
control over thoughts and actions by resolving conflicts
between stimulus or response elements. Convergent evidence
suggests that supervisory control of attention depends on inter-
actions between dorsolateral and anterior cingulate areas of
prefrontal cortex. Cognitive control reflects the ability to orches-
trate thought and action in relation to internal goals. In a sense,
it is an emergent property of working memory and supervisory
attentional control. Cognitive control is thought to depend on
lateral prefrontal cortex and to exert its effects through top-
down control of premotor and posterior association areas
of cerebral cortex. Cognitive control appears to be organized
hierarchically in a caudal to rostral fashion with more caudal
areas mediating sensory control and more rostral areas mediat-
ing higher-level aspects of cognitive control.
See also: Amnesia and the Brain; Appetite; Attention; Glial Cells;
Hippocampal Formation; Memory, Neural Substrates; Neuroexecutive
Function; Sense of Smell; Sense of Taste (Effect on Behavior); Visual
Perception.
Further Reading
Blumenfeld H (2010) Neuroanatomy Through Clinical Cases. Sunderland, MA: Sinauer.
Breedlove SM, Watson NV, and Rosenzweig MR (2010) Biological Psychology.
Sunderland, MA: Sinauer.
Squire LR, Berg D, Bloom F, du Lac S, and Ghosh A (2008) Fundamental Neuroscience,
3rd edn. San Diego, CA: Academic Press.
Relevant Websites
http://sig.biostr.washington.edu/projects/da/ - The Digital Anatomist Project,
University of Washington, Seattle.
http://www.med.harvard.edu/AANLIB/home.html - The Whole Brain Atlas, Harvard
Medical School.
Brain and Behavior Relationships
B T Legesse, B H Price, and E D Murray, McLean Hospital, Belmont, MA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Achromatopsia Also known as cortical colorblindness, it
refers to the loss of ability to perceive colors.
Agnosia It is loss or diminution of the ability to recognize
familiar objects or stimuli. Agnosia may refer to visual,
auditory, or tactile perceptions.
Alexia It is the loss of ability to read.
Amusia It is the inability to produce or comprehend music.
Aphasia It is the loss or impairment of the power to use or
comprehend words.
Aprosody It is the absence of the normal pitch, rhythm, and
variations in melodic intonations of speech.
Declarative memory A type of memory that requires
conscious recollection of information previously
learned.
Dysarthria It is the difficulty in articulating words.
Hierarchical (serial) processing A type of processing where
information is relayed in hierarchy from one center to
the next.
Neuronal/synaptic plasticity The property of synapse and
neurons to change their functionality, connectivity, and
internal parameters based on activity level.
Nondeclarative memory A type of memory that does not
require conscious or intentional recollection of what
was learned.
Nuclei A collection of central nervous system gray matter
(neuronal cell bodies) embedded in white matter.
Parallel-processing model Processing of different aspects
of information simultaneously in different regions of the
brain.
Prosopagnosia It is the loss or inability to recognize
human faces.
Historical Background
From the time of Aristotle to the modern age, human behavior
has been understood in the context of the ideological trends and
philosophy of the historical period. During the last millennium,
the natural sciences have developed testable theories, which
explain observed phenomena in terms of physical laws. The
development of this scientific method of inquiry has resulted
in a shift from exclusively theological or philosophical explana-
tions to those that are subject to measurement and testing.
The Greek physician, Hippocrates (460-377 BC), main-
tained that "not only our pleasure, our joy and our laughter
but also our sorrow, pain, grief and tears rise from the brain and
brain alone" (Figure 1). Hippocrates' ideas were overshadowed
by those of the influential Greek philosopher, Aristotle (384-322
BC), who proposed that the heart was the organ of thought and
mental processes while relegating the brain to the role of a
cooling organ for hot blood from the heart. The second century
AD physician, Galen of Pergamum, advanced the theory that
the brain was the seat of mental activity, and like the rest of the
body, it functioned by the balance of the four vital humors.
The bodily humors hypothesis proposed that differences in
human moods come as a consequence of imbalances in one
of the four bodily fluids: blood, yellow bile, black bile, and
phlegm. According to Galen, an imbalance of each humor
corresponded with a particular human temperament (blood-
sanguine, black bile-melancholic, yellow bile-choleric, and
phlegm-phlegmatic). Individuals with sanguine temperaments
are extroverted and social. Choleric people have energy, passion,
and charisma. Melancholics are creative, kind, and considerate.
Phlegmatic temperaments are characterized by dependability,
kindness, and affection.
For thenextl500 years until the time of the Enlightenment,
the understanding of brain function was highly influenced by
the teachings of Galen. As Western Europe gradually emerged
from the Dark Ages, the dissection of human cadavers became
more common. From dissections, artists like Leonardo da Vinci
prepared anatomic atlases of the human body and nervous
system. The seventeenth century French philosopher and
mathematician, Rene Descartes, argued that physical laws
could not account for the full range of human behavior such
as the intellect. From this hypothesis arose the modern formu-
lation of the mind - body dichotomy, known as Cartesian
Dualism, which to this day continues to be a point of philo-
sophical debate (Figure 2).
The proponents of the mind-brain dichotomy have con-
tended that human consciousness is unlike any other physical
phenomena, and cannot be fully accounted for by mere
biological processes. This mind-brain division led to the diver-
gence of psychiatry and neurology in twentieth century medi-
cine, with psychiatry focusing on the psychological and
emotional underpinnings of behavior while neurology focused
on brain physiology and pathology. Sigmund Freud (1856-
1939), the founder of the psychoanalytic school of psychiatry
and a neurologist by training, focused mainly on the psycholog-
ical foundations of behavior (Figure 3). Freud did, however,
believe that brain physiological studies would eventually lead to
a better understanding of mental processes. Contemporary neu-
roscience views the brain as the source of all behavior and
mental functions. Human cognitive abilities and behaviors are
seen as having varying degrees of representation and homology
with those found in other members of the animal kingdom.
By the nineteenth century, the gross anatomy of the human
nervous system had been well documented. Through careful
microscopic examination of brain sections, Santiago Ramon y
Cajal (1852-1934) prepared drawings of brain cells, which con-
tinue to be referenced today. Korbinian Brodmann (1868-1918)
defined regions of the cerebral cortex based on histological
386
Brain and Behavior Relationships 387
Figure 1 Hippocrates. Engraving by Peter Paul Rubens, 1638. Courtesy
of the National Library of Medicine.
Figure 2 Portrait of Rene Descartes by Frans Hals, Louvre.
characteristics. These regions, named Brodmann areas, corre-
spond to functional regions (see Figure 4). The work of Paul
Broca (1824-1880) and Carl Wernicke (1848-1905) contrib-
uted to our understanding that the brain has regions dedicated
to the performance of discrete functions. The origin of such
intricate specialization was a mystery but it would soon be less
so. In 1859, in his book On the Origin of Species, Darwin
proposed that natural selection was the means by which nature
crafted biological specialization. This theory proposed that the
Figure 3 Sigmund Freud. Photograph by Max Halberstadt, 1921.
characteristics of organisms result from a long process of natu-
ral selection. Organisms, and therefore the nervous system,
were shaped by adaptation to ever-changing environmental
pressures. Since the introduction of this theory, scientists
have analyzed physical and behavioral characteristics in terms
of their adaptive advantages to the organism (Figure 5).
The substance that encoded biological evolutionary
changes and allowed them to be transmitted from generation
to generation was still unknown. In the mid-nineteenth cen-
tury, a Catholic monk, Gregor Mendel, while experimenting
with horticulture and insects, observed that organisms inherit
traits via discrete units of inheritance. Mendel's observations
served as a foundation for the modern science of genetics. It
was not until 1953 that James Watson and Francis Crick, using
X-ray diffraction data, deduced the double helix structure of the
DNA molecule. In 1963, they were awarded the Nobel Prize for
their discoveries which ultimately revealed how information
was transferred from generation to generation in living organ-
isms. Subsequent research has identified a genetic basis for
numerous neurobehavioral disorders such as Huntington's
disease, frontotemporal dementia, Down's syndrome, and
Wilson's disease. The genetic factors that alter susceptibility to
many diseases, such as Alzheimer's disease, schizophrenia,
mood disorders, alcohol dependence, and certain personality
disorders, have also been identified.
The germ theory of disease and the development of micro-
biology helped elucidate brain and behavior relationships as
well. Although many individuals have made monumental con-
tributions to the development of microbiology, it was Robert
Koch who published a series of proofs in 1890 verifying the
germ theory of disease. Microbiology research led to identifica-
tion of the Treponema pallidum bacterium in 1905 by Shaudinn
and Hofmann as the causative agent for syphilis. Eight years
later, Hideyo Noguchi demonstrated the presence of the syph-
ilis spirochete in the brain of a patient with generalized paresis
of the insane (GPI). Before this, GPI was largely thought
to be a mental illness caused by weakness of personality or
defective moral character. After this discovery, it became the
first neuropsychiatric syndrome caused by an identifiable
infectious agent. A study published in 1930 looking at more
388 Brain and Behavior Relationships
Selected Brodmann regions of the cortex and their related
functions
Brodmann's #
Name
Function
17
Occipital lobe
Visual projection cortex
18
Occipital lobe
Visual association cortex
19
Posterior parietal lobe
Visual association cortex
37
Tempero-parietal-
occipital area
General sensory
association cortex
39
Angular gyrus
Word recognition
40
Supramarginal lobe
Somatosensory
association cortex
1,2,3
Postcentral gyrus
Somatosensory projection
cortex
5, 7
Superior parietal lobule
General sensory
association cortex
22
Superior temporal gyrus Auditory association cortex
21, 20, 38
Inferior temporal
cortex
General sensory
association cortex
4
Precentral gyrus
Primary motor cortex
1,2,3
Postcentral gyrus
Somatosensory projection
cortex
6,8,9
Premotor cortex
Motor association cortex
41, 42
Middle 1/3 of superior
temporal cortex
Auditory projection cortex
44,45,46
Broca's area
Motor association cortex -
specific to speech
10
Prefrontal cortex
General motor association
cortex
11
Orbital gyri
General motor association
cortex
Figure 4 Selected Brodmann regions of the cortex and their related functions. Reproduced with modification from Mesulam MM (ed.) (2000)
Principles of Cognitive and Behavioral Neurology. New York: Oxford University Press.
Figure 5 Charles Darwin 1868.
than 2000 consecutive admissions to an insane asylum esti-
mated that one in six patients may have had syphilis infection.
The 1930s and 1940s saw the beginning of the antibiotic
revolution in medicine. The introduction of penicillin, an
effective treatment for syphilis, resulted in a dramatic reduction
of syphilis-related psychiatric illness.
Principles of Behavioral Neuroanatomy
From simple reflexes to complex reasoning, behavior is gener-
ated in a hierarchal fashion in the nervous system. Many aspects
of behavior are mediated almost reflexively, with more complex
mental tasks requiring engagement and coordination of multi-
ple higher brain centers. Simple reflexes, such as tendon stretch
reflexes, are mediated at the level of the spinal cord, while brain
stem structures control semiautomated motor aspects of beha-
viors such as crying, laughing, chewing, and sucking without the
absolute requirement of cortical involvement. Anencephalic
babies, born without the cerebral cortex, are able to suckle,
chew, and cry just like their healthy counterparts. Nevertheless,
there is a bidirectional modulation between subcortical and
cortical centers, with the cortex exerting regulatory control over
lower-level reflexes under normal circumstances. The hypothal-
amus is a subcortical structure which mediates homeostatic
functions such as temperature, appetites, and sexual drive. The
cortex regulates behaviors which address the biological needs
arising from lower centers.
Cerebral function may be analyzed from the vantage point
of a number of disciplines: neurochemistry, neurophysiology,
Brain and Behavior Relationships 389
histology, and at the gross anatomical level. None of these
alone is sufficient for achieving a comprehensive understand-
ing of behavioral neurobiology. Neurotransmitters, receptors,
synaptic plasticity, and neural connectivity form a continuum
of elements which are the basis of nervous system function. On
a higher organizational level, corticocortical and corticosubcor-
tical brain networks link and coordinate distant brain regions.
These networks function in parallel to generate our mental
representations of the internal milieu and the external world.
Chemical and Neural Basis of Behavior
Neurotransmission
For millennia, various cultures have used perception and
mood-altering substances in religious or shamanic rituals to
facilitate healing, divination, communication with spirits, and
to mark coming-of-age ceremonies. Some plant derivatives,
such as opiates and cocaine, have been used in modern
times for their medicinal benefits and abilities to induce
euphoria. The misuse of drugs has resulted in medical, legal,
and social problems for which society has not been able to
find adequate solutions. Yet, they also have opened up new
and profound opportunities to study human behavior. The
mechanism of the profound behavioral effects of psychoactive
ethnobotanical substances did not receive wide scientific
attention until the twentieth century when, in 1938, Albert
Hofmann synthesized a hallucinogenic drug, lysergic acid
diethyl amide (LSD), from the ergot fungus. In 1943, Hof-
mann became inadvertently intoxicated by LSD and is remem-
bered for his colorful description of his bicycle ride home
while in an intoxicated condition. Early researchers observed
that the drug's effects were similar to the naturally occurring
symptoms of schizophrenia. In the decades that followed,
psychiatrists developed an interest in the properties of halluci-
nogenic drugs as a model for understanding psychosis. The
structural similarity of serotonin (5-HT) to LSD offered sup-
port to researchers who hypothesized that alteration of brain
5-HT might be involved in mental illness.
The early 1950s witnessed the birth of psychopharmacology
with the serendipitous discovery that the administration of
chlorpromazine reduced psychotic symptoms in patients. The
introduction of first-generation antipsychotics, over three dec-
ades, significantly reduced the number of chronically psychotic
patients institutionalized in long-term treatment facilities and
returned them to the community. The antidepressant effects of
antihistamine derivatives with a tricyclic structure led to a
subsequent revolution in the pharmacotherapy of depressive
illness. Julius Axelrod and his colleagues received the Nobel
Prize for their work performed in the late 1950s describing
the synthesis, storage, release, and inactivation of catechol-
amine (i.e., epinephrine, norepinephrine, and dopamine) neu-
rotransmitters. Many classes of medications have since been
developed for treatment of psychosis, depression, mania, anxiety,
obsessive-compulsive disorder, and insomnia: selective seroto-
nin reuptake inhibitors, serotonin-norepinephrine reuptake
inhibitors, second-generation (atypical) antipsychotics, tetracyc-
lic antidepressants, and monoamine oxidase inhibitors. This rev-
olution in pharmacology opened new avenues for investigating
the chemical mediation of behavior. Research clarified that
antipsychotic medications exert benefits through their ability to
block dopamine D2 receptor binding. Amphetamines were
demonstrated to increase brain norepinephrine and dopamine
levels, and large doses or prolonged use of this medication were
observed to cause psychosis. These lines of evidence converged to
support the 'dopamine hypothesis' of schizophrenia. This
hypothesis postulated that psychosis results from disturbed and
hyperactive dopaminergic signal transduction. Another model
based partly on the mind-altering effects of drugs, namely, phen-
cyclidine (PCP) and ketamine, which act upon glutamatergic
signaling systems, is that of the 'glutamate hypothesis.'
Learning and Memory
Connectivity between neurons is a dynamic process, continu-
ally being modified by the level of neuronal activity. Synapses
with high activity levels are maintained and strengthened,
while low activity leads to weakening of synaptic transmission
or even dissolution of the connection. This process continues
throughout life and is a critical substrate for learning, memory,
and the brain's ability to adapt to environmental demands. In
2000, Eric Kandel was awarded the Nobel Prize in medicine for
his work describing the physiological processes by which
memory and learning occur in aplysia (sea slug). This work
ultimately shed light on how individual personality styles
might be shaped by experience, why children start to speak
the language they are most exposed to, and why practice
improves subsequent performance.
Localization of Brain Functions
The cortex has discrete regions dedicated to the processing of
language, visual, auditory, and olfactory information as well as
areas for cognitive and emotional processing. There are higher-
order cortical association areas that are specialized to process
color, form, fine details of visual objects, motion, and even
areas specific to human face recognition. The brain has seman-
tic stores, located in association cortices, where knowledge
about the world is categorized and organized. These stores
attach semantic knowledge to primary modalities of percep-
tion and action located in primary sensory and motor cortices.
Hence, the brain is a collection of highly subspecialized,
interconnected functional regions (see Figure 6). The proces-
sing of different characteristics of a stimulus in discrete brain
regions may seem counterintuitive to our subjective experi-
ence. After all, when we see a red car speeding down the street,
we do not perceive the color, shape, noise, or speed of the car
separately. Rather, we see and hear the car as an integrated
whole. How the brain binds these various percepts into one
experience is not fully understood. It is clear that it is depen-
dent on the rich reciprocal connections among distributed
specialized cortical and subcortical regions which create a
type of parallel processing network.
Early functional maps of the cerebral cortex were prepared
from postmortem studies of patients with brain insults. Later,
the famous neurosurgeon, Wilder Penfield (1891-1976),
refined a technique of testing the function of various brain
regions. While experimenting with ways to limit brain injury
during surgery, he applied electrical stimulation to specific
brain regions via small probes while the patients were
390 Brain and Behavior Relationships
Figure 6 Positron emission tomography (PET) imaging of the brain
demonstrating that different brain regions are activated when people
perform four different language tasks, showing anatomical localization of
the various language tasks. Courtesy of Marcus Raichle.
conscious on the operating table under only local anesthesia.
Penfield observed their responses and accurately mapped the
major sensory, motor, and language areas, thereby allowing
some ability to minimize injury to these areas during surgery.
He also charted the connections from sensory and motor cor-
tices (homunculus) to the various limbs and organs of the
body. Strikingly, he observed that in his epilepsy patients,
stimulation applied to regions of the temporal lobes could
trigger vivid recall of past experiences. As of yet this has not
been reproduced in patients without epilepsy. The functional
maps Penfield produced are still in use today. Functional mag-
netic resonance imaging (fMRI) and positron emission tomog-
raphy (PET) scans are imaging technologies which have led to
refinements of functional brain maps (see Figure 6). fMRI uses
signal changes which arise from blood flow and/or oxygena-
tion changes caused by local neural activity. These signal
changes serve to indicate regional brain activity during the
performance of a defined task. PET scans use radio-labeled
tracer molecules which may target receptors, molecules, or be
incorporated into metabolic processes to allow visualization.
Cognitive-Behavioral Neuroanatomy
Two general principles of brain organization can be discerned
(see Figure 4). Perceptual areas such as the somatosensory, visual,
and auditory cortices are located posterior to the central sulcus,
while effector (output) areas such as the prefrontal and motor
cortices are found anterior to the central sulcus. Speech produc-
tion (Broca's area) and speech perceptual areas (Wernicke's area)
also follow this anterior-posterior pattern of organization. In
addition to anterior-posterior localization, the brain also shows
lateralization of function between the hemispheres. The left and
right hemispheres of the brain look similar but they are not exact
mirror images of each other in terms of anatomy or function.
Each hemisphere receives sensory information from and controls
motor activity to the opposite side of the body.
The left hemisphere is dominant for language function in
over 90% of all right-handed and about 60% of left-handed
people, while the right hemisphere is involved in the emotional
content (prosody) of language. Hence, damage to the left Broca's
area results in hesitant, agrammatical speech with short utter-
ances consisting of few words. A lesion in the homologous right
hemisphere will result in impairment of emotional inflection in
speech. In such patients, speech may be flat, monotonous, and
lacking in gestures, but with fairly normal content and grammar.
The parietal, frontal, and cingulate cortices interact to generate
attention to the opposite side of space. Lesions in these cortices
produce contralateral sensory neglect, directional hypokinesia,
and reduced motivational value, respectively. Neglect most often
occurs with respect to the left side of space, indicating the impor-
tance of the nondominant hemisphere (usually, the right hemi-
sphere) in spatial attention. Neglect syndromes show that focal
lesions may affect specific aspects of self-awareness while sparing
others. Focal lesions resulting in neglect syndromes illustrate that
self-awareness is a composite consisting of the interconnected
functions of many brain regions.
Cortical Organization
The cerebral cortex can be divided into five major functional
subtypes: primary sensorimotor, unimodal association, hetero-
modal association, paralimbic, and limbic (see Figure 7). The
primary sensory areas are the point of entry for sensory infor-
mation into the cortical circuitry. The primary motor cortex
conveys motor programs to motor neurons in the brain stem
and spinal cord. Further processing of sensory information
occurs as information moves from primary sensory areas to
the adjacent unimodal association areas. The unimodal and
heteromodal cortexes are involved in perceptual processing
and motor planning. The complexity of processing increases
as information is then transmitted to heteromodal association
areas, which receive input from more than one sensory modal-
ity. Examples of heteromodal association areas include pre-
frontal cortex, posterior parietal cortex, parts of the lateral
temporal cortex, and portions of the parahippocampal cortex.
The highest level of cognitive processing occurs in regions
referred to as transmodal areas. These areas are composed of
heteromodal, paralimbic, and limbic regions, which are collec-
tively linked in parallel to other transmodal regions. Intercon-
nections among transmodal areas such as Wernicke's area,
posterior parietal cortex, and the hippocampal-entorhinal cor-
tex allow the integration of distributed perceptual processing
systems resulting in perceptual recognition, such as scenes and
events becoming experiences or words taking on meaning.
The limbic region has strong connectivity with the prefron-
tal cortex and is intimately involved with regulation of emo-
tion, memory, motivation, and autonomic and endocrine
function. Limbic system dysfunction has been described in
several psychiatric syndromes such as schizophrenia, bipolar
illness, and depressive disorders. The amygdala, located in the
mesial temporal lobe, is important for both the acquisition and
expression of emotions. Its strong connectivity with the hypo-
thalamus is the underpinning for the autonomic and visceral
manifestations and sensations accompanying emotions such as
fear. The critical role of the amygdale for emotional processing
is demonstrated in individuals who have Kliiver-Bucy syn-
drome. Kliiver-Bucy syndrome is a behavioral syndrome char-
acterized by psychic blindness (visual agnosia), a tendency to
Brain and Behavior Relationships 391
Central
sulcus
Supplementary ^
* motor cortex
motor area (SMA)
Broca’s
area
Premotor cortex
Prefrontal
association
cortex
Limbic cortex
Primary visual
cortex
Primary auditory cortex
(in sylvian fissure)
Auditory
association cortex
Visual association cortex
Primary
somatosensory cortex
Somatosensory
association cortex
Secondary somato-
sensory cortex
(in parietal operculum)
Wernicke’s area
□ Primary motor or sensory cortex □ Heteromodal association cortex
I I Unimodal association cortex □ Limbic cortex
Figure 7 Showing major functional areas of the brain. Reproduced with permission from Blumenfeld H (201 0) Neuroanatomy Through Clinical Cases,
2nd edn. Sunderland, MA: Sinauer Associates.
examine objects orally, distractability, hypersexuality, loss of
aggressiveness (placidity), and a striking inability to display or
experience fear or anxiety. Kliiver-Bucy syndrome occurs with
injury to the amygdale or medial temporal lesions on both
sides. Potential causes include traumatic brain injury, anoxia,
herpes encephalitis, and frontotemporal dementia. In contrast,
seizures arising from the amygdala often produce intense anxi-
ety or fear, through overactivation of this circuit.
Cortical Networks
Five distinct cortical network regions govern specific aspects
of cognitive functioning: (1) the language network, which
includes transmodal regions in Broca's and Wernicke's areas;
(2) the spatial awareness network dependent on transmodal
regions in the posterior parietal area, frontal eye fields; (3) the
memory and emotional network located in the hippocampal
entorhinal region and amygdala; (4) the executive function-
working-memory network dependent on transmodal regions
in the lateral prefrontal cortex and possibly in the inferior parie-
tal cortices; and (5) the face-object recognition network based
in the temporopolar and midtemporal cortices. Lesions of
transmodal cortical areas result in global impairments such as
hemineglect, anosognosia, amnesia, and multimodal anomia.
Disconnection of transmodal regions from a specific unimodal
input will result in selective impairments such as category-
specific anomias, prosopagnosia, pure word deafness, or pure
word blindness (alexia).
Cortical networks underlie the ability to empathize with
another person's psychological and physical circumstances.
Empathy is a necessary foundation for social and moral behav-
ior. Originally identified in monkey premotor cortex, mirror
neurons are a specific class of visuomotor neuron that dis-
charges in a subject when a particular action is performed and
when an action is observed in another individual. The human
mirror neuron system is the neural network now postulated
to be involved in comprehending the actions of others and
the intentions behind the actions. It may also provide the
basis for observational learning. The parietofrontal mirror sys-
tem, which includes the parietal lobe and the premotor cortex
plus the caudal part of the inferior frontal gyrus, is involved in
recognition of voluntary behavior in other people. The limbic
mirror system, formed by the insula and the anterior mesial
frontal cortex, is devoted to the recognition of affective behav-
ior. Dysfunction of this system may underlie deficits in theory
of mind and has been proposed as an explanation for the
social deficits seen in autistic disorders. Theory of mind is a
model of how a person understands and infers other people's
intentions, desires, mental states, and emotions.
Frontosubcortical Networks
Cognition, behavior, and movement are dependent on five
frontosubcortical circuits. All share the same principal compo-
nents: (1) frontal cortex; (2) striatum (caudate, putamen, ven-
tral striatum); (3) globus pallidus and substantia nigra; and
(4) thalamus. While two of these circuits are concerned with
somatic motor and oculomotor functions, the other three are
intimately involved in cognition and behavior. These circuits
originate in the dorsolateral prefrontal, orbitofrontal, and
anterior cingulate cortices.
The dorsolateral prefrontal circuit governs executive func-
tions, including the ability to plan and maintain attention,
problem-solve, learn, retrieve remote memories, sequence the
temporal order of events, shift cognitive and behavioral sets,
and generate motor programs. Executive dysfunction is a prin-
cipal component of subcortical dementias. Common deficits
in subcortical dementias include slowed information proces-
sing, memory retrieval deficits, mood and behavioral changes,
gait disturbance, dysarthria, and other motor impairments.
Vascular dementias, Parkinson's disease, and ffuntington's dis-
ease are examples of conditions that affect this circuit.
392 Brain and Behavior Relationships
The orbitofrontal circuit connects frontal monitoring func-
tions to the limbic system. This circuit governs appropriate
responses to social cues, empathy, social judgment, and inter-
personal sensitivity. It pairs thoughts, memories, and experi-
ences with corresponding visceral and emotional states. This
circuit is heavily involved in the process of decision-making
and evaluating the costs and benefits of specific behavioral
responses to the environment. The medial orbitofrontal cortex
(OFC) evaluates reward, whereas the lateral OFC monitors
and decodes punishment as it pertains to motivating behav-
ioral change. There is an anterior-posterior gradient in which
the reward value for more abstract and complex secondary
reinforcing factors, such as money, is encoded in the anterior
regions, and more concrete factors such as touch and taste are
encoded in the posterior OFC areas. The posterior OFC is
thought to have an important role in evaluating the emotional
significance of stimuli. Dysfunction in this circuit can lead to
disinhibition, irritability, aggressive outbursts, inappropriate
social responses, and impulsive decision-making. Persons
with OFC lesions may show deficits in both the production
and recognition of emotional expression from the face, voice,
or gestures. Bilateral OFC lesions may result in theory of mind
deficits.
The anterior cingulate circuit includes the nucleus accum-
bens and has both afferent and efferent connections to the
dorsolateral prefrontal cortex (DLPFC) and amygdala. It is
involved in motivated behavior. Lesions in this circuit result
in apathy, loss of the ability to act or to make decisions (abu-
lia), and akinetic mutism (loss or impairment of voluntary
activity and communication). Apathy can have affective (flat
and unchanging facial expressions), emotional (loss of interest
or excitement and lack of emotional reactivity), cognitive (lack
of interest in learning and a decreased ability to understand
new thoughts and participate in creative thought processes),
and motoric dimensions (lack of effort, reduced productivity,
and decreased ability to sustain activities). The medial prefron-
tal cortex is thought to play a significant role in the generation
of emotions related to empathy, cognitive functions related
to 'theory of mind,' and the ability to recognize a moral
dilemma. The ventromedial frontal lobe evaluates the current
relative value of stimuli helping to guide decision-making
by determining the goals toward which behavior is directed
and through judging outcomes. Various conditions that may
affect this circuit include Alzheimer's disease, frontotemporal
dementia, Parkinson's disease, and Huntington's disease, head
trauma, brain tumors, cerebral infarcts, and obstructive
hydrocephalus.
Cerebrocerebellar Networks
Once thought to mainly coordinate movement, the cerebellum
has been shown to have a role in the regulation of cognition,
mood, and behavior. Cognitive processing tasks such as lan-
guage, working memory, and spatial and executive tasks are
associated with activation of the posterior cerebellar lobes. The
posterior cerebellar vermis may function as a putative limbic
cerebellum modulating emotional processing. Distractibility,
executive and working memory problems, impaired judgment,
reduced verbal fluency, disinhibition, irritability, anxiety,
emotional lability or blunting, and obsessive-compulsive
behaviors as well as depression and psychosis have been
reported in association with cerebellar pathology.
Neuropsychiatry
Neuropsychiatric symptoms can result from dysfunction in
specific anatomic regions. Any disease, toxin, drug, or process
that affects a region can manifest with changes in behavior
mediated by the circuitry dependent upon that region. Brain
pathologies may result in loss of function (negative symptoms,
e.g., apathy), appearance of symptoms which should not be
there (positive symptoms, e.g., hallucinations), or a combina-
tion of the two. Table 1 summarizes neuropsychiatric symp-
toms and their anatomic correlates.
Not all brain lesions, however, will have observable behav-
ioral consequences. There is also wide individual variability in
the behavioral manifestations of focal brain lesions. Observa-
tional studies have shown that depression is common follow-
ing left anterior or left subcortical stroke, but not all patients
with such lesions develop depression. When temporal lobe
epilepsy or Huntington's disease begins in adolescence, a
higher incidence of psychosis is noted than when their onset
occurs later in life. This variability in behavioral manifestation
of brain lesions may be due to factors specific to the lesion
(e.g., size, location, laterality) and factors related to the indi-
vidual such as the cerebral reserve, genes, and developmental
stage of the brain (Table 2).
The case of Phineas Gage illustrates how regional brain injury
may result in profound behavioral changes in the absence of
Table 1 Neuropsychiatric symptoms and proposed corresponding
neuroanatomy
Symptom
Neuroanatomic region
Depression
Frontal lobes, left anterior frontal cortex,
anterior cingulate gyrus, subgenu of the
corpus callosum, basal gangia, left caudate
Mania
Inferomedial and ventromedial frontal cortex,
right inferomedial frontal cortex, anterior
cingulate, caudate nucleus, thalamus, and
temporothalamic projections
Apathy
Anterior cingulate gyrus, nucleus accumbens,
globus pallidus, thalamus
Obsessive-
Orbital or medial frontal cortex, caudate
compulsive
disorder
nucleus, globus pallidus
Disinhibition
Orbitofrontal cortex, hypothalamus, septum
Paraphilias
Mediotemporal cortex, hypothalamus, septum,
rostral brain stem
Hallucinations
Unimodal association cortex, orbitofrontal
cortex, paralimbic cortex, limbic cortex,
striatum, thalamus, midbrain
Delusions
Orbitofrontal cortex, amygdala, striatum,
thalamus
Psychosis
Frontal lobes, left temporal cortex
Reproduced with permission from Murray ED and Price BH (2008) Depression and
psychosis in neurological practice. In: Bradley WG, Daroff RB, Fenichel GM, and
Jankovic J (eds.) Neurology in Clinical Practice, 5th edn., pp. 101-120. Philadelphia,
PA: Butterworth-Heinmann.
Brain and Behavior Relationships 393
Table 2 Selected neurobehavioral manifestations of focal brain lesions
Hemispheric side of the lesion
Left
Right
Bilateral
Temporal lobes
Mesial
Anterograde amnesia for verbal
Anterograde amnesia for nonverbal
Severe anterograde amnesia for verbal
material
material
and nonverbal material
Occipitotemporal
Prosopagnosia
Transient or mild prosopagnosia
Prosopagnosia, visual agnosia
junction
Occipital lobes
Ventral
Right hemiachromatopsia, alexia
Left hemiachromatopsia, visual agnosia
Full-field achromatopsia, visual agnosia
Parietal lobes
Temporoparietal
Wernickie’s aphasia
Amusia
Auditory agnosia
junction
Inferior parietal
Tactile agnosia
Neglect, anosognosia, tactile agnosia
Body schema disturbances,
lobule
anosognosia
Frontal lobes
Frontal operculum
Broca’s aphasia
Expressive aprosody
Broca’a aphasia
Superior mesial
Akinetic mutism
Akinetic mutism
Severe akinetic mutism
region
Basal forebrain
Anterograde and retrograde amnesia
Anterograde and retrograde amnesia
Anterograde and retrograde amnesia
with confabulation
with confabulation
with confabulation
Orbital (inferior
Defective social contact, acquired
Defective social contact, acquired
Defective social contact, acquired
mesial region)
sociopathy
sociopathy
sociopathy
Dorsolateral
Impaired working memory for verbal
Impaired working memory for
Impaired working memory for verbal
prefrontal region
material, impaired executive function
nonverbal material, impaired
and nonverbal material, impaired
executive function
executive function
Subcortical structures
Basal ganglia
Dysarthria, aprosody, impaired
Dysarthria, aprosody, impaired
Dysarthria, aprosody, impaired
nondeclarative memory
nondeclarative memory
nondeclarative memory
Thalamus
Thalamic aphasia, amnesia with
Thalamic aphasia, amnesia with
Thalamic aphasia, amnesia with
confabulation, impairments in
confabulation, impairments in
confabulation, impairments in
attention and ‘executive functions’
attention and ‘executive functions’
attention and ‘executive functions’
Refer to Glossary for definition of medical terms. Reproduced with permission from Yudofsky SC and Hales RE (eds.) (2002) Textbook of Neuropsychiatry and Clinical Neurosciences,
5th edn. Arlington, VA: American Psychiatric Publishing, Inc.
memory dysfunction, language deficits, or sensorimotor symp-
toms. Gage was the foreman of a railroad construction crew
working in Vermont on September 13, 1848. He was known to
be the company's 'most efficient and capable foreman' with a
well-balanced mind and a shrewd business sense. He was tamp-
ing an explosive charge into the bottom of a borehole with a
tamping iron when the charge accidentally exploded resulting in
the tamping iron rocketing through his left cheekbone, prefron-
tal cortex, and anterior dorsal skull finally landing 25-30 yards
away. Gage survived the massive penetrating injury but dis-
played personality changes which his physician, Dr. John
M. Harlow, described in 1848 as follows: "He is fitful, irreverent,
indulging at times in the grossest profanity (which was not
previously his custom), manifesting but little deference for his
fellows, impatient of restraint or advice when it conflicts with his
desires." According to his friends, Gage 'was no longer Gage.'
Demasio and colleagues performed an analysis using clues from
the remains of Gage's skull and found that Gage's brain injury
likely involved bilateral anterior orbitofrontal cortex, polar
and anterior mesial frontal cortices, and the rostral portion
of the anterior cingulate gyrus, with underlying white matter
more extensively involved in the left hemisphere than in the
right (see Figure 8).
Depression and Schizophrenia: Nature and Nurture
Depression and schizophrenia are conditions which both
appear to result from a complex interplay of environmental
and biological factors. Depression occurs with a prevalence of
20-60% in persons with stroke, neurodegenerative diseases,
multiple sclerosis, headache, HIV, traumatic brain injury,
epilepsy, chronic pain, obstructive sleep apnea, intracranial
neoplasms, and motor neuron disease. It has been proposed
that primary depression is due to dysfunction in a network
that includes two known pathways: the orbitofrontal - basal
ganglia - thalamic circuit and the basotemporal limbic circuit
that links the orbitofrontal cortex and the anterior temporal
cortex by the uncinate fasciculus. This has been expanded into a
unifying depression circuit model which consists of four
interconnected functional compartments. Each functional com-
partment consists of strongly interconnected anatomic structures
upon which that compartment is dependent. The functional
compartments are as follows: mood regulation (medial frontal,
medial orbital-frontal and pregenu anterior cingulate cortex),
mood monitoring (ventral striatum-caudate, amygdale, dor-
somedial thalamus, midbrain-ventral tegmental area), intero-
ception (subcallosal cingulated, ventral-anterior hippocampus,
anterior insula, brain stem, hypothalamus), and exteroception
394 Brain and Behavior Relationships
Figure 8 Damasio and colleagues’ digital reconstruction of the
penetrating brain injury sustained by Phineas Gage in 1848 which
led to profound personality changes likely due to orbitofrontal cortical
damage. (The metal rod shown in red.)
(prefrontal, premotor, parietal, mid-cingulate, and posterior
cingulate cortices with dorsal-posterior hippocampus).
Efforts to link single genes to major depressive disorder
(MDD) have been unsuccessful, despite exhaustive mapping
attempts. Consequently, behavioral geneticists have turned
to the study of genetic polymorphisms in establishing a predis-
position to depression and in shaping the response to envi-
ronmental stressors. The most extensive studies in this field
have focused on polymorphisms in the serotonin transporter
(5-HTT) gene. Behavioral genetics research based on diathesis-
stress models of depression demonstrate that the risk of depres-
sion after a stressful event is enhanced in populations carrying
genetic risk factors and is diminished in populations lacking
such risk factors. Genotype-environment interactions are
ubiquitous, because genes not only impact the risk for depres-
sion by creating susceptibility to specific environmental stres-
sors but also cause individuals to persistently place themselves
in highly sttessful environments.
Among several etiological hypotheses for schizophrenia,
the neurodevelopmental model is most prominent. This model
generally posits that schizophrenia results from processes that
begin long before the onset of clinical symptoms and is caused
by a combination of environmental and genetic factors. Envi-
ronmental factors include being a first-generation immigrant or
the child of a first-generation immigrant, urban living, drug use,
head injury, prenatal infection, maternal malnunition, obstetri-
cal complications during delivery, and winter birth. Genetic risks
are clearly present but not well understood. The majority of
patients with schizophrenia lack a family history of the disorder.
The population lifetime risk for schizophrenia is 1%, 10% for
first-degree relatives, and 4% for second-degree relatives. There
is an —50% concordance rate for monozygotic twins compared
to that of -15% for dizygotic twins. It is most likely that many
different genes make small, but important, contributions to sus-
ceptibility. The disease manifests only when these genes are
combined or certain environmental factors are present. A num-
ber of susceptibility genes show association with schizophrenia.
A small proportion of schizophrenia incidence maybe explained
by genomic structural variations known as copy number variants
(CNVs). CNVs consist of inherited or de novo small duplica-
tions, deletions, or inversions in genes or regulatory regions.
These genomic structural variations conttibute to normal varia-
bility, disease risk, and developmental anomalies as well as act-
ing as a major mutational mechanism in evolution. The most
common CNV disorder, 22qll.2 deletion syndrome (velocar-
diofacial syndrome), has a 20-fold increased risk for schizophre-
nia and constitute about 0.9- 1.0% of schizophrenia patients.
Functional imaging studies in schizophrenics show de-
creased cerebral blood flow in the DLPFC during specific cogni-
tive tasks. Schizophrenic patients with prominent negative
symptoms display reduced glucose utilization in the frontal
lobes. Functional imaging studies suggest that disruption in
distributed functional circuits is important in the development
of schizophrenia. These functional circuit locations include the
DLPFC, OFC, mediofrontal cortex, anterior cingulate gyrus, thal-
amus, temporal lobe subregions, and the cerebellum. Several
conditions which may manifest psychosis, such as Huntington's
disease, Parkinson's disease, frontotemporal dementia, and
sttoke, are commonly associated with frontal and subcortical
dysfunction. Dorsolateral and mediofrontal hypoperfusion
on functional imaging has been demonstrated in a subset of
Alzheimer's disease patients with delusions.
Wisdom
The neurobiology of wisdom warrants special mention since it
lies on the exciting frontier of our ability to model sophisti-
cated cognitive processes. A consensus definition for wisdom is
not available. Since the dawn of civilization, definitions have
been advanced from different cultural, religious, and philo-
sophical viewpoints. A theoretical definition that takes into
account many cultural, religious, and philosophical themes
is that wisdom represents a demonstrated superior ability to
understand the nature and behavior of things, people, or
events. This results in an increased ability to predict behavior
or events which then may be used to benefit self or others.
There is more often a desire to share the accrued benefits with a
larger group for the purpose of promoting survival, cohesion,
or well-being of that group. The benefits do not result from
malicious or antisocial intents or inequitable behavior. Envi-
ronmental factors, such as family, education, socioeconomic
status, culture, and religion, are involved in generating the
milieu in which the personal value system develops. Many of
these same factors also influence how a given community
decides whether wisdom is present or not.
This model of wisdom relies on the individual's ability to
generate a mental representation of the self (cognitive, emo-
tional, and physical), the external world, and the dynamic
relationship of the self with the external world. The neural
systems critical to enable these functions are distributed but
heavily dependent on those that support memory, learning,
understanding other people's mental states (Theory of Mind),
and assigning relative value to information. The three
Brain and Behavior Relationships 395
frontosubcortical neural networks, the limbic system, and the
mirror neuron system are of particular importance for support-
ing these activities. These neural systems work in concert to
weigh and estimate the risks and benefits of various mentally
modeled courses of action to generate wisdom. The neural
substrates of empathy may be conceptualized as biasing the
information processing network in favor of valuing others,
interpersonal communication, cooperation, and community.
Summary
Advances in the natural sciences and medicine have resulted in
a shift from analysis with theological or philosophical expla-
nations of human behavior to those that are subject to scien-
tific methods. With this revolution in scientific thought has
come the understanding that emotions, behavior, and cogni-
tion are produced and controlled by distributed neural net-
works. The brain and, in turn, behavior are shaped by genetics
and a host of environmental factors throughout life. New
technologies, such as fMRI and PET, enable scientists to study
brain function in vivo and are beginning to yield insights into
mental illness and normal cognitive functions such as empathy
and wisdom which have traditionally been left to philosophi-
cal speculations. Advances in neuroscience will continue to
change our understanding of ourselves and our relationship
with the world around us.
See also: Agnosia (including Prosopagnosia and Anosognosia);
Agraphia and Alexia; Aphasia; Behavioral Pharmacology; The Brain;
Brain Chemicals: Global Projections of Ancient Aromatic
Neurotransmitters; Depression; Drugs, the Brain, and Behavior;
Empathy; Empirical Challenges to Conventional Mind-Brain Theory;
Memory, Neural Substrates; The Mind-Body Problem; Neuroexecutive
Function; Neurotechnologies; Schizophrenia.
Further Reading
Blumenfeld H (201 0) Neuroanatomy Through Clinical Cases, 2nd edn. Sunderland, MA:
Sinauer Associates.
Caspi A, Sugden K, Moffitt T, Taylor A, et al. (2003) Influence of life stress on
depression: Moderation by a polymorphism in the 5-HTT gene. Science
301: 386-389.
Cummings JL and Mega MS (eds.) (2003) Neuropsychiatry and Behavioral
Neuroscience. New York: Oxford University Press.
Damasio H, Grabowski T, Frank R, Galaburda A, and Damasio A (1994) The return of
phineas gage: Clues about the brain from the skull of a famous patient. Science 264:
1102-1105.
Devinsky 0 and D’Esposito M (2004) Neurology of Cognitive and Behavioral Disorders.
New York: Oxford University Press.
Kandel ER, Schwartz JH, and Jessell TM (eds.) (2000) Principles ot Neural Science,
4th edn. New York: McGraw-Hill.
Mesulam MM (ed.) (2000) Principles of Behavioral and Cognitive Neurology.
New York: Oxford University Press.
Murray ED and Price BH (2008) Depression and psychosis in neurological
practice. In: Bradley WG, Daroff RB, Fenichel GM, and Jankovic J (eds.)
Neurology in Clinical Practice, 5th edn., pp. 101-120. Philadelphia, PA:
Buttorworth-Heinmann.
Price BH, Raymond AD, and Coyle JT (2000) Neurology and psychiatry: Closing the
great divide. Neurology 54: 8-14.
Proescher F and Arkush A (1 930) Incidence of syphilis in insanity. American Journal of
Psychiatry 87: 245-253.
Stoodley CJ and Schmahmann JD (2010) Evidence for topographic organization in the
cerebellum of motor control versus cognitive and affective processing. Cortex
46: 831-844.
Yudofsky SC and Hales RE (eds.) (2002) Textbook of Neuropsychiatry and Clinical
Neurosciences, 5th edn. Arlington, VA: American Psychiatric Publishing, Inc.
Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters
E C Azmitia, New York University, New York, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Aromatic structures They are usually six member carbon
(benzene) or five (pyrrole) or six member (Pyridine)
carbon-nitrogen ring structures having alternating single
and double bonds. These ring structures are biologically
important because they are able to absorb the energy from a
photon which has the energy equivalent to that of a
hydrogen electron. Aromatic (ring) hydrocarbons can be
used as electron donors aerobically. The benzene-based
amino acids (tyrosine, phenylalanine) do not lead to
initiation of photosynthesis. However, indole-based amino
acids (tryptophan), which have a five member aromatic ring
with a nitrogen molecule attached to the benzene, donate an
electron to a heavy metal and begin the process of
photosynthesis.
Glial cells There are three main types of glial cells: the
oligodendroglia cells, the astrocytes, and the microglia cells.
These cells made their appearance in the brain as it became
more complex during animal development. The
oligodendroglia cells wrap themselves around axons and
provide guidance, stability, and insulation to the axons of
PTP neurons. The astrocytes function to support neurons by
providing trophic factors (e.g., S100B, GFAP, and GDGF),
energy (e.g., glucose), ions (e.g., K + ), and guidance factors
(e.g., laminin). Microglial cells are of mesoderm origin
(as opposed to the ectodermal origin of most other brain
cells - blood vessels are also of mesodermal origin).
These aggressive cells multiply, are mobile, and contribute
principally to inflammation processes which occur after
brain injury.
Global neurons These neurons are evolutionarily ancient
and have extensive connections throughout the brain. They
have their cells of origin in the brainstem regions but appear
to connect with most neurons in the cortex, cerebellum, and
spinal cord. These neurons use monoamines as
neurotransmitters. These chemicals, made from indole or
benzene structures, have powerful actions using a multiple
number of specific receptors. Brain chemicals are indole and
benzene ring-containing amino acids.
Interneurons These neurons are almost always inhibitory,
using GABA as their neurotransmitter, and are mainly
involved in the regulation of the local circuitry. The
interneurons are involved in feedback inhibition and in
defining the limits of neuron activity as best seen in sensory
systems. The interneurons develop by a different pattern
than that seen with either the global or PTP neurons. The
interneurons principally use GABA, a derivative of the
principal excitatory PTP chemical glutamate.
Limbic system James Papez first conceived the idea of a
limbic system in 1937 after reviewing brain injuries in
soldiers that resulted in emotional problems. The Papez
circuit consisted of the hippocampus, mammillary bodies,
anterior thalamus, and the cingulate cortex. Over the years,
many regions have been added to the emotional brain as
part of the limbic system. These include the raphe serotonin
system, amygdala, and septum; some even include the
cerebellum, hypothalamus, parahippocampal gyrus,
olfactory bulbs, and orbitalfrontal gyrus. As the number of
structures increases in this 'emotional circuit,' its utility as an
interesting concept decreases.
Medial forebrain bundle and lateral forebrain
bundle These two, largely unmyelinated, ancient fiber tracts
have fibers extending to and from the midbrain-pons and
the forebrain of the reptilian and mammalian brain. These
tracts are used by the aromatic neurotransmitters systems
(serotonin, dopamine, and noradrenaline) to reach all areas
of the telencephalon and diencephalon, brain regions rostral
to the midbrain. All cells of the forebrain are near one of
these aromatic neurotransmitter axons.
Neuromuscular neurons These neurons project from the
spinal cord and brainstem to innervate and activate skeletal
muscle. These neurons use the brain chemical acetylcholine.
Acetylcholine is also used by other PTP systems in the limbic
system and also as a global cortical projecting system from
the rostral brainstem.
Neurosecretory neurons These cells project from the
hypothalamus to direcdy reach the posterior pituitary or the
blood vessels going to the endocrine cells of the anterior
pituitary. These neurons are part of the limbic pathway or in
close association with the global neurons. The brain
chemicals used by these secretory cells are almost
predominately large peptides such as ACTH, oxytocin,
anandamide, or vasopressin.
Photosynthesis This is a process whereby solar energy is
converted to biological energy that converts carbon dioxide
into organic compounds, especially sugars, resulting in
the release of 0 2 and the creation of ATP and reduced
cofactors such as NADH. In the second stage of
photosynthesis, ATP and NADH can be used to convert
C0 2 to sugars.
Point to point (PTP) neurons These neurons are
evolutionarily a more recent addition in that they are part of
what is referred to as the hard wiring of the brain. The
neurons have specific and localized projections and are used
to relay specific bits of information into and out of the brain.
These neurons use glutamate as the excitatory and glycine
as the inhibitory neurotransmitters. Brain chemicals are
amino acids with simple side chains such as glutamate
and glycine.
Terminal neurons There are two types of neurons that are
involved in transmitting the results of brain activity to the
rest of the body. Motoneurons serve to move muscles and
cause the body to move. The neurosecretory neurons release
hormones (i.e., steroids and peptides) into the blood.
396
Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters 397
I was sitting writing at my textbook but the work did not progress;
my thoughts were elsewhere. I turned my chair to the fire and dozed.
Again the atoms were gambolling before my eyes. This time the
smaller groups kept modestly in the background. My mental eye,
rendered more acute by the repeated visions of the kind, could now
distinguish larger structures of manifold confirmation: long rows,
sometimes more closely fitted together all twining and twisting in
snake like motion. But look! What was that? One of the snakes had
seized hold of its own tail, and the form whirled mockingly before
my eyes. As if by a flash of lightning I awoke; and this time also I
spent the rest of the night in working out the rest of the hypothesis.
Let us learn to dream, gentlemen, then perhaps we shall find the
truth . . . But let us beware of publishing our dreams till they have
been tested by waking understanding. (August Kekule, 1890)
Introduction
In the first edition of the Encyclopedia of Human Behavior, I
presented an article on brain chemicals that focused on neuro-
transmitters and neurotrophic molecules. Neurotransmitters
were placed into three different categories depending on the
size, rate of synthesis, and binding characteristics to receptors.
A fast system used acetylcholine, glutamate, or GABA, which
are small molecules rapidly available to the neuron with mini-
mal synthetic processing. These chemicals could bind to their
receptors with relatively low affinity to enable quick on and off
states. The slow systems used peptides such as enkephalin,
oxytocin, substance-P, or anandamide, which are large mole-
cules stored in dense core vesicles and slowly available for
synaptic release. The synthesis of these peptides begins in the
neuronal soma and involves extensive axonal processing.
These chemicals bind to their receptors with high affinity to
produce very slow on and off states and prolonged activation.
The intermediate system used the aromatic monoamines
such as serotonin, dopamine, and noradrenaline. These aro-
matic neurotransmitters are available in limited quantities
because the amino acids tryptophan (indole ring) and tyrosine
(benzene ring) must be obtained from the diet. Once the
aromatic amino acids are stored in the neurons, the synthesis
requires molecular oxygen and reduced cofactors (both chemi-
cals important for mitochondrial production of ATP) to pro-
duce the final monoamine chemical. These transmitters are
ancient and have multiple binding proteins in the human
brain, ranging from low to high affinity to permit a wide
range of actions. The intent of the present article is not to
expand on the previous edition, but to focus on the brain
aromatic chemicals required for the intermediate system as
these brain chemicals form a global projecting system involved
in brain homeostasis.
Obviously, the range of human behavior is dependent on
the neural circuits contained in the human brain. Most of the
research strategies utilized to study human behavior focus on
the point to point (PTP) circuits (Figure 1). For example, to
understand vision, the processes begin in the eye, but then the
connections of the eye with the lateral geniculate and the visual
cortex are followed and examined. Similar pathways are avail-
able for the behaviors involved in all the senses where there is a
PTP progression from the principal sensory receptors to a final
stop in a specific region of the sensory cortex, with relays
occurring in thalamus or brainstem. The ancient system of
olfaction is different, but still possesses a PTP circuit that can
be drawn, usually utilizing black arrows. PTP diagrams are also
available to illustrate the limbic system and cortical processing,
with information transfer moving neatly from one point to
another. The brain chemicals required for this type of network
would be as vast as all the receptors needed, neurotransmitters
released, and trophic factors required to build and maintain
each PTP connection. It is not the intention of this article to
discuss the neurotransmitters of the PTP circuits.
Rather, I will focus on the aromatic compounds - indole
and benzene - which form the backbone of the amino acids of
tryptophan and tyrosine. These are the essential chemicals
needed by the global systems that utilize the monoamines.
The unique properties of the aromatic chemicals were central
to the unicellular process of photosynthesis, and regulated
such processes as cell division and motility. The presence of
these aromatic amino acids (AAA) and their derivates enable
the survival and growth of solar powered oxygen producing
First order system
global projecting neuron
Second order system
point-to-point network
Figure 1 The figure illustrates drawings showing global projecting system (a) and point-to-point (PTP) systems (b). The global systems are
proposed to be more primitive and are called first-order systems while the PTP systems are a more recent addition and considered as
second-order systems.
398 Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters
organisms. However, when the animal kingdom separated
from the plant kingdom, the AAA could no longer be produced
endogenously because of the loss of chloroplast and genes
necessary for indole and benzene synthesis. These chemicals
have to be obtained from the diet.
During the evolution of the brain, the AAA developed a
global system for the distribution of aromatic neurotransmit-
ters that depended on the medial and lateral forebrain bundles
to reach target areas and then on multiple receptors to transmit
their physiological actions. The important aspect of receptor-
mediated actions is that the neurotransmitters are not
degraded, and reuptake mechanisms are used to return these
chemicals to the aromatic neurotransmitter axon. This exten-
sive distribution of the global neurotransmitter system pro-
vides insights into many of the human behaviors influenced
by these chemicals. Not only are many of these behaviors
linked to the primitive networks of the limbic system but
they also regulate many glial trophic factors necessary for the
maintenance of the PTP networks throughout the forebrain.
Thus, the concept is that these global neurons are involved in
the homeostatic regulation of the brain itself, and its interac-
tion with the rest of the body functioning within the external
environment. Furthermore, an understanding of these brain
chemicals in the global neurons provides fresh insights into
concepts such as awareness, depression, and consciousness.
Aromatic Chemicals
Many of the chemicals found in the brain that are involved in
important behaviors in humans also served important roles in
the life, growth, and division of unicellular organs. Three bil-
lion years ago, unicellular organisms began to produce oxygen
by the process of photosynthesis. To make the monoamine
neurotransmitters, oxygen is needed (Figure 2). In the earliest
geological times, the earth's atmosphere had little oxygen.
Thus, the monoamines were made specifically in unicellular
systems capable of photosynthesis and the cellular production
of oxygen. The conserved monoamine biosynthetic pathway
began in the unicellular systems of cyanobacteria, green algae,
and fungi, and continually evolved to its current position in
the human brain.
Benzenes
Two amino acids have a benzene backbone: phenyalanine and
tyrosine. The delocalization of electrons is one explanation for
the thermodynamic stability of benzene and related ringed
molecules. It is likely that it is this stability that contributes to
the peculiar molecular and chemical properties known as aro-
maticity, a term originally used to describe their fragrance. The
concept of quantum-mechanical resonance was developed by
W. Heisenberg in 1926 and applied to the benzene molecule
by L. Pauling. Pauling wrote that benzene bonds 'would then
be expected to be a sort of average of the properties of the
individual molecules.'
Catechols
Benzenes can be oxidized by both bacteria and eukaryotes. In
bacteria, the dioxygenase enzyme adds two oxygen molecules
Benzene
Figure 2 This figure shows the molecular structures of benzene,
pyrrole, and indole rings. Double bonds are illustrated as localized in the
first two structures, and as being nonlocalized in the third structure of
each set.
to the ring, and the unstable product is immediately reduced
(by the cofactor NADH) to a cyclic diol with two double
bonds, breaking the aromaticity. Next, the diol is reduced by
NADH to form catechol. Catechol can be metabolized to acetyl
CoA and succinyl CoA, used by organisms mainly in the Krebs
Cycle for energy production. The catechol skeleton occurs in
catecholamines, drugs derived from them (such as MDMA),
urushiols (skin-irritating poisons found in plants like poison
ivy), and catechin (found in tea).
Pyrrole
Pyrrole is an aromatic five-member ring containing four car-
bons and a single protonated nitrogen. There are two double
bonds which alternate between the five atoms. It is an impor-
tant component of such complex light gathering macromole-
cules as chlorophyll.
Indole
This ring structure is an organic compound consisting of a
benzene ring and a pyrrole ring (Figure 3). The most reactive
position on indole for electrophilic aromatic substitution is
C-3, which is 10 13 times more reactive than benzene. The
origin of the first photosynthetic pathway is proposed to be
the photooxidation of uroporphyrinogen (an indole-rich com-
pound) by sunlight and this oxidation is accompanied by the
release of molecular hydrogen. The indole structure can be
found in the amino acid tryptophan. This molecule also
forms the light-sensitive component of such molecules as,
niacin, melatonin, tryptophan-containing enzymes and recep-
tors, alkaloids, and pigments. The indole ring is electron rich
and will easily lose a nitrogen electron (oxidized) to an elec-
trophilic compound, like a heavy metal.
Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters 399
Tyrosine
0 Tyrosine hydroxylase
H 4 -biopterin H 2 -biopterin
Dihydroxyphenyalanine
(DOPA) 0
Pyridoxal
phosphate
Aromatic
amino acid
decarboxylase
Dopamine
0 2 ascorbic
acid
Dopamine
B-hydroxylase
H 2 0 dehydro-
ascorbic acid
S-adenosyl
Phenylethanolamine
Epinephrine N-methyltransferase Norepinephrine
Figure 3 The biosynthetic steps are shown for the enzymatic conversion of tyrosine to dihydroxyphenyalanine (DOPA), dopamine, norepinephrine
(noradrenaline), and epinephrine (adrenaline). The names of the enzymatic steps as well as the main cofactors are given.
Biosynthesis of Aromatic Amino Acids
Hydroxylation
Hydroxylase is the first enzyme in the synthesis of the aromatic
neurotransmitters serotonin and catecholamines. In animals,
tryptophan, phenylalanine, and tyrosine are obtained exclu-
sively from the diet. Tyrosine can be converted from phenylal-
anine by the enzyme phenylalanine hydroxylase. The next
step is also mediated by a hydroxylase enzyme, tyrosine
hydroxylation. This enzyme produces l-DOPA, which retains
the catechol structure.
The three enzymes involved in aromatic hydroxylation
evolved from a single primitive enzyme that reacted with phe-
nylalanine, tyrosine, and tryptophan, which were all sensitive
to light. The hydroxylase enzyme was involved in the synthesis
of large number of complex alkaloids in plants, all of which are
potent antioxidants in their own right. Cellular oxidation is
important for cell maturation and division, but excess oxida-
tion results in cell death. The synthesis of pharmaceutically
important indoles also involves decarboxylase enzymes (3.2),
discussed below.
The hydroxylase enzyme, which utilizes molecular oxygen
and H 4 -biopterine, is the first enzyme in the pathway for
both serotonin and catecholamine synthesis. 5-HTP, the
immediate precursor of serotonin, is formed from tryptophan
and l-DOPA, the immediate precursor of noradrenaline, is
formed from tyrosine. The ubiquitous carboxylase enzyme,
working in reverse to function as an aromatic amino acid
decarboxylase, rapidly converts both these molecules to
serotonin or dopamine (Figure 4).
Decarboxylase
The second enzyme in serotonin biosynthesis is tryptophan
decarboxylase. The first organisms to contain this enzyme may
have been the bacteria, Enterobacteria cloacae strains, which are
normally associated with plant roots, and help produce the
plant hormone auxin by using the enzyme indolepyruvate
decarboxylase. Aromatic L-amino acid decarboxylase occurs in
plants. For example, tryptophan decarboxylase in plants leads
to the biosynthesis of pharmaceutically active indole alkaloids.
These compounds are abundant throughout the plant king-
dom, and are known to produce strong psychic or hallucino-
genic responses in humans (see Figure 3). In animals, this
enzyme loses its specificity for tryptophan and becomes a
general aromatic acid decarboxylase. It is likely that as animal
cells had a marked reduction in tryptophan because of loss of
tryptophan synthesizing enzymes, there was less evolutionary
need for a specific tryptophan decarboxylase enzyme.
Thus, enzymes commonly used in anaerobic organisms
before 0 2 was formed inside cells produce serotonin and the
catecholamines. Besides the algae, fungi, and molds, the most
efficient generators of 0 2 and monoamines are plants. In
humans, the collection of the AAA is quite complex because
these structures are no longer made endogenously. A single trans-
port system exists for phenylalanine, tyrosine, and tryptophan
400 Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters
Tryptophan
Tryptophan hydroxylase
H 4 -biopterin H 2 -biopterin
5-OH-tryptophan
Pyridoxal
phosphate
Aromatic
amino acid
decarboxylase
5-OH-indole aldehyde
NH3+H2O2 H2O + O2
Monoamine
oxidase
HO
ch 2 ch 2 — nh 2
5-OH-tryptamine (5-HT)
serotonin
Figure 4 The biosynthetic steps are shown for the enzymatic conversion of tryptophan to 5-hydroxytryptamine, serotonin, and 5-OH-indole
aldehyde (5-HIAA). The names of the enzymatic steps are given as well as the main cofactors.
to carry these molecules to the brain. Once in the brain, the
amino acids have specific uptake into the global neurons where
they are converted into the neurotransmitter chemicals. This
conversion occurs in the axon in the terminal areas of the brain.
The neurotransmitters are released either in vesicles after an
action potential or from the cytoplasm not in vesicles by a
Ca 2+ -independent process. These chemicals have multiple recep-
tor systems to translate their biological actions without the
metabolism of the neurotransmitter. The neurons also contain
a very effective reuptake protein that removes the neurotransmit-
ter from the synapse and either releases it again or retrogradely
transports it back to the neuronal soma in the brainstem.
Derivatives of Aromatic Neurotransmitters
As bacteria and plants became very efficient at converting light
into biological energy (NADH and ATP), the excess production
of 0 2 became a biochemical problem: oxygen can easily gener-
ate free radicals that are very destructive to the interior of the
cell. The remedy was to have 0 2 bind directly to tryptophan
and tyrosine. Many of the biochemical derivatives of trypto-
phan and tyrosine evolved to function as antioxidant mole-
cules in simple cells long before they assumed more complex
functions in animals and humans as hallucinogens and mind-
altering drugs.
Catecholamine
Catecholamine derivatives are known to serve important
regulatory roles in plants and in humans. The sympathetic
component of the peripheral nervous system (PNS) uses nor-
adrenaline and the adrenal gland secretes adrenaline during
stress response. Many of the benzene based alkaloid and syn-
thetic compounds are also hallucinogenic. These chemicals act
as dopamine and noradrenaline agonists with actions on both
the central nervous system (CNS) and PNS. In addition to their
hallucinogenic effects, they produce nausea, psychosis, racing,
mania, and panic attacks (Figure 5).
Catecholamines
They are called catecholamines because they contain a catechol
or 3,4-dihydroxylphenyl group. They are derived from the
amino acid tyrosine. Catecholamines are produced mainly by
the chromaffin cells of the adrenal medulla and the postgan-
glionic fibers of the sympathetic nervous system. Noradrena-
line neurons, part of the global neurons, are largely confined to
melanin-pigmented cell bodies of the locus ceruleus in the
Pons. Dopamine, a global neurotransmitter in the CNS, is
largely produced in neuronal cell bodies in two areas of the
brainstem: the substantia nigra and the ventral tegmental area.
Epinephrine (adrenaline)
It is a hormone and neurotransmitter. It increases heart rate,
contracts blood vessels, dilates air passages, and participates in
the fight-or-flight response of the sympathetic nervous system.
Epinephrine acts on nearly all body tissues. Its actions depend
on the receptor expression of the tissue. For example, epineph-
rine causes smooth muscle relaxation in the esophagus, but
causes contraction of the smooth muscle that lines arterioles.
Adrenaline is a nonselective agonist of all adrenergic receptors.
Mescaline
Peyote is found in high levels in the buds from the cactus pants
in Mexico. This drug can produce deep introspection and
insight described as being of a metaphysical or spiritual nature.
This hallucinogen is rich in both visual and auditory effects
(synesthesia). Mescaline was first isolated and identified in
1897. Its side effects include paralysis, breathing difficulty,
vomiting, tremors in the legs, and anorexia.
Phenethylamine
The group of phenethylamine derivatives is referred to as the
phenethylamines. Substituted phenethylamines, substituted
amphetamines, and substituted methylenedioxyphenethyla-
mines are a series of broad and diverse classes of compounds
derived from phenethylamine that include stimulants, psyche-
delics, and entactogens, as well as anorectics, bronchodilators,
decongestants, and antidepressants, among others.
Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters 401
nh 2
Mescaline
Methamphetamine
MDMA (ecstasy)
Ephedrine Amphetamine Selergilin (deprenyl)
Figure 5 The physiologically relevant derivatives from catecholamines are shown. These include the principal neurotransmitters dopamine and
noradrenaline. Mescaline is a strong hallucinogen while the phenethylamines include ephedrine and selergilin. These are all alkaloids with mind-altering
properties. Amphetamine, DOM (2,5-dimethoxy-4-methylamphetamine), methamphetamine, and ecstasy are designer drugs. They are synthetic
phenethylamines with strong sympathetic and mind-altering properties.
Ephedrine
Chemically, it is an alkaloid derived from various plants in the
genus Ephedra (family Ephedraceae). Ephedrine is a sympa-
thomimetic amine. The principal mechanism of its action is on
the nonvesicular release of noradrenaline in the sympathetic
nervous system of the PNS. Ephedrine crosses the blood-brain
barrier weakly. Behavior effects of this drug include mania,
hallucinations, delusions, agitation, confusion, mild euphoria,
paranoia, hostility, and panic.
Selegiline (deprenyl)
It is a synthetic phenethylamine which functions as a potent
inhibitor of the enzymes MAOA and MAOB, the principle
routes for the degradation of the aromatic neurotransmitters.
The drug results in intraneuronal concentration of these aro-
matic neural transmitters and is commonly used as an adjunct
to L-DOPA therapy or ecstasy and amphetamine abuse. Effects
of these sympathomimetic stimulants include hallucinations,
dizziness, dry mouth, severe headache, tachycardia, arrhyth-
mia, and panic.
Designer drugs
They have been synthesized from using the amphetamine
backbone. Many of these compounds were first synthesized
by Alexander Shugun and are discussed in his book PiHKAL
(Phenethylamines i Have Known And Loved).
Amphetamine
It is a synthetic psychostimulant drug of the phenethylamine
class. This drug effects the release of dopamine and noradrena-
line in the CNS and PNS by a Ca 2+ -independent nonvesicular
release mechanism. The drug is a stimulant coupled with a
feeling of euphoria. The class includes prescription CNS
drugs paradoxically used to treat attention-deficit hyperactivity
disorder (ADHD).
Methamphetamine
It has high potential for abuse and addiction by activating the
psychological reward system via increasing levels of dopamine,
noradrenaline, and serotonin in the brain. The US Food and
Drug Administration (FDA) has approved methamphetamine
for the treatment of attention-deficit hyperactivity disorder
(ADHD) and exogenous obesity. Methamphetamine is mar-
keted in the United States under the trademark name Desoxyn.
DOM (2,5-dimethoxy-4-methylamphetamine)
It is also known as STP, an abbreviation for 'serenity, tranquil-
ity, and peace.' This is a powerful psychedelic (hallucinogenic
drug) and a substituted amphetamine which has its hallucino-
gen properties by stimulating serotonin receptors. DOM is a
very potent agonist of the serotonin 5-HT family of receptors;
mainly of the 5-HT2 subtype. Effects of this drug include
substantial perceptual alterations such as visual hallucinations,
double vision, distortions, vibration of objects, slowing of
time, and panic.
Serotonin
This is the principle substance used as a neurotransmitter in the
nervous system of all animals, and has been found in nearly all
photosynthetic organisms. Two major compounds with struc-
tural similarities to serotonin are auxin and melatonin, which
serve important regulatory processes in plants and humans.
Other stmctural derivatives of serotonin function as antioxi-
dants, trophic molecules, or strong hallucinogenic drugs.
402 Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters
Auxin (hydroxyindoie acetic acid)
It is also a protective, antioxidant compound found in unicel-
lular organisms and functions to stimulate growth and sur-
vival. Auxin acts as an important phototrophic hormone in
plants and induces cell elongation and cell division, with all
the subsequent results for plant growth and development. Its
specific function in leaves is to move the entire photogenerat-
ing organ toward the source of sunlight.
Melatonin
It functions as a powerful antioxidant. Melatonin is found in
all plants and animals, including algae and is involved in
microtubule function in flagella and in mitosis. Melatonin
levels have a diurnal rhythm in many protists and plants, as
well as in humans. Melatonin functions as an antioxidant in
plants and as a biological signal for the organization of day-
length-dependent annual functions such as reproduction,
behavior, coat growth, and camouflage coloring.
Indolamines
There are many natural psychoactive drugs synthesized in
plants from tryptophan. All these compounds function as
antioxidants in single-cell organisms, and have been used by
man as entheogens for thousands of years. The ability of these
drugs to induce a feeling of closeness to God is a special
property of the indoles and this property is attributed to acti-
vation of the cortical 2A serotonin receptor. Substituted
indoles are derivatives of the tryptophan-based tryptamine
alkaloids, such as serotonin, melatonin, and auxin; the hallu-
cinogens psilocybin, DMT, and 5-MeO-DMT; and the ergots,
such as ergotamine and LSD. N-Formylkynurenine, 5- methox-
ytryptamine, auxin, and melatonin in prokaryotes function as
photosensors, antioxidants, and pattern generators for flagella
movement, and in defense mechanisms. These alkaloids pro-
duced from tryptophan have powerful actions on the human
brain by acting through the serotonin receptors or reuptake
proteins. The structures of some of these compounds are
shown in Figure 6.
Psilocybin mushrooms
Psilocybin mushrooms are found mainly in South and Central
America. Native Americans (e.g., Aztecs, Mayans) used this
hallucinogenic drug for thousands of years. It was almost
always used in religious ceremonies with the users believing
Figure 6 The physiologically relevant derivatives from indolamines are shown. Auxin is a powerful plant hormone. Melatonin is a powerful hormone in
humans and found in plants. The indoleamines are powerful hallucinogens. The alkaloids include psilocybin, /V,/V-dimethyltryptamine (DMT),
5-methoxytryptamine, ibogaine, and yohimbin. LSD is a semisynthetic psychedelic drug of the ergoline family, which is one of the most powerful
mind-altering drugs known.
Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters 403
they were moving between the earthly and supernatural
realms. Its properties are due to activation of the cortical 2A
serotonin receptor.
I' i,H-Dimethyltryptamine (DMT)
This alkaloid occurs in many species of plants and in trace
amounts in the human brain. N,N-Dimethyltryptamine (DMT)
is a naturally occurring hallucinogenic drug of the indoleamine
family and functions as a nonselective agonist at most or all of
the serotonin receptors (including 5-HT2A) with some affinity
for dopamine adrenergic receptors. DMT can produce deper-
sonalization, near-death experiences and mystical states.
Ibogaine
It is found in a number of plants, principally in a member
of the dogbane family known as iboga (Tabernanthe iboga).
Ibogaine-containing preparations are used in medicinal and
ritual purposes within African spiritual traditions. Ibogaine
experience occurs as two distinct phases: the visual phase and
the introspective phase. The visual phase is visual distortion
often accompanied by laughing, euphoria, fear, and temporary
short-term memory impairment. The introspective phase is
elevated mood, calm, and euphoria, and intellectual and
emotional clarity.
Yohimbine
It is the principal alkaloid of the bark of the West-African
evergreen Pausinystalia yohimbe. In Africa, yohimbine has tradi-
tionally been used as an aphrodisiac. Yohimbine is used as a
dietary supplement in herbal extract form and as a prescription
medicine in pure form for the treatment of sexual dysfunction.
Yohimbine can produce hallucinations, anxiety reactions,
rapid heart rate, high blood pressure, insomnia, seizures, and
panic attacks.
Lysergic acid diethylamide (LSD)
LSD is a semisynthetic psychedelic drug of the ergoline family.
The parent compound of the major hallucinogen LSD has
long been known to be produced by a fungus, genus Claviceps.
The hallucinogen was synthesized by the addition of diethyla-
mide to ergotamine by the chemist Albert Hofmann at the
Sandoz Laboratory. LSD is believed to be one of the most
potent mind-altering compounds discovered to date. LSD
causes hallucinations, distortions, and an altered experience
of senses, emotions, memories, time, and awareness. There can
be distortions of sensory perception such as synaesthesia, the
experience of additional spatial or temporal dimensions, and
temporary dissociation.
Homeostasis
In a discussion of brain chemicals and human behavior, the
aromatic neurotransmitters are an important topic not only for
the variety of behavior under their control, but also for the
intensity and 'mind-altering' aspects of the induced states.
Serotonin is involved in aggression, dominance, memory,
mood, appetite, and sleep. Biologically, it is implicated in
neurogenesis, movement, cell division, and apoptosis. Dopa-
mine is directly implicated in addiction, pleasure, and repeti-
tive motor activity. Noradrenaline has actions on all the
sympathetic responses in the PNS and CNS. The derivatives
of these compounds, many found in plants, produce halluci-
nations and altered states of consciousness. To appreciate the
spectrum and nature of these actions in human behavior, it
helps to have an evolutionary perspective.
The aromatic neurotransmitters were found in very high
levels in unicellular organisms and plants because of the avail-
ability of aromatic amino acids and oxygen. The aromatic
amino acids were largely synthesized in plastids (chloroplast),
which are the major organelles in the cells of plants and algae.
Most proteins have an intrinsic UV fluorescence as a result of
photoabsorption because of the aromatic amino acids in their
structure, particularly phenylalanine, tyrosine, and tryptophan.
Comparing the latter two aromatic amino acids, tryptophan has
the higher fluorescence quantum yield, overshadowing mark-
edly the emissions of the tyrosine. The amino acid tryptophan
contains an indole backbone with a reactive pyrrole ring and
absorbs light very effectively. Free tryptophan has characteristic
fluorescence absorption at UVB (450-480 nm wavelength), and
the fluorescence emission is in the range of UVA-blue light. This
amino acid in proteins is needed for solar energy to drive
photosynthesis in cyanobacteria, algae, and plants. The absorp-
tion of a UVB photon (energy of a photon of blue light of
wavelength 450 nm = 2.76 eV) is equivalent to the energy of a
hydrogen electron. Therefore, it appears that the primary pho-
toionization pathway involves base ionization of tryptophan
followed by deprotonation. The energy equivalent of solar
energy conversion in photosynthesis involves electron transfer
between an excited donor molecule and an acceptor molecule
that are contained in the reaction center, an intrinsic membrane
protein pigment complex. In the photosynthetic reaction cen-
ters of Rhodobacter sphaeroides, there are 39 tryptophan resi-
dues. The importance and abundance of tryptophan, as well as
the catecholamines, have made these chemicals central to many
cellular functions such as energy production and cytoskeleton
stability that regulates cell mobility and cell division (Figure 7).
As cells began to produce large amounts of 0 2 , the forma-
tion of free oxygen radicals became a serious problem. The
hydroxylation of aromatic amino acids removed not only a
molecule of 0 2 , but also the aromatic amino acid. Over time,
many of these aromatic amines became effective antioxidants
and served to regulate the homeostasis of algae and plants.
When animals diverted from plants, the plant plastid
organelle (chloroplast) was lost and with it the ability to pro-
duce aromatic amino acids. Animals could only survive if they
were symbiotic to plants or algae or digest their material. The
animals were still largely dependent on the aromatic chemi-
cals, but now developed a new strategy for maximizing the
effects from these compounds by evolving receptor molecules
that could produce a metabolic response without metabolizing
the neurotransmitter. The neurons developed global projec-
tions throughout the brain using the ancient forebrain bundles
and the largely unmyelinated axons also traveled near blood
vessels and in the ventricular space. The aromatic neurotrans-
mitters were synthesized from products obtained in the target
regions. They evolved reuptake proteins for retrieving any
secreted aromatic amino acid and are able to transport chemi-
cals made in the terminals back to the cell somas in the mid-
brain and pons. Because of this extensive network and multiple
receptors, the human brain has maintained a homeostatic role
404 Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters
Indole
Chloroplast
Chorismate
Mitochondria
Catechol
Hallucinogens
Tryptophan
Melatonin \
Photosynthesis
Oxygen
Jl
Hydroxylase enzyme
Hallucinogens
Epinephrine
S'
Norepinephrine
Serotonin
Dopamine
Figure 7 This is a line diagram illustrating the close biological interactions between the aromatic amino acids in the production of neurotransmitters
(serotonin and catecholamine) and hormones (melatonin and adrenaline). The key process of photosynthesis in producing oxygen and NADH in
the biosynthesis of these chemicals is shown. The importance of chloroplast (plant plastid) and mitochondria organelles is illustrated. The
hallucinogenic derivatives are included as these compounds are alkaloids and serve antioxidant functions in the algae and plant cells.
Figure 8 This drawing shows the central role of aromatic global
neurons in the homeostasis of the brain. The various biological
components that interact with these neurons are shown in small type and
include vascular and digestive processes. The biological impact of these
neurons can be seen on tissues from all three types of cells: endoderm,
mesoderm, and ectoderm. The influence of these neurons affects CNS,
PNS, and endocrine systems.
for aromatic neurotransmitters, despite an inability to synthe-
size the precursor amino acids. This is consistent with the early
proposal of Brodie and Shore (1964). In their hypothesis,
noradrenaline and serotonin modulated opposite systems in
the brain on the basis of Walter Hess's (1949) concept of
the functional integration of the autonomic system with the
CNS. The PNS uses acetylcholine to mediate the sympathetic
response and noradrenaline to mediate parasympathetic
response. Serotonin was the modulator of the trophotrophic
system, which integrates behavioral patterns that are recupera-
tive in nature. This was considered a recessive system, which
normally functions during sleep or hibernation (Figure 8).
A new concept of consciousness proposes that the mind is
an amalgamation of the actions of the PTP and global neurons.
Therefore, not only do the aromatic amino acids serve homeo-
stasis involved in cellular and behavioral processes, but deal
with inner process of the mind. Many of the alkaloid and
synthetic derivatives have mind-altering properties in humans.
The actions of the aromatic amino acids in early life to
capture solar energy may explain why decreases in aromatic
neurotransmitters in the human brain lead to feelings of
worthlessness and suicidal ideation. In certain conditions,
these states in humans are elevated by exposure to blue light.
Brain chemicals are found in all the primitive cells, and the
most interesting of these may be the aromatics.
Acknowledgment
I would like to thank Mr. Raymond Xu, a neuroscience student
at NYU who assisted me in writing and editing this manuscript.
I would also like to thank Daniel Stein, Science Dean at NYU,
for his support.
See also: Behavioral Pharmacology; Brain and Behavior
Relationships; Catecholamines and Behavior; Depression; Drugs, the
Brain, and Behavior; Evolutionary Developmental Psychology;
Evolutionary Psychology; Hallucinations; Suicide.
Further Reading
Araneda S, Gamrani H, Font C, Calas A, Pujol JF, and Bobillier P (1980) Retrograde
axonal transport following injection of [3H]-serotonin into the olfactory bulb II.
Radioautographic study. Brain Research 196: 417-427.
Azmitia EC (2001) Modern views on an ancient chemical: Serotonin effects on
cell proliferation, maturation, and apoptosis. Brain Research Bulletin
56: 413-424.
Azmitia EC (2007) Cajal and brain plasticity: Insights relevant to emerging concepts
of mind. Brain Research Reviews 55(2): 395-405.
Azmitia EC (2010) The evolution of the serotonin system. In: Muller C and Jacobs B
(eds.) Handbook of the Behavioral Neurobiology of Serotonin, pp. 3-22, Chapter 1
Amsterdam: Elsevier.
Beaudet A and Descarries L (1978) The monoamine innervation of rat cerebral cortex:
Synaptic and nonsynaptic axon terminals. Neuroscience 3(10): 851-860.
Brodie BB and Shore PA (1 957) A concept for a role of serotonin and norepinephrine as
chemical mediators in the brain. Annals of the New York Academy of Sciences
66: 631-642.
Cannon WB (1929) Organization for physiological homeostasis. Physiological Reviews
9: 399-431.
Fernando H, Papadantonakis GA, Kim NS, and Lebreton PR (1998)
Conduction-band-edge ionization thresholds of DNA components in aqueous
solution. Proceedings of the National Academy of Sciences USA 95: 5550-5555.
Hess WR (1949) Nobel Lecture. Available at nobelprize.org/nobel_prizes/medicine/
laureates/1 949/hess-lectu re.html .
Brain Chemicals: Global Projections of Ancient Aromatic Neurotransmitters 405
Hofmann A (1 983) LSD My Problem Child: Reflections on Sacred Drugs, Mysticism
and Science. New York: McGraw-Hill.
Iversen L, Iversen S, Bloom FE, and Roth RH (2008) Introduction to
Neuropsychopharmacology. USA: Oxford University Press.
Pauling Linus and Wheland GW (1933) The nature of the chemical bond. V. The
quantum-mechanical calculation of the resonance energy of benzene and
naphthalene and the hydrocarbon free radicals. Journal of Chemical Physics
1: 362-374.
Peroutka SJ and Howell TA (1994) The molecular evolution of G protein-coupled
receptors: Focus on 5- hyd roxytry ptam i ne receptors. Neuropharmacology
33: 319-324.
Protacio CM, Dai YR, Lewis EF, and Flores HE (1992) Growth stimulation by
catecholamines In plant tissue. Annals of the New York Academy of Sciences
66: 631.
Russo S, Kema IP, Bosker F, Haavik J, and Korf J (2009) Tryptophan as an
evolutionarily conserved signal to brain serotonin: Molecular evidence and
psychiatric Implications. World Journal of Biological Psychiatry
10(4): 258-268.
Shulgin A (1991) Pihkal: A Chemical Love Story. Berkeley, CA: Transform Press.
Smiley JF (1996) Monoamines and acetylcholine in primate cerebral cortex: What
anatomy tells us about function. Revista BrasileiradeBiologia 56(Supplement 1 , Pt 1):
153-164.
Brainwashing and Totalitarian Influence
S M Andersen, New York University, New York, NY, USA
S ASaribay, Bogazici University, Bebek, Istanbul, Turkey
© 2012 Elsevier Inc. All rights reserved.
Glossary
Attitudes The evaluations, whether positive or negative, of
social objects or persons, as linked to specific beliefs and
ideas, and an alteration in this response based on exposure
to new information or social cues.
Automatic thought Thought and association that occurs
with little investment of mental effort, attention, and
without the need for conscious control.
Conscious thought It is the processing of information in
awareness versus without it.
Cult A group, usually of a religious or political nature, that
tends to be all-consuming for members and is highly
controlling of members' thoughts and actions, even if
participation is voluntary.
Dissociative states Atypical states of consciousness with
little or no self-awareness of one's own ordinary identity and
that often accompany behavior that is atypical for the
individual.
Heuristic processing This means responding to new
information by taking recourse to simple, common rules
(e.g., likeability = credibility) rather than engaging in careful
systematic processing.
Hypersuggestibility It is the unusual suggestibility in one's
own beliefs and actions to influence by the beliefs, actions,
or commands of another.
Identity It is the person's own sense of self as an individual,
and also in terms of his or her important relationships
(relational identity) and identification with groups
(social identity).
New religious movements These are new groups
reflecting a nonconventional approach to religion and
spirituality, not necessarily highly controlling and thus
not precisely cults.
Systematic processing It is the conscious evaluation,
integration, and weighing of new information.
Brainwashing can be defined, in succinct terms, as a process by
which an individual's values, beliefs, and behaviors, usually of a
religious or political nature, are systematically, intentionally,
and forcefully compromised and replaced with those deter-
mined by another person by deceptive, manipulative, and coer-
cive means. The term is most likely to be evoked when the
brainwashing agent has the intention of totally controlling the
other person, inducing him or her to engage in a sequence of
actions at variance with the individual's prior beliefs, particu-
larly in a relatively short time frame and with little awareness on
the part of the person or preexisting familiarity with the change
agent. The aim of brainwashing is to break down the indivi-
dual's capacity for critical thought, systematic decision-making,
and self-reflection involving long-held values, beliefs, prefer-
ences, and norms. At the same time, brainwashing remains
sufficiently ill-defined that its existence is open to question.
The term brainwashing was first used by journalist and US
intelligence agent Edward Hunter, who derived it from the
Chinese term > 7*0 Ugj (xl nao), and popularized it through news
articles and a book in the early 1950s. The term was purport-
edly used by a Chinese citizen, whom Hunter interviewed, to
refer to reeducation experiences in Communist China, and
although it was brought to the attention of Western society in
a negative light, as a kind of mental rape of the person, adher-
ents of Chinese Communism apparently viewed it in a more
positive light, as a 'morally uplifting' process by which the
person is freed from the grip of imperialist ideology.
Uses of the Term
Brainwashing has been applied to influence exerted by both
small and large groups of people, including religious 'cults' and
political organizations, for example, Jim Jones' Peoples Temple
movement in Jonestown where the mass suicide/massacre
occurred in 1978, as well as one-to-one dyads, such as the
case of Lee Boyd Malvo, the teenage sniper in the Washington,
DC, suburbs in 2002. It has been termed a social weapon aimed
at nonmainstream movements, such as religious 'cults,' or new
religious movements. Legally, the concept has been evoked to
defend the crimes committed by members of cults (or other
kinds of groups) that allegedly employ brainwashing, as was
presumed in the case of the members of Aum Shinrikyo, a
Japanese group responsible for the 1995 Tokyo subway gas
attacks.
Brainwashing is a concept commonly evoked by the media
and the lay public to fill in gaps in their understanding of
extreme experiences and transformations of individuals they
care about as intimates, community members, or well-known
individuals, including someone like Patty Hearst, the young
American heiress kidnapped by the Symbionese Liberation
Army in the 1970s who then confounded the public and police
by proceeding to actively participate in a bank robbery with her
captors. The term has been raised to try to explain any number
of unconventional or hard-to-comprehend acts perpetrated by
members of a particular group (e.g., by cult members), and as a
label, it creates an illusion of understanding and delineates
perpetrators relative to victims.
The concept has also been evoked to justify kidnapping and
deprogramming of cult members who are loved ones and are
presumed to have succumbed to untoward influence, and then
this process is carried out with the aim of erasing the effects of
the putative brainwashing and bringing the individual back
into alignment with mainstream values, itself potentially a
kind of brainwashing.
406
Brainwashing and Totalitarian Influence 407
The Mythic Definition: Induced by Exotic Techniques
The most exotic techniques that laypeople assume are in-
volved in brainwashing, whether severe exposure to drugs,
sensory deprivation, or hypnosis, are often presumed to have
the power to transform the human being into a programmed
machine that enters dissociative states when cued by physical
triggers (e.g., a code word) and will follow directives that
would otherwise be refused based on individual identity and
morality, and with no memory of doing so, as a kind of
Manchurian Candidate. Although controversial at best, this
definition of brainwashing involves exotic techniques that
were purportedly developed, used, and researched by the
erstwhile Soviet Union (e.g., revealed in the Moscow show
trials of the 1930s), by German Nazi intelligence agencies in
the Second World War, by China during the Communist
Revolution and Korean War, by the Central Intelligence
Agency (CIA) in research projects (e.g., MKULTRA) con-
ducted after the Second World War, and by some religious
or political cults. Since such activities tend to occur in secret,
behavioral science researchers are generally not in a position
to judge the veracity and comprehensiveness of the available
accounts.
A distinguishing aspect of this definition of brainwashing
seems to be its complete lack of regard for ethics, human rights,
and the personhood of the individual, as evidenced by the
extreme and sensational nature of the techniques used and
their presumed results. The hallmarks are the complete control
of thought and behavior that is sought, the direct harm to the
recipient and to others, and the brainwasher's intent to use
targeted individuals to engage in violent acts. The most vivid
feature of such brainwashing is the intent of the brainwasher to
systematically and persistently manipulate, through any avail-
able means, the individual's attitudes and behaviors to align
them with the brainwasher's vision. Beyond changing the sub-
ject's attitudes and behaviors and achieving complete control
over the individual, the goal can also be extracting secret infor-
mation from a captive or creating a deployable agent who will
carry out pro-organization tasks (e.g., a suicide attack) regard-
less of personal costs or moral or legal concerns. The results of
the search for methods of brainwashing, defined as such, have
at best been mixed.
These oft-discussed techniques include
• Pharmacological substances such as hallucinogens
• Hypnosis to produce dissociative states
• Physical exhaustion, sleep deprivation, heavy labor, with-
holding food and water, etc.
• Abuse and torture in the form of sensory deprivation, sensory
overstimulation, extended confinement, spinning, super-
sonic vibrations, electroshock, lobotomy, extreme tempera-
tures, cold-heat alternation, near-death experiences, oxygen
deprivation, gasses, etc.
• Psychic driving (i.e., exposure to looped audio messages
during drug-induced coma)
There is little doubt that efforts have been made to brain-
wash specific individuals using such procedures and to study
the consequences of it. Existing evidence, however, does not
support the claim that these efforts uncovered a set of exotic
techniques that will render a person sufficiently robotic that he
or she can be programmed to carry out an elaborate sequence
of actions without awareness, upon command, that he or she
otherwise would not. This version of brainwashing, then, is
essentially a myth. Such interventions in fact are better known
for disorienting, disabling, or killing the person rather than
effectively programming him or her to do one's bidding.
The Social-Psychological Definition: Coercive
Persuasion Through Mundane Social Influences
Exercise of influence over the mental processes and behaviors
of other individuals may be more profound and controlling
when based on mundane social-psychological processes com-
mon in daily life (such as those typically employed by legiti-
mate socializing agents like parents, the state, etc., and others in
relevant social roles such as salespersons) than when reliant on
exotic techniques. Although there is no consensually used term
for this social-psychological understanding of brainwashing as
contrasted with the mythic sense of this concept, the terms,
coercive persuasion and mind control, are sometimes used to
denote well-understood social influence processes that have
been heavily researched in social psychology, as noted by
Andersen and Zimbardo. The term coercive persuasion is the
scientific term most commonly in use and we restrict ourselves
to that term here. This conceptualization of coercive persuasion
builds on classical work in social psychology, such as that of
Asch, Sherif, Festinger, Milgram, and Zimbardo, whose research
on conformity, compliance, obedience, the power of social
roles, and cognitive dissonance is well known. Contemporary
scientific research on social influence, persuasion, and attitude
change is also central. For instance, Cialdini has identified
numerous 'weapons' of social influence that can be considered
the core aspects of coercive persuasion when used in combina-
tion and to an exaggerated extent, including use of a persuader
who is an authority and is likeable, playing on norms of reci-
procity, focus on the scarcity of the desired outcome, use of the
social proof that 'everyone else is doing it, including those you
trust,' and fostering a relevant commitment in order to tap
people's need to respond consistently. These techniques are
crucial in social influence and thus in coercive persuasion.
Whether this is an appropriate replacement for the other-
wise mythic concept of brainwashing depends on the perceived
goals of the brainwasher. For instance, Solomon has argued
that the foundational techniques used by the Chinese should
be distinguished from those used by the Russians because the
former sought conversion, not mere compliance. In this view,
the Chinese model may be better understood from a social-
psychological perspective than from the perspective of the
mythic brainwashing concept.
Social influence stems from the desire of a social system
(e.g., a religious organization) to propagate itself, socializing
recruits into its network of values and behavioral norms via
systematic use of a combination of ordinary social influence
tactics. An individual may choose to subject himself or herself
to such tactics when he or she expects benefits such as in
purchasing a used car or joining like-minded others for spiri-
tual guidance and growth. However, when a social system, for
various reasons (e.g., high ambitions and incentives for
growth), comes to exert an unacceptable level of influence
and control through the systematic and combined use of
408 Brainwashing and Totalitarian Influence
nonexotic social-psychological influence techniques at exag-
gerated levels, this may be tantamount to coercive persuasion.
Such influence techniques may be viewed as safe and legitimate
under normal conditions, such as when used separately and at
milder levels. Such coercive persuasion may or may not involve
untoward, malignant intent or complete disregard for the rights
of the target individual, unlike the more dramatic definition or
version of brainwashing, which is virtually always considered to
involve those factors and leads to no positive outcome for the
individual.
While severe distress and duress would be a core aspect of
brainwashing, coercive persuasion can proceed more subtly,
relying exclusively on the psychological tactics. Through
more subtle means of coercive persuasion, there is the poten-
tial to socialize the person into the organization in a way
that leaves largely intact the individual's psychological integ-
rity and his or her decision-making abilities, even while
significantly limiting his or her personal freedom of choice
and action.
Identifying Coercive Persuasion
When attempting to identify agents of coercive persuasion,
a persistent problem is overcategorization of deviance from
mainstream culture, religion, ethicality as involving unto-
ward influence. When there is demonstrable harm to the
individual or to society, however, the question of whether
or not there is an intolerable control exerted over individual
freedom, based, for instance, on the joint use of social-
psychological techniques noted below, this can be a viable
alternative. Coercive persuasion arises from a strong and
uniform authority structure, and the elicitation of complete
commitment by members, creating the potential for exploi-
tation of individuals. When coercion and deception are also
involved, the totalitarian aspects of the organization become
increasingly evident. This may also be considered cult-like,
though the use of this term, like brainwashing, is also
fraught. Such an organization can be religious or political
or may involve a dyadic familial structure, as with battered
spouses.
Mundane influences on persuasion and compliance
• The illusion of common ground, that is, perceived similar-
ity fostering the belief that the persuader (and other group
members) shares the individual's values, needs, goals, etc.,
whether or not this is tme
• Fostering a sense of belonging, acceptance, and relation-
ship, appealing to the inner pressure to bond with others
• The illusion of free choice through offers to the individual
of only prosystem choices, while legitimate options that do
not fit the group's goals are omitted
• Limiting the information available (external sources of
information), even as the individual is asked to choose
freely
• Inducing complexity, uncertainty, and confusion via jar-
gon, euphemisms, false analogies, rhetorical labels, seman-
tic distortions, misleading abstractions, and so on to
capitalize on the natural tendency to seek simple answers
(e.g., from a strong leader or cohesive group, or simply for
convenience), to engage in heuristic processing rather than
reasoning critically
• Playing on norms about reciprocity (i.e., to reciprocate
disclosures or gestures of another)
• Use of contingent social reward (e.g., love, support, emo-
tional comforting) that depends on expressed beliefs and
actions reflecting the preferred worldview or ideology
• Assigning the individual to one or more roles, ranks, or
positions to bestow significance and guide behavior, possi-
bly capitalizing on the tendency to take on a complemen-
tary social role (e.g., to become strong or the helper when
somebody else is weak)
• Use of a step-by-step indoctrination strategy to render
recruits unaware of the intended end results of the recruit-
ment process
• Emotions, especially negative ones, like fear or guilt,
induced, and then resolved, as when an insurance salesper-
son first makes the customer feel fearful and then offers the
solution of an insurance policy
• Use of artificial pressures to create group unanimity (e.g.,
censoring, intolerance of diversity, discouraging questions),
potentially resulting in groupthink, and poor decisions
under the guise of consensus
• Restriction of communication and prohibition of certain
linguistic expressions, in favor of a coded language, labels,
or jargon
• Evoking reactance, as when a behavioral choice is presented
as scarce and limited to only a select few, in order to induce
the person to become more intrigued by it, desiring it more
Psychological coercion
In some combinations, social-psychological influence strate-
gies can approach psychological coercion, if used systemati-
cally by the persuader, and hence, these are worthy of note as
particularly controlling:
• Isolation of the individual from his or her usual social
network and prior sources of social support
• Stripping the individual from familiar environments, sym-
bols, and possessions that remind the individual of his or
her identity
• Systematic induction of relatively intense psychological
duress, whose simplest resolution is a further commitment
to the group
• Division of the world into the 'good us' and the 'evil them,'
and the dehumanization and possibly demonization of the
other using facile, derogatory labels for outsiders
• Elicitation of confessions, potentially to obtain knowledge
of the individual's deepest vulnerabilities, which can then
be abused, possibly for blackmail
• Overstimulation by prosystem information (e.g., interactions
or rehearsals with members, proselytizing activities, etc.)
• Exhaustion through extended labor and sleep deprivation
• Lack of or elimination of privacy
• Threat of sanctions for failure to adhere to the group view,
whether subtle or overt
• Threat of psychological or spiritual punishment (e.g., to
'burn in hell')
• Threat of physical harm
Brainwashing and Totalitarian Influence 409
The relevance of deception
The use of outright deception may also be present and exacer-
bate psychological coercion, making it most insidious. This can
include
• Providing inaccurate information about the group in the
form of overt lies about the group (e.g., via a front organi-
zation) or other misleading identification about the group
presented to outsiders, for example, to potential recruits,
family members of existing recruits, etc.
• Omitting crucial information about the group and its aims
and activities.
Other factors
The groups or organizations that grab hold of an individual
and exert special control over him or her may also enable
persuasion by other factors, including
• The presence of a charismatic leader or guru, and referring
to him or her as God or God-like
• The fostered belief that there is only one true authority (e.g.,
the group leader)
• Special appeals to individuals not anchored in social net-
works, who are alienated from society, and/or who feel
excluded
• Love bombing (showering with love and praise).
Dyadic Coercive Persuasion
Coercive persuasion is not limited to groups. Spousal battery
in marital relationships, for example, may make use of forms
of coercive persuasion, by involving (1) verbal and/or physical
dominance; (2) isolation or imprisonment; (3) fear arousal
and maintenance; (4) guilt induction; (5) contingent expres-
sions of 'love'; (6) enforced loyalty to the aggressor and
self-denunciation; (7) promotion of powerlessness and help-
lessness; (8) pathological expressions of jealousy; (9) hope-
instilling behaviors (that somehow the battering will end
if. . .); and (10) required secrecy.
Coercive Persuasion in Organizations Involved in Terrorism
While coercive persuasion has often been discussed in the con-
text of religious cults, and to a lesser extent, spousal abuse, the
potential application of these ideas to terrorism is of special
contemporary interest. In recent years, scholars have paid
increasing attention to terrorism, especially in the Islamic
world, following events such as 9/11 and the bombings in
Madrid, London, and elsewhere. What is known about coercive
persuasion has potential applicability to recruitment and train-
ing of individuals by extremist organizations, luring young peo-
ple and readying them to conduct suicide attacks. Kruglanski and
colleagues have argued that personal qualities (e.g., trauma,
social exclusion, alienation, etc.) may heighten individuals'
quest for personal significance and meaning. In the context of
high levels of intergroup conflict and threat to the ingroup,
terrorism-justifying ideologies may provide the means to restore
significance and overcome death anxiety that stems from earlier
traumas and lacking of a sense of belonging. Consistent with the
foundational work of Lifton, Schein, and others, it is this
particular constellation of personal and social factors that may
create vulnerability in the individual to the indoctrination tech-
niques of terrorist organizations. Terrorist organizations are
likely to make use of the same social-psychological influence
strategies as other kinds of organizations do (see above) and to
deliberately engineer their messages to fulfill the significant
quest for potential recruits, seeking out individuals from the
backgrounds noted (e.g., traumatized, alienated individuals).
Like some religious groups, terrorist organizations may be led
by a charismatic and megalomaniacal leader aiming to under-
take spectacular acts.
Coercive Persuasion versus Ordinary Social Influence
The aforementioned techniques, used in moderation, may be
common, if not ubiquitous, in modern society. However,
when many social influence strategies are combined in extreme
form and used systematically, this level of influence may con-
stitute a qualitative difference that demarcates what is an
acceptable level of relatively nonmanipulative social influence.
This is akin to a mild versus severe exposure model, as in the
example of radiation in which mild exposure can be tolerated
well by the person (the body), whereas exposure from multiple
sources over extended periods or in the extreme may result
in cancer.
One reason that a combination of techniques results in a
qualitatively different kind of social influence is their effect on
attentional capacity. Robert Baron has argued, on the basis of
social-psychological research, that the stress and emotional
arousal that is a part of intense indoctrination experiences
leads to significant decreases in attentional capacity, which in
turn makes subjects highly vulnerable to social influence tac-
tics. Research shows that factors that lessen attentional capacity
increase (a) compliance, (b) heuristic (vs. systematic) message
processing, (c) conformity, (d) stereotypical thinking, and
(e) cognitive dissonance effects. For instance, individuals
under intense indoctrination, even when they are motivated
to do otherwise, should be less able (have less capacity) to
process messages systematically (in the sense of Chaiken's
attitude theory) and should be more influenced by heuristic
qualities of persuasion and cues associated with heuristic pro-
cessing. Minimally, the lure of engaging in relatively little
thought (processing information automatically) without sys-
tematic deliberation is quite considerable in most situations,
and further enabling this is not particularly difficult. Indivi-
duals under intense indoctrination should be more influenced
by heuristic qualities in a persuasion context, such as the
responses of others around them (who may be well-socialized
members who have been trained to give a uniform response)
and how confident or likeable the persuader appears, rather
than critically scrutinizing the substance of the information
provided. This kind of heuristic processing can lead to attitude
change based on little conscious thought or deliberation,
which is solidified later, for example through continued social-
ization into the group and reinforcing activities and discus-
sions that enable the new information to be incorporated into
aspects of one's own identity. Increased stereotypical thinking
among indoctrinees suggests that individuals are easily led to
endorse a stark 'us versus them' distinction and to have a
fostered paranoia about outsiders and the outside world,
410 Brainwashing and Totalitarian Influence
which the stereotyping measure may tap. Intense indoctrina-
tion situations increase conformity and reliance on authority,
as manifested perhaps in the Stockholm Syndrome in which
a captive becomes sympathetic to and collaborates with his or
her captors.
Theories of Brainwashing
Foundational Work
The foundational research of Robert J. Lifton and Edgar Schein
on Korean War prisoners of war (POWs) (and also those of
others in the same era, such as Hinkle and Wolff) reached the
conclusion that the Communists had no especially influential
techniques at their disposal, and whoever among the POWs
converted, defected, or refused repatriation (22 out of about
4500 American soldiers; none of the Turkish soldiers) under
their influence did so either out of emotional and physical
needs or due to prior ideological leanings. For instance, captors
placed POWs under harsh conditions and rewarded them for
their pro-Communist behavior by improving their conditions,
such as giving them better food, blankets, and a more comfort-
able sleeping arrangement. While the behavior of some POWs
changed to align with their captors' desire, for most of them,
cognitive (e.g., beliefs, values) changes did not accompany.
None of Lifton's and Schein's subjects actually converted to
Communism, though a small number can be said to be influ-
enced in the direction of their captors. It must be noted, however,
that foundational research has been criticized for its small sam-
ples (e.g., N = 40 in Lifton's study). This foundational research
attributed such influence (and conversion to Communist ideol-
ogy, when it occurred) among POWs to personality characteris-
tics that existed prior to the exposure to Communist messages
and not to induced dissociation or to other more exotic factors.
Powerful social influence strategies are thus not particularly
exotic, but rather, mundane in their influence on behavior and
central to a social-psychological conceptualization of coercive
persuasion. While early research did suggest a variety of poten-
tially influential techniques, it also refuted an initial allegation
by Hunter that a person's belief system can be completely
replaced against his or her will by powerful, external influ-
ences. For this reason, both of the foundational researchers
stayed away from the term brainwashing and instead used
the term coercive persuasion (Schein) and the terms thought
reform and totalism (Lifton). Despite this, the term brainwash-
ing stuck in the public consciousness and continued to fuel
paranoia about the Eastern bloc in the Western hemisphere, in
the United States during the Cold War.
Foundational researchers like Schein also viewed thought
reform or coercive persuasion in terms of stages such as
unfreezing, changing, and refreezing (see also Kruglanski's
work). Lifton in particular asserted eight stages of the process
of thought reform and ideological totalism beginning with
assault on identity and ending with final confession and
rebirth. In particular, these proposed stages track well with
techniques identified as influential earlier, and involve the
following:
• Milieu control: Control of information and also internal
dialog of the subject
• Mystical manipulation: Covertly orchestrating events while
simultaneously making them appear spontaneous to argue
for the special powers or divine status of the organization/
leader
• Demand for purity: Imposing a dualistic view of the world
in terms of us/them, good/evil, saved/damned, etc., and
arguing that outgroups should be eliminated or otherwise
prevented from contaminating the ingroup
• Cult of confession: Use of public confessions to destroy
privacy and exploit the subject
• Sacred science: Imposing a view of organizational ideology
as morally unchallengeable and scientifically exact
• Loading the language: Use of simple language (thought-
terminating cliches) to reduce complexities regarding orga-
nizational ideology, thereby discouraging critical thinking
and creating the illusion of group unanimity
• The primacy of doctrine over person: The view that nothing
the person experiences can have more importance than,
and potential to discredit, organizational ideology
• Dispensing of existence: The view that the organization has
the right to determine the fates of its members and also its
opponents.
Cultic Brainwashing Theory
A number of scholars, such as Singer and Ofshe (in the United
States) and Abgrall (in Europe), have characterized brainwash-
ing in the context of conversion to cults (and in related legal
cases), and have referred to their approach as 'cultic brainwash-
ing theory.' Owing to its tacit assumptions about personhood
and free will, it has also been referred to as Robot Theory or the
Involuntarist Theory of Brainwashing. While it references
the foundational work of Lifton and Schein, its formulation
is closer to the mythic definition of brainwashing than to the
social-psychological definition. The gist of its understanding
of brainwashing is that the unsolicited use of a set of techni-
ques by cults on recruits can put the latter into disoriented
or dissociative states of consciousness in which their capacity
to make rational choices for themselves is vastly diminished
or destroyed, and hypersuggestibility reigns that leads to the
uncritical adoption of any propaganda imposed on them. The
victim then may become addicted to the new set of beliefs
programmed into his or her mind and thus may not be able
to rationally contradict these beliefs unless deprogrammed by
experts. Such brainwashing is assumed to be irresistible and
potentially irreversible, and this is the approach that has been
central in the anticult literature and movement, although
empirical evidence for it is lacking and it has been roundly
criticized.
Primitive states of consciousness (e.g., the dissociative, the
hypnotic) were not central to or a part of Communist indoctri-
nation and thus, the foundational researchers accorded no
special role to them. The foundational researchers also recog-
nized more complexity in human behavior and thus offered
a 'soft determinism' rather than an exotic model in which
human beings can be transformed into tools or robots ('hard
determinism') by means of hypnosis. As criticism of the cultic
brainwashing theory, scholars like Robbins and Richardson
have argued that there is little evidence of any special illegality
of religious 'cults' in terms of unacceptable means or levels of
Brainwashing and Totalitarian Influence 41 1
influence (e.g., physical coercion) or other illegality, that these
are protected under the First Amendment, and that the theory
ignores findings showing that participation in some unconven-
tional religious movements is beneficial to the well-being of
recruits. These authors also argue that the theory overempha-
sizes and makes overly broad claims about hypnosis and its
influence in the light of research that has discredited the notion
that hypnosis can be used to obtain complete control over an
individual's behavior, that it underemphasizes predisposing
characteristics of religious converts, and that the recruitment
process is typically entirely volitional (voluntary). Indeed,
Richardson argues that parents, fraternities and sororities,
marine boot camp, and juvenile detention halls - all use
isolation for attitude and behavior change, so even the more
dramatic strategies are not unprecedented. Use of such strate-
gies, as in preventing a recruit from being exposed to dissen-
ters, should thus be treated as no more problematic than a car
salesman preventing a customer from talking to someone who
previously refused the same offer, the great deal. Some argue
then that cubic brainwashing theory is not a proper scientific
theory in that it is ambiguous, internally contradictory, and
not falsifiable.
Other scholars have sought to study brainwashing from a
more scientific perspective. For example, Zablocki has argued
that both those who claim that there is a pandemic of brain-
washing and those who claim that brainwashing does not exist
at all are both incorrect, and has attempted to construct a precise
and falsifiable definition of brainwashing, reviewing ethno-
graphic evidence on cults and interviews with ex-members and
leaders to support it. This approach has been strongly criticized
as well by anticultic brainwashing scholars, though Zablocki
argues that ignoring the situation of former members of cults
who were psychologically and/or physically harmed prioritizes
legal proceedings and concerns about religious liberty over
scientific study.
Activist View of Conversion: Eschewing the
Brainwashing Concept
Based on such criticism, Richardson offered a view of the
recruitment and conversion as an interactive negotiation, draw-
ing heavily on Moreland and Levine's work on socialization in
small groups. Other scholars, mostly sociologists of religion
(vs. mental health professionals), support such a view. The
main contrast between passivist views (such as cubic brain-
washing theory) and the more activist (subject-centered) view
of conversion is that the individual's capability to act autono-
mously and make choices to advance his or her own goals
is intact and not entirely compromised, though it may be
manipulated to whatever degree. This includes the active seek-
ing of recruitment and the ability to withdraw from the group.
The group is not seen as omnipotent. The group can also choose
to withdraw from the effort to recruit the individual or may kick
existing members out. The levels of commitment of individual
to group and group to individual fluctuate over time, and
recruits can exert minority influence on the group once they
are members. This offers a more balanced view of a greater
proportion of recruit-group interactions as may occur in non-
traditional religious movements that does not take recourse in
the brainwashing concept.
Difficulties with the Scientific Study of Brainwashing
There are a number of difficulties with the scientific study of
brainwashing whether defined in mythic or social-psychological
terms. The concept has suffered from loose definition as well as
distortion in the mass media, in popular culture, and in politi-
cal propaganda. As a result, its definition is less coherent and
less widely agreed upon than is the case for other scientific
constructs, and its scientific status is debatable. Some scholars
have thus suggested that the concept be abandoned. Concepts
at the heart of the brainwashing debate are themselves contro-
versial, such as the nature of free will, consciousness, dissocia-
tive states, and the definition of pathological. Relatively few
solid theoretical advances have emerged since the so-called
foundational theories of brainwashing (such as Lifton's and
Schein's) were developed, and these foundational theories
themselves are often misinterpreted or presented in misleading
ways. As a result of court cases making reference to brainwash-
ing and in which a binary decision needs to be reached, scien-
tific attempts to answer questions about brainwashing have
also taken on an either/or flavor (e.g., Does it exist or not? Is
it effective or not? Do religious movements use it or not?),
whereas the answer to most pertinent questions along these
lines is likely to lie in the gray area as with every other reason-
able scientific question.
Because brainwashing cannot be created in the laboratory,
for obvious ethical reasons at minimum, direct scientific evi-
dence pertaining to it is scant. There remains no consensus on
how to determine whether or not a person has been brain-
washed, making the concept highly problematic to assess. For
example, determining through interviews whether or not an
individual who joined a movement did so by consent will
be highly problematic after the individual is already a well-
socialized member of the group. Indeed, some scholars have
argued that the more successful the brainwashing is in any
given case, the fewer the symptoms the person should show,
making assessment all but impossible. Further, if brainwashing
is thought to be psychologically damaging, those who were
allegedly brainwashed may be seen as lacking psychological
integrity and credibility. And, even if not psychologically com-
promised, a person who has had negative experiences in a
religious group and thus exits the group may be motivated to
go against it or even to take revenge on it, thus making evalua-
tion of the veracity of their accounts highly problematic. Sound
evidence on the matter thus clearly calls for longitudinal (pre-
post) data collection, and yet it remains unclear how research-
ers should know in advance whom to follow based on the
individual being likely to join such a movement.
Scholars interested in the brainwashing concept have
tended to focus on movements that have clearly harmed
individuals, such as Aum Shinrikyo, People's Temple, Hea-
ven's Gate, Solar Temple, and Charles Manson's group,
whereas those opposing the use of the brainwashing concept
have tended to focus instead on movements that obviously
contribute positive value and innovation to society. The diffi-
culty in establishing a scientific consensus on the brainwash-
ing concept is evidenced by the rejection, in 1987, of the
report of the american psychological association's (APA)
Task Force on deceptive and indirect techniques of persuasion
and control (DIMPAC) by APA's board of social and ethical
412 Brainwashing and Totalitarian Influence
responsibility for psychology (BSERP) on the basis that it did
not provide sufficiently rigorous and evenhanded analysis of
the issue.
Conclusion
Brainwashing is a sufficiently ill-defined concept that it may
not be scientifically useful. There is a lack of evidence for it in
its mythic form which is presumed to rely on exotic techni-
ques and to lead surreptitiously to total control. Even half a
century ago, the foundational researchers in this area
eschewed the term brainwashing for good reason. Instead,
the most fruitful approach has been to examine mundane
but powerful social influence processes, which may, under
some circumstances, take an extreme form in a particular
social group or organization, whether religious, political, or
otherwise, or even in a dyadic relationship. An alternative
term sometimes used, rather than brainwashing, is coercive
persuasion or thought reform, with grounding in basic social
influence processes and factors that have an impact on human
thought processes and persuasion, and the operation of which
is well understood on the basis of the science in social
psychology.
See also: Anxiety and Fear; Attitude Change; Attitude Formation;
Child Abuse; Cognitive Dissonance Theory; Confabulation and
Reality Filtering; Conformity and Obedience; Crowding: Effects on
Health and Behavior; Delusions; Dissociative Disorders; Drugs, the
Brain, and Behavior; Eyewitness Identification; Interpersonal
Perception and Communication; Jury Psychology; Perceived
Control; Psychology and Religion; Psychology, Science, and
Astrology; Risk-Taking Behavior (Young Male Syndrome); Sense of
Taste (Effect on Behavior); Terrorism; Violence; War.
Further Reading
Andersen SM (1985) Identifying coercion and deception in social systems. In:
Kilbourne B (ed.) Scientific Research and New Religions: Divergent Perspectives,
pp. 12-23. Washington, DC: American Association for the Advancement of Science.
Anthony D (1999) Pseudoscience and minority religions: An evaluation of the
brainwashing theories of Jean-Marie Abgrall. Social Justice Research 12: 421-457.
Baron RS (2000) Arousal, capacity, and intense indoctrination. Personality and Social
Psychology Review 4: 238-254.
Boulette TR and Andersen SM (1985) “Mind Control" and the battering of women.
Community Mental Health Journal 21: 109-118.
Cialdini RB (2001) Influence: Science and Practice, 4th edn. Boston, MA: Allyn & Bacon.
Hood RW, Hill PC, and Spilka B (2009) Psychology of Religion, 4th edn. New York:
Guilford Press.
kilbourne B and Richardson JT (1988) Paradigm conflict, types of conversion, and
conversion theories. Sociological Analysis 50: 1-21.
Litton RJ (1 961 ) Thought Reform and the Psychology of Totalism: A Study of
‘Brainwashing in China. New York: W. W. Norton.
Richardson JT (1993) A social-psychological critique of ‘brainwashing’ claims about
recruitment to new religions. In: Hadden J and Bromley D (eds.) The Handbook of
Cults and Sects in America, pp. 75-97. Greenwich, CT: JAI Press.
Schein EH, Schneier I, and Barker CH (1961) Coercive Persuasion. New York:
W. W. Norton.
Singer MT (1 995) Cults in Our Midst: The Hidden Menace in Our Everyday Lives.
San Francisco, CA: Jossey-Bass.
Solomon T (1983) Programming and deprogramming the ’Moonies’: Social psychology
applied. In: Bromley D and Richardson J (eds.) The Brainwashing/Deprogramming
Controversy. New York: Edwin Mellen.
Streatfeild D (2008) Brainwash: The Secret History of Mind Control. New York: Picador.
Taylor KE (2004) Brainwashing: The Science of Thought Control. Oxford, UK: Oxford
University Press.
Zablocki Band RobbinsT (2001) Misunderstanding Cults: Searching for Objectivity in a
Controversial Field. Toronto, ON: University of Toronto Press.
Zimbardo PG and Andersen SM (1993) Understanding mind control: Mundane and
exotic mental manipulations. In: Langone M (ed.) Recovery from Cults,
pp. 104-125. New York: W. W. Norton.
Relevant Websites
http://www.icsahome.com/ - International Cultic Studies Association.
http://www.cesnur.org/ - Center for Studies on New Religions.
http://www.apologeticsindex.org/-Apologetics Research Resources on religious
movements, cults, sects, world religions and related issues.
Bulimia Nervosa
J E Mitchell, K Steffen, T Ertelt, and J Marino, University of North Dakota, Fargo, ND, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Binge eating Consuming a large amount of food in a
short period of time accompanied by a sense of loss of
control.
DSM-IV The 4th edition of the Diagnostic and Statistical
Manual of the American Psychiatric Association.
Eating disorders not otherwise specified (EDNOS) An
eating disorder that is clinically significant but does not meet
criteria for anorexia nervosa or bulimia nervosa.
Incidence New cases per period of time.
Pharmacotherapy Drug treatment.
Prevalence Total cases at any given time.
Introduction
Bulimia nervosa (BN) involves three behavioral and cognitive
components: binge eating, inappropriate compensatory mea-
sures, and body image disturbance. Binge eating is defined as
(1) eating a large amount of food in a discrete amount of time
which would be atypical of others in similar circumstances and
(2) a sense of loss of control during these eating episodes.
Binge-eating episodes are often followed by a negative emo-
tional state, and often occur in response to stressors or after a
period of fasting. The food consumed during binge-eating
episodes is usually high in calories and often has fat content,
for example, sweets. Research also suggests that the disorder
has a significant financial cost (i.e., $1599.45 per year in the
United States) associated with purchasing the food that
patients consume during their eating episodes.
Compensatory behaviors can involve multiple methods
including vomiting; laxative, diuretic, or enema misuse; fasting;
or overexercising. DSM-IV suggests that binge-eating and purg-
ing behaviors occur in combination and at least twice weekly
for 3 months. Body image disturbance is defined such that self-
evaluation is unduly influenced by body shape and weight.
These symptoms must not occur during the course of anorexia
nervosa (AN), which is marked by maintenance of low weight.
BN can be specified in two ways: purging and nonpurging.
Purging behavior includes vomiting, and misuse of laxative,
diuretics, or enemas. Nonpurging BN is diagnosed when the
patient only engages in fasting or overexercise.
BN includes ~25.5% of outpatient eating-disorder (ED)
diagnoses. Many ED patients tend to fall into the ED not
otherwise specified (EDNOS) category, as many patients'
symptoms do not meet the full criteria for BN. EDNOS is the
most common ED diagnosis, constituting nearly 60% of out-
patient ED cases. Given this large grouping which lacks speci-
fying detail, many clinicians and researchers are looking to the
DSM-V to define the diagnostic criteria for subgroups of
EDNOS. A future direction for the next edition of the DSM
may be to operationalize the amount of food needed to define
an 'objectively' large binge-eating episode. Some ED patients
report experiencing 'subjective' binge-eating episodes, wherein
the amount of food consumed is not necessarily considered
larger than what others would eat (e.g., two cookies); however,
the patient reports a feeling of loss of control during the
consumption of the food. Further research is needed to better
understand the caloric differences between objective and sub-
jective binge-eating episodes. Additionally, recent attention
has been paid to 'purging disorder,' which is similar to BN
but is marked by compensatory behavior in the absence of
objectively large binge-eating episodes. Limited research has
explored this disorder, which is currently classified as a form of
EDNOS.
Clinical Characteristics
BN most often occurs in females (i.e., 90%), although a
research literature has developed on males and eating disor-
ders. EDs have generally been studied, and likely are most
prevalent in western cultures. Research has suggested, however,
that the cultural bias on thinness is impacting other regions of
the world as Western influences become more prevalent. Indi-
viduals suffering from BN are typically of normal weight.
Patients can experience comorbid substance abuse, especially
alcohol abuse, mood disturbance, anxiety disorders, and per-
sonality disorders.
BN often first appears during adolescence or in the early
20s. Patients tend to begin restricting food intake, which is
followed by episodes of overeating and subsequent purging.
These behaviors become cyclical. As mentioned, the binge-
purge cycle appears to become a means to cope with negative
emotional states. The most commonly used method of com-
pensating for binge eating is vomiting, which can occur some-
what effortlessly after recurrent vomiting. Oftentimes, the
pattern of purging can lead to medical complications such as
electrolyte imbalances, edema, and dental complications such
as erosion of dental enamel. Laxative abuse can engender
various gastrointestinal complications. Given the sometimes
chronic course of BN, ongoing medical complications can
occasionally become life threatening.
Epidemiology
The DSM-IV estimates the prevalence of BN to be between 1
and 3% for women and between 0.1 and 0.3% for men.
As numerous epidemiological studies have indicated, the prev-
alence and incidence of BN and its subthreshold variants are
considerably more complicated issues. There have been a
413
414 Bulimia Nervosa
number of significant difficulties and methodological limita-
tions inherent in studying the epidemiology of EDs, including
the relatively low prevalence of eating disorders in the overall
population and the general reluctance of individuals affected
with EDs to seek professional help. Considerations of this
nature must be taken into account when discussing epidemio-
logical issues in BN.
The incidence of BN has been estimated in several studies in
industrialized nations. In the United States, the incidence of
BN has been estimated at 13.5 affected individuals per 100 000
people for the period between 1980 and 1990. An incidence of
12.2 per 100 000 was observed in the United Kingdom in 1993
by Turnbull and colleagues.
Several prevalence studies of BN have also been conducted
in industrialized nations. Fairburn and Beglin in 1990 provided
the generally accepted prevalence of estimate of 1.0% when
taking into account BN as it was defined in DSM-I1I, DSM-III-R,
and Russel's original description in 1979. Kendler and collea-
gues in 1991 reported a lifetime prevalence of the narrow BN
phenotype at 2.8% in the United States. In 2006, a replication
of the National Comorbidity Survey reported in a paper by
Hudson and colleagues a lifetime prevalence of 1.5% for
women and 0.5% for men and a point prevalence of 0.5%
in women and 0.1% in men.
In addition, it is important to consider individuals affected
by subclinical variants of BN when discussing epidemiological
issues. The prevalence of subclinical BN is ~3-4 times that of
full-syndrome BN.
Longitudinal Course
The course of BN is highly variable. One-quarter of individuals
with BN who do not receive treatment still improve, while
upwards of 70% who are treated respond. Literature has sug-
gested that there may be considerable crossover among AN,
BN, and EDNOS. In fact, it is likely that many individuals
fluctuate between these disorders, and in the severity of their
psychopathology. Research generally has focused on full-
threshold disorders (e.g., AN and BN). However, it has been
suggested that AN and BN may in fact exist on a continuum
with EDNOS. Indeed, patients with AN often cross over to BN.
Although BN patients can also cross to AN, it is less common.
Periods of symptom remission are also possible. For example,
research suggests that nearly 30% of BN patients and 12-30%
of patients with EDNOS resembling BN or purging disorder
had a previous diagnosis of AN. Most women who transition to
AN from BN appear to meet criteria for AN-B/P, and fewer
transition from AN-R. It appears most likely that patients with
BN tend to fluctuate to partial remission or full resolution of
symptoms. These changes in diagnostic classification and in
symptom severity speak of the importance of continued moni-
toring of symptoms and assessment of treatment effectiveness
throughout the course.
Psychotherapy
Cognitive behavioral therapy (CBT) is generally considered the
treatment of choice for patients with BN. The treatment can be
delivered in either group or individual formats. Recently, the
NICE guidelines in the United Kingdom assigned the grade of
A for CBT for BN, attesting to the efficacy of this treatment.
Interpersonal therapy (IPT) has also been monitored in two
large studies, and although the effectiveness is substantial, the
therapy appears to take longer to have an effect. From a theo-
retical standpoint, this is quite interesting. IPT was developed
to be used as a maintenance treatment for depression and
does not include a focus on the specific symptoms of BN.
A third treatment, which has also been modified from its
original application for use with ED patients, is dialectic
behavior therapy (DBT). This therapy was originally given to
patients with self-injurious behavior and suicidality but has
subsequently also been shown to be effective in BN. While
CBT focuses on specific bulimic behaviors and IPT focuses on
interpersonal relationships, DBT focuses on both behavioral
change and self-acceptance. Several other treatments have
been suggested in the literature but have not been empirically
validated.
Unfortunately, the treatment of choice, CBT, often results
in low rates of complete remission, usually in the range of
30-40% in most studies. For this reason, increasing emphasis
is being placed on the development of new therapies. One
such therapy is an enhanced form of CBT (CBT-E) developed
by Fairburn, Cooper, and Shafran at Oxford University. This
therapy expands on the existing form of CBT to embrace four
additional domains that are assumed to be maintaining
mechanisms: clinical perfectionism, interpersonal difficulties,
low core self-esteem, and mood intolerance. This therapy is
designed to be transdiagnostic and may be used for patients
with any ED. Another new therapy developed at the Neuro-
psychiatric Research Institute and the University of Minnesota
also appears promising. This therapy, called Integrative Cogni-
tive-Affective Therapy (ICAT), expands on traditional CBT by
placing greater emphasis on cultural factors, self-oriented cog-
nitions, interpersonal schemas, interpersonal patterns, and
emotional experience.
CBT rests on the cognitive view of the maintenance of
BN, which assumes that low self-esteem generates extreme
concerns about weight and shape; this leads to strict dieting
which subsequently leads to binge eating and then to self-
induced vomiting. Each level of this cascade then feeds back
on the other levels. Traditional CBT is delivered in 20 sessions
over 16 weeks with twice-weekly appointments in the initial
month. There are generally three treatment phases. The first
8 sessions focus on explaining the rationale for CBT and repla-
cing binge-eating behavior with a stable eating pattern. The
second treatment phase, from sessions 9-16, continues to
focus on behavioral strategies but also focuses on cognitive
issues, including an examination of all forms of 'dieting,'
weight and shape issues and cognitive distortions, and includes
training in behavioral problem solving. The third phase, from
sessions 17-20, focuses on relapse prevention. There is a strong
emphasis on self-monitoring in this treatment.
Pharmacothearpy
The first report of pharmacotherapy as a treatment method
for BN appeared in 1983, 4years after Gerald Russell first
Bulimia Nervosa 415
described the condition. Over the last 30 years, several phar-
macological agents have been shown to have efficacy for BN,
beginning with early trials with tricyclic antidepressants (TCAs)
and continuing through more recent trials with the antiepilep-
tic drug topiramate. Despite the presence of positive efficacy
data with several classes of medications in BN, significant
room for improvement in the pharmacological treatment of
this disorder remains, as remission rates are low, often only
around 10-15%.
Typically, patients with BN fall within the normal weight
range and are likely to have at least one comorbid psychiatric
condition. For example, approximately half the patients with
BN have a lifetime diagnosis of major depressive disorder. In
addition to targeting the recurrent binge eating and compensa-
tory mechanisms that characterize this condition, pharmaco-
therapy trials have also investigated the impact of medication
on comorbid psychopathology and body weight.
A number of placebo-controlled trials of TCAs were pub-
lished in the 1980s and early 1990s, including imipramine,
desipramine, and amitriptyline. With a few exceptions, the
results of these trials demonstrated the ability of TCAs to
reduce the frequency of binge eating in patients with BN.
A secondary outcome measure in many of these studies was a
reduction in depression symptoms. The TCAs generally,
although not always, led to a significantly greater reduction
in depression symptoms relative to placebo.
Despite research showing overall favorable efficacy with the
TCAs, tolerability was not optimal. With varying degrees of
potency, the TCAs act as antagonists at muscarinic, a- 1 adren-
ergic, and histamine- 1 receptors. These pharmacological attri-
butes can generate adverse effects including anticholinergic
effects (e.g., dry mouth, urinary retention, and constipation),
onhostatic hypotension, sedation, and weight gain. These drugs
are also known to have serious or fatal consequences in over-
dose. Also beginning in the mid-1980s, several conUolled trials
with monoamine oxidase inhibitors (MAOIs) were reported.
Later, the selective, reversible MOAIs brofaromine and moclo-
bemide were investigated. Neither of the latter two MAOIs are
currently available in the United States. Similar to the TCA trials,
most, although not all, of the placebo-conuolled MAOI trials
demonstrated efficacy for reducing the frequency of binge
eating. Also in similarity to the TCAs, the MAOIs have unique
tolerability and safety concerns. While the TCAs and MAOIs
may remain effective and appropriate options for selected
patients, the marketing of the first selective serotonin reuptake
inhibitor (SSRI), fluoxetine, in the late 1980s resulted in a shift
toward the use of newer antidepressant medications for BN.
Investigations shifted to the SSRIs in the 1990s. The US
Food and Drug Administration approved fluoxetine 60 mg
day“* for BN treatment. Romano and colleagues examined
whether fluoxetine could prevent relapse in BN following suc-
cessful acute treatment. Although the data were positive, the
study suffered from a high drop-out rate.
Two novel treatment approaches, ondansetron and topira-
mate, have also been investigated in controlled trials over the
past decade. Ondansetron, a serotonin-3 receptor antagonist
used for the prevention and treatment of nausea and vomiting,
was shown to significantly reduce binge-eating and vomiting
episodes in a 4-week, placebo-controlled trial at a daily dose of
24 mg, divided into six 4-mg doses.
Topiramate, an anticonvulsant medication with a complex
mechanism of action, has been examined in two 10-week,
placebo-controlled trials for BN. Results from both these trials
showed a reduction in binge-eating and purging frequency
along with a reduction in body weight with topiramate relative
to placebo.
Interspersed throughout the timeline of the previously dis-
cussed studies were trials that examined an assortment of other
compounds. Results from clinical trials with the opioid recep-
tor antagonist naltrexone have been mixed. Reboxetine, a selec-
tive norepinephrine reuptake inhibitor that is not currently
available in the United States, has been reported to reduce
symptoms of binge eating, vomiting, and depression in a
small case series and subsequent uncontrolled pilot study.
A variety of other medications have also been explored in
BN, including D-fenfluramine, flutamide, phenytoin, an anti-
androgenic oral contraceptive, spironolactone, and methyl-
phenidate. Concerns over efficacy, safety, tolerability, or the
need for additional data exclude these compounds from rou-
tine clinical use for the treatment of BN at the present time.
Although pharmacotherapy trials for BN have consisted of
a variety of medications and have often resulted in positive
efficacy data, the majority have been of short duration. For
many of these medications, it is unclear whether improve-
ments gained during the acute treatment are sustained over a
longer duration of treatment.
Psychobiology
EDs in general and BN in particular in our discussion here are
probably best considered as examples of gene-environment
interactions. Clearly environmental factors are important, but
the available study suggests a strong influence of genetic factors
for the liability for this disorder. Relative to specific genetic risk
factors, seven linkage regions have been identified in genome-
wide screens of patients with AN and BN, but no consistent
association between a candidate gene and BN has been identi-
fied. These problems can be overcome only with larger
genetic association studies involving multicenter collabora-
tions. Also, the heterogeneity of eating disorders (ED) pheno-
types represents a major problem, and it will be necessary for
future studies to focus on more homogenous groups using
selective phenotypes and perhaps focusing on specific ED traits.
Also, there is some interesting work suggesting poor treatment
response in patients with bulimic syndromes who have low
functioning alleles of 5-HTTLPR and -1438 G/A, suggesting
hereditary factors involving the serotonin system. Relative to
the psychobiology of BN, several areas have developed over the
last few decades. One such area has been an examination
of baseline concentrations and physiological responses of
various gut peptides in patients with eating disorders. This
work has been recently reviewed. Results in these studies have
been somewhat inconsistent as have the conclusions that
have been drawn from these studies. Some results suggest
elevations in gut peptides such as ghrelin, PYY, and cholecysto-
kinen. Others have not replicated these findings. There has also
been an interest in examining biogenic amines and their meta-
bolites in various bodily fluids. Also various brain imaging
techniques have been used to examine possible structural and
416 Bulimia Nervosa
functional changes in the CNS including magnetic resonance
imaging, positron emission tomography and single photon
emission commuted tomography. CT studies have generally
found some evidence of decreased cortical mass in patients
with BN. There have also been studies in general showing the
significant correlations between measures of harm avoidance
and binding for the 5HT 1A and 5HT 2A / D2/D3 receptors in
certain brain areas such as the mesial temporal lobe, and harm
avoidance appears to be elevated after recovery from BN. There
also appear to be interesting changes in frontal striatal circuits
across multiple childhood psychopathologies including BN as
well as altered striatal responses to reward in recovered women
with BN.
Conclusions
BN is a prevalent, serious, ED among young women. It can be
associated with serious medical and psychosocial complica-
tions. Many patients with BN have other psychiatric problems.
Antidepressant drugs improve the symptoms of BN. However,
CBT appears to be more effective and is the treatment of choice.
Both genetic and environmental factors play a role in the
etiology of this disorder.
See also: Body Image; The Mind-Body Problem; Obsessive-
Compulsive Disorder; Self-Esteem.
Further Reading
Agras WS (2001) The consequences and costs of the eating disorders. Psychiatric
Clinics of North America 24: 371-379.
American Psychiatric Association (2002a) Treatment of Patients with Eating Disorders,
3rd edn. Arlington, VA: American Psychiatric Association.
American Psychiatric Association (2002b) Diagnostic and Statistical Manual of
Mental Disorders, Revised, 4th edn. Arlington, VA: American Psychiatric
Association.
Becker A (2004) Television, disordered eating, and young women in Fiji: Negotiating
body image and identity during rapid social change. Culture, Medicine and
Psychiatry 4: 533-559.
Bulik CM, Hebebrand J, Keski-Rahkonen A, et al. (2007) Genetic epidemiology,
endophenotypes, and eating disorder classification. International Journal of Eating
Disorders 40: 52-60.
Fairburn CG and Beglin SJ (1990) Studies of the epidemiology of bulimia nervosa.
American Journal of Psychiatry 147: 401-408.
Fairburn CG, Cooper Z, and Shafran R (2003) Cognitive behavior therapy for
eating disorders: A 'transdiagnostic’ theory and treatment. Behavioral Research
Therapy 41: 509-528.
Fluoxetine Bulimia Nervosa Collaborative Study Group (1992) Fluoxetine in the
treatment of bulimia nervosa. Archives of General Psychiatry 49: 139-147.
Hoek HW and van Joeken D (2003) Review of the prevalence and incidence of eating
disorders. International Journal of Eating Disorders 34: 282-296.
Hudson ME, Hiripi E, Pope HG, and Kessler RC (2006) The prevalence and correlates of
eating disorders in the National Comorbidity Survey Replication. Biological
Psychiatry 61 : 348-358.
Kaye W (2007) Neurobiology of anorexia and bulimia nervosa. Physiology and Behavior
94: 121-135.
Kendler KS, MacLean C, Neale M, Kessler R, Heath A, and Eaves L (1991)
The genetic epidemiology of bulimia nervosa. American Journal of Psychiatry
148: 1627-1637.
Mitchell JE, Agras S, and Wonderlich S (2007) Treatment of bulimia nervosa:
Where are we and where are we going? International Journal of Eating Disorders
40: 95-101.
Romano SJ, Halmi KA, Sarkar NP, Koke SC, and Lee JS (2002) A placebo-controlled
study of fluoxetine in continued treatment of bulimia nervosa after successful acute
fluoxetine treatment. American Journal of Psychiatry 159: 96-102.
Russell G (1979) Bulimia nervosa: An ominous variant of anorexia nervosa.
Psychological Medicine 9: 429-448.
Relevant Websites
www.aed.org - Academy for Eating Disorders.
www.nationaleatingdisorders.org - National Eating Disorders Association.
Bullying
S M Swearer, A Collins, and B Berry, University of Nebraska - Lincoln, Lincoln, NE, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Bully/victim continuum Bullying behaviors are
depicted on a continuum of four behaviors: bullying,
victimization, experiencing both, and watching
bullying occur.
Bullying A relationship that occurs between two or
more individuals, with repeated interactions
characterized by an imbalance of power and the
intent to harm.
Cyberbullying An individual or a group willfully using
information and communication involving electronic
technologies to facilitate deliberate and repeated harassment
or threat to another individual or group by sending or
posting cruel text and/or graphics using technological
means.
Dating aggression Physical, sexual, and/or emotional
aggression carried out by a dating partner.
Direct behaviors A form of bullying that includes behaviors
that are overt, usually involve a face-to-face interaction,
and can be either physical or verbal.
Indirect behaviors A form of bullying that includes
behaviors that are covert, possibly relationally aggressive,
including cyberbullying.
Intimate partner aggression Physical, sexual, and/or
emotional aggression carried out by an intimate partner in
the context of control.
Definition
Bullying is most commonly defined as any form of aggression
in which one or more individuals repeatedly verbally or physi-
cally harass a victim without provocation. According to the
American Psychological Association, bullying is persistent,
threatening, and aggressive behavior directed toward others,
especially those who are smaller or weaker. There are three
common characteristics found in most definitions of bullying:
(a) an imbalance of power, (b) the intent to harm, and (c) the
display of repetitive behavior. An imbalance of power is ste-
reotypically characterized as a physical issue, with the bully
being much larger and stronger than the victim. While this is
the stereotypical image of bullying, an imbalance of power can
also include characteristics such as intellectual ability, social
status, or social prowess. To illustrate this issue, consider the
example of a high school senior who is physically shorter and
weaker than a high school freshman; however, the freshman
does not believe he can defend himself because he is an under-
classman. Because of the senior's greater social status, he is able
to repeatedly make the freshman stand in a hallway trashcan.
Thus, the senior's behavior is considered as bullying, despite
his physical size.
The behaviors that constitute bullying can be divided into
two forms: (a) direct behaviors and (b) indirect behaviors.
Direct behaviors are more overt, can be both physical and ver-
bal, and involve face-to-face interactions between the bully and
the victim (e.g., hitting, shoving, and calling names). Indirect
bullying behaviors can be relationally aggressive, are typically
covert in nature (e.g., spreading rumors, excluding someone
from a group), and can also include cyberbullying (e.g., bullying
through technology such as cell phones and computers). The
underlying factor distinguishing direct forms of bullying from
indirect forms is whether or not there is a face-to-face interaction
between the bully and the victim.
Bullying behaviors exist on a continuum and are not
permanently set in fixed roles in individuals. There are four
behaviors along the bully/victim continuum: (a) bullying,
(b) victimization, (c) experiencing both (bullying-victimization),
and (d) watching bullying take place. Bullies are those indivi-
duals who exhibit only bullying behaviors, while victims are
those who are targeted by bullies. Bully-victims are those
individuals who both bully others and are bullied by others.
Relative to only bullies or only victims, bully-victims are typi-
cally more impulsive, have greater difficulty regulating their
emotions and reactions, and experience the greatest psycholog-
ical impact from their involvement in the bullying continuum.
Bystanders are those who witness bullying but are involved
neither as a bully nor as a victim. Individuals involved in bully-
ing may move between roles or hold multiple roles on the
bully/victim continuum. Thus, bullying/victimization should
be considered a dynamic, evolving experience, and not a static,
fixed experience.
Prevalence
Bullying is a worldwide phenomenon. Studies have shown
that bullying behaviors exist with similar prevalence rates
across countries (e.g., Australia, United States, Japan, Italy, the
Netherlands). Bullying occurs in rural and urban populations,
across social classes, large and small schools, single-sex
and coeducational schools, romantic relationships, and the
workplace. In short, bullying is an issue that is present in the
majority of the world. Although the fact that bullying exists
is clear, the determination of reliable and valid prevalence rates
of bullying is more difficult.
The prevalence rates of bullying in schools vary based on
the definition and methodology used in the research. One of
the largest samples of bullying behavior in the United States
found that, of 15 686 students in grades 6 through 10, 13%
reported frequent involvement as bullies, 10.6% as victims,
and 6% as bully-victims for a total rate of 29.9% for frequent
involvement in bullying. However, when examining students
417
418 Bullying
who were bullied 'regularly,' a separate study found victimiza-
tion rates to be between 5% and 9%. Participants across cul-
tures or countries may also report differing prevalence rates.
In Australia, for example, within a sample of 2680 students
with an average age of 13, 33% of students reported being
bullied at one time. A study of over 5000 Swedish students,
however, found that 20.2% were repeatedly victimized every
month. Cultural differences may partially explain the differ-
ences in prevalence rates found between countries. For example,
Japan is considered to be a more collectivist country compared to
the United States and the topography of bullying is influenced
by this cultural difference. In Japan, bullying is referred to
as 'ijime' and is rarely a one-to-one interaction. The majority
of bullying incidents that occur in Japanese schools involve
multiple students directly or indirectly bullying an individual
student. Given these differences, perhaps it is not surprising
that such different prevalence rates have been found across
studies.
At this time, there is no universally accepted method
for measuring bullying. The most common methods used for
measuring bullying behaviors are self-report from students,
teachers, and parents, direct observations, and teacher and
peer nominations. Each method has strengths and limitations,
making it difficult for one method to be considered the 'gold
standard.' Additionally, other factors can influence reported
prevalence rates. For instance, some studies assess relational
bullying, while others only examine more direct behaviors
such as hitting or name-calling. Despite the differing cultures,
methodologies, and definitions, the overall range of prevalence
rates in the majority of research on bullying suggests that
bullying is an important issue.
Development
When examining bullying across students' school career, we
see that it typically increases at the end of elementary school
(grades 4 and 5), peaks in the middle school years (grades
6-8), and begins to decline in high school (grades 9-12).
Despite this developmental trend, research on the bullying
behaviors of younger children (under age 6) and adults is less
abundant. The method of bullying may also change depending
on the age of a student. In earlier grades, bullying is typically
more direct and physical, while relational bullying increases
as students progress through school. Furthermore, as children
become more cognitively and socially mature, the topography
and methods they use to bully others change.
Although the majority of research begins measuring bully-
ing in middle childhood, bullying behaviors have been found
as early as preschool. Relative to older students, preschoolers
show higher rates of physical aggression. This is often explained
by the lack of social skills and problem-solving abilities children
have at younger ages. Supporting this argument is the finding
that at the beginning of the school year, preschool children
typically show increased rates of physical and verbal aggres-
sion, which decline over the year. Relational bullying does
occur during preschool, but it is less frequent and more direct
than in older children (e.g., 'You can't play with our group
unless you give me your toy'). Preschool bullying behavior has
also been linked to future behavioral problems. More
specifically, preschool children who bullied others were more
likely to be diagnosed with Conduct Disorder in adolescence.
This suggests that as early as preschool, children engage in
bullying behaviors, and the effects of those experiences are
carried forth into adolescence.
As students enter middle school, the topography of bully-
ing begins to change. By middle school, bullying tends to
become more covert and relational. A student might replace
the overt physical bullying behavior of hitting with the more
covert behavior of spreading rumors. As students enter early
adolescence, their social group is increasingly used for emo-
tional support. Thus, bullying behaviors target these social
areas as they are important for adolescents. It is also during
this time that cyberbullying becomes more prevalent. As par-
ents give their children more independence and privileges (e.g.,
cell phones, less restrictive Internet access), the opportunity for
bullying increases. Additionally, romantic relationships begin
to play a role in the experience of bullies at this time. Bullies
have been shown to begin romantic relationships earlier than
students not involved in bullying. While bullies begin dating
earlier, and in more advanced forms, they also report less
emotional support in their relationships. On the basis of this
research, bullies appear to be at a greater risk for developing
unhealthy romantic relationships (e.g., domestic violence and
sexual harassment).
During middle to late adolescence, bullying behaviors once
more begin to change. Physical bullying becomes even less
frequent as the consequences for physical aggression become
more severe. With the increased cross-sex interactions that
occur during adolescence, a new context for bullying emerges.
For example, in high school, rumors may be spread that a
female student is pregnant. While these types of bullying beha-
viors do occur in earlier grades, it is during high school that
they become more prevalent. Also during this time, and into
adulthood, bullying behaviors may transfer into romantic rela-
tionships. Research has suggested, for example, that bullying
may turn into sexual harassment as students become more
aware of their own and others' sexuality. Additionally, bullying
in the workplace does occur, although less frequently than
during the primary and secondary school years.
Impact
The experience of being victimized has been found to be
associated with a wide range of consequences. Children who
are victimized by their peers have been found to show more
internalizing (e.g., anxiety and depression) and externalizing
(e.g., delinquent and aggressive behaviors) problems. Similarly,
adolescent victims have reported more trauma symptoms, a
higher rate of substance abuse, lower levels of self-esteem,
poorer health, and higher rates of suicidal ideation. The expe-
rience of being a victim of bullying may be linked to students'
decisions to carry out physical attacks, and even lethal attacks.
Clearly, there are a variety of negative, even potentially fatal,
short- and long-term effects of bullying. However, the adverse
effects of bullying may be dependent on the type of victimi-
zation one experiences. It has been found, for example, that
students who experienced verbal victimization, physical
victimization, or sexual harassment had a higher risk of
Bullying 419
experiencing behavioral problems, while being relationally
victimized was not associated with these negative effects.
Research has shown that individuals who bully are also at
risk for adverse effects. Both bullies and victims of bullying
have been found to be more likely than their nonbullying peers
to experience depressive symptoms. Additionally, students
who bully others have shown higher general misconduct in a
variety of settings, greater levels of impulsivity, significantly
greater levels of anger, and more feelings of depression com-
pared to peers who do not bully. Such negative effects may be
even more pronounced for students who are bully-victims, as
one study found that such students were significantly more
likely to show symptoms of depression compared to their
peers who bullied, were victims, or were not involved. In fact,
compared to other groups, bully-victims were four times more
likely to lack a positive self-image and three times more likely
to report feeling insecure within the school setting.
Research suggests that negative psychosocial outcomes exist
for those individuals who bully, those who are victimized, and
those who are bully-victims. The same appears to be true for
bystanders. One study found that adults who witnessed work-
place bullying were more likely to suffer poor mental and
physical health compared to employees who neither experi-
enced nor witnessed workplace bullying. Additionally, a recent
study found that people who had witnessed a great deal of
bullying in their youth reported clinical levels of distress and
psychological trauma similar to the levels exhibited by people
who had witnessed catastrophic events such as earthquakes.
This evidence is used to support the claim that if bullying was
truly a normal, natural part of growing up, it would not result
in such devastating levels of distress and psychological trauma.
Thus, it is clear that bullying can cause serious harm to all
the parties involved, even to those who 'simply'witness the
bullying incident.
Gender Differences
At this point of time, there is little consensus regarding whether
or not boys and girls differ in the frequency with which they
bully. After analyzing seventh- and eighth-grade students'
responses to questionnaire items assessing the frequency with
which they bullied, one study found that males were signifi-
cantly more likely to bully others than their female peers.
However, Crick and Grotpeter found that approximately
equal numbers of boys and girls (27.0% and 21.7%, respec-
tively) were classified as aggressive, meaning that they engaged
in some type of bullying. Therefore, it is not clear whether boys
and girls bully at similar rates.
Furthermore, it is not clear whether boys and girls use
different types of bullying with similar frequencies. When
measuring overt types of aggression, which included physical
and verbal aggression, Crick and Grotpeter found that 15.6%
of boys but only 0.4% of girls were classified as overtly
aggressive. In addition, when examining relational aggression,
defined as attempts or threats to harm another's peer relation-
ship, 17.4% of girls but only 2.0% of boys were classified
as relationally aggressive. Thus, boys were significantly more
likely than girls to be overtly aggressive, while girls were sig-
nificantly more likely than boys to be relationally aggressive.
This is consistent with other research findings that females are
more likely to be involved in relational and indirect forms of
bullying, while males are more likely to be involved in direct
forms of bullying.
However, other research studies suggest that gender differ-
ences in relational aggression are not as clear. When the study
undertaken by Crick and Grotpeter was replicated using a sam-
ple of third- and sixth-grade students, the results showed that
there were no gender differences in the percentage of boys and
girls found to be relationally aggressive. Furthermore, although
the replicated study, like that of Crick and Grotpeter, found that
boys were more overtly aggressive than girls, it acknowledged
that their definition of overt aggression included only physical
aggression and theorized that had they included verbal aggres-
sion in their measure of overt aggression, they would not have
found gender differences either.
Aside from school-age children and adolescents, gender
differences with respect to bullying have been studied among
adult populations as well. One study found that, within a
sample of male and female university students, males reported
using physical aggression significantly more often than females
but there were no gender differences in the reported use of
relational aggression. A separate study measuring workplace
bullying among a sample of adults employed in nine compa-
nies in Iceland, found that men reported bullying significantly
more often than women. Although men and women reported
being victimized at similar rates, gender differences emerged
in the manner in which these adults responded to such victim-
ization. Specifically, women were more likely to seek help,
while men were more likely to confront the perpetrator(s) of
the aggression.
Additional evidence shows that there may be gender dif-
ferences in the effects of bullying. While some studies have
found a significant relationship between bullying others and
depressive symptoms for both genders, others have found
that females were more likely to develop internalizing symp-
toms due to bullying. Looking at specific types of bullying
separately, Crick and Grotpeter found that relationally aggres-
sive girls reported significantly higher levels of loneliness
and significantly lower levels of peer acceptance compared
to relationally aggressive boys, nonaggressive girls, and nonag-
gressive boys. Relationally aggressive boys did not experience
these problems at significantly higher levels than their peers.
It appears as though girls, but not boys, who perpetrate acts
of relational aggression, are at a higher risk for feelings of
loneliness and low peer acceptance. Furthermore, in one
study, female university students who were physically aggres-
sive were more likely to be depressed than their nonaggressive
peers, while physically aggressive males were no more likely
than peers to be depressed. Thus, physical aggression, too,
seems to be associated with negative outcomes for females
but not for males.
Although it is difficult to conclusively say that gender differ-
ences exist in any area of bullying, it is clear that (a) both boys
and girls engage in bullying, (b) the use of physical, verbal,
or relational aggression is not limited to just one gender,
and (c) the act of bullying is associated with negative outcomes,
females being more likely to experience these outcomes. Further
research is needed to clarify the remaining questions surround-
ing potential gender differences in bullying.
420 Bullying
Cyberbullying
Cyberbullying is a relatively new medium through which bul-
lying occurs (e.g., chat rooms, text messages). Cyberbullying
has been defined as an individual or a group willfully using
information and communication involving electronic technol-
ogies to facilitate deliberate and repeated harassment or threat
to another individual or group by sending or posting cruel text
and/or graphics using technological means. Many of the meth-
ods used in traditional bullying are used in cyberbullying.
Direct cyberbullying can occur when one person calls another
a name through an electronic message. Relational bullying can
also occur online. For example, with the numerous social
networking sites now online (e.g., Facebook, MySpace), 'hate
groups' have become a popular approach to bullying. In a hate
group, a student creates an online social group against a
schoolmate, allows others to join, and collectively the group
posts negative comments about the student. Fortunately, social
networking sites have begun taking action against the creation
of hate groups. When creating a group on Facebook, for
instance, a warning is placed near the bottom of the page that
reads, "Note: groups that attack a specific person or group of
people (e.g., racist, sexist, or other hate groups) will not be
tolerated. Creating such a group will result in the immediate
termination of your Facebook account." There are several simi-
larities between cyberbullying and traditional bullying. Exam-
ining the overlap between cyberbullying and traditional
bullying, one study found that 56% of online victims were
also victims of traditional bullying. Unfortunately, no infor-
mation exists on whether the cyberbully was the same person
who also bullied the victim in traditional ways. Interestingly,
this also means that some victims are targets of only online
bullying. One explanation that has been proposed for this
phenomenon involves the anonymity an individual has on
the Internet. In traditional bullying, the person bullying a
victim is almost always known. Online, however, individuals
are represented by screen names that do not necessarily corre-
spond with their real-life identity. The psychological outcomes
of bullies and victims of cyberbullying are also similar in
relation to traditional bullying. Approximately 38% of youth
who are harassed online report emotional distress about the
event. Additionally, the type of harassment perpetrated has an
effect on the degree of emotional distress one experiences. One
of the most emotionally distressing types of harassment men-
tioned by youth includes being propositioned to send a picture
of oneself.
Some research has shown that there are also differences
between cyberbullying and traditional forms of bullying.
As with traditional bullying, prevalence rates of cyberbullying
vary depending upon the method of assessment and the
definition of cyberbullying. A synthesis of research findings
suggests that 10-30% of youth are involved in cyberbullying.
There are also gender differences in cyberbullying. One study
found that the gender of online harassers was split evenly
between males and females. These findings, however, may
be influenced by the lack of physical bullying that can occur
on the Internet. In another contrast to traditional bullying, in
which prevalence rates decrease with age, cyberbullying has
been shown to increase with age. This may be due to the
increased availability of the Internet to older students. With
an increasing number of adolescents and younger children
going online, cyberbullying is an important area for further
research.
Workplace Bullying and Sexual Harassment
Bullying is a very common occurrence in schools, but it is not
limited to this domain. In fact, bullying has been found to
occur in a variety of contexts, including among adults within
the workplace. A 2001 study found that nearly half (i.e.,
46.5%) of a sample of 5288 employees reported being bystan-
ders of workplace bullying in the past 5 years. Workplace
bullying is known by many names, such as mobbing and
harassment, and has been defined in a multitude of ways.
It is a complex phenomenon that has yet to be consistently
defined.
For their purposes, Baillien, Neyens, De Witte, and De
Cuyper defined workplace bullying as negative behaviors in
the workplace that persist for a minimum of 6 months and may
be psychological, physical, or sexual. To explore individual
and work-related factors that may lead to workplace bullying,
Baillien et al. interviewed 87 informants who reported that
they had knowledge of bullying occurring in the workplace.
The results showed three 'pathways' that could lead to the
development of workplace bullying: (a) frustration and inef-
fective coping, (b) interpersonal conflict, and (c) a negative
culture and/or harmful habits within the organization. Abetter
understanding of these antecedents may lead to more efficient
ways of preventing workplace bullying from occurring.
In order to further understand workplace bullying, it has
sometimes been studied in conjunction with sexual harassment,
another form of workplace aggression. Sexual harassment has
been defined as behaviors with sexual content that negatively
affect one or more people by humiliating, ridiculing, and/or
intimidating them. Sexual harassment can be verbal (e.g.,
telling sexual stories or making sexual comments about a
person), nonverbal (e.g., wolf whistling or other gestures
with sexual connotations), or physical (i.e., unwanted touch-
ing) and can range from telling jokes with sexual content to
criminal actions such as rape. When considering the range of
aggressive behaviors in the workplace, it has been theorized
that these could be thought of as residing upon a continuum,
with acts as harmless as channeling aggression into effective
work practices on one end of the continuum and something
as violent as homicide in the workplace on the other end.
Workplace bullying would fall somewhere in the middle of
this continuum, as would sexual harassment.
To explore the relationship between sexual harassment
and workplace bullying, one study queried male and female
employees at a large university to complete questionnaires
that asked about several work-related subjects. Path analysis
revealed that the occurrence of workplace bullying was closely
related to aspects of work organization for both men and
women but that sexual harassment was related to work organi-
zation for women only. Specifically, the relationship between
workplace bullying, conflicts at work, and gender differences
was significant for men and women, although it was stronger
for women. Informal decision structures were related to bully-
ing for men, while lack of equality and conflicts were related to
Bullying 421
sexual harassment for women. In addition to these gender
differences, a greater number of women experienced both
sexual harassment (9% of women vs. 2% of men) and bullying
(13% of women vs. 9% of men).
The relationship between sexual harassment and bullying
has also been studied among adolescents in the school context.
In one study, middle and high school students completed
questionnaires measuring bullying, sexual harassment, and
sexual orientation. The results were analyzed to explore whether
the frequency and/or health effects of bullying and sexual
harassment differed by gender and/or sexual orientation. Bully-
ing was found to occur more frequently than sexual harass-
ment and members of sexual minorities reported higher rates
of both bullying and sexual harassment. While both bullying
and sexual harassment had adverse effects on students' health,
sexual harassment had more negative effects on health than
bullying, especially for girls and members of sexual minorities.
Although this study is limited by the small numbers of ethni-
cally and racially diverse students as well as students of sexual
minorities, it helps to demonstrate the need to study bullying
in conjunction with sexual harassment.
Intimate Partner Aggression
Like sexual harassment, intimate partner aggression and dating
violence are unfortunate realities for adults as well as adoles-
cents. These forms of aggression are of particular concern
because, in addition to causing direct harm to the victims,
children who witness such aggression appear to be more likely
to engage in bullying (see section 'Cycle of Aggression').
Within a sample of 1059 elementary and middle school chil-
dren in one study, a substantial proportion (17.4%) reported
witnessing intimate partner aggression between their parents at
home. Rates of intimate partner aggression appear to be even
higher when women are directly asked about their experience
or lack thereof with such aggression over their entire lifetimes.
On interviewing 1401 women from 18 to 65 years of age, over
half (55.1%) reported having experienced intimate partner
aggression with a past or present male partner. Clearly, a
woman's lifetime risk of experiencing intimate partner aggres-
sion is alarmingly high.
Equally disturbing is the fact that such aggression is not
limited to adults. According to ninth- through twelfth-grade
students' responses on the Youth Behavior Risk Survey, 10.3%
of girls had experienced dating aggression, defined as being
intentionally, physically hurt by a dating partner. Dating
aggression was also found to be associated with involvement
in other types of violence, substance use/abuse, risky sexual
behavior, experiencing sad/hopeless feelings, and contemplat-
ing or having attempted suicide. Espelage and Holt found that
the experience of dating aggression was also closely related
to the experience of bullying for a sample of 685 male and
female seventh- through twelfth-grade students. Specifically,
students who reported being bully-victims reported experien-
cing higher levels of physical dating aggression than students
who reported being bullies, victims, or uninvolved in bullying.
Bully-victims also reported experiencing higher levels of sexual
harassment compared to members of the other three groups
and higher emotional abuse within a dating relationship
than victims and students uninvolved in bullying. Given this
overlap between bullying, sexual harassment, and dating
aggression, it appears as though a substantial proportion of
adolescents are being victimized in numerous ways.
Cycle of Aggression
Existing research has shown that bullying, as well as other types
of violence and aggression, may follow a cycle of aggression
that begins at home. For instance, one study found that young
adults who were maltreated as children were more likely to
perpetrate acts of general aggression as well as intimate partner
aggression. Furthermore, among a sample of 43 death row
inmates convicted of homicide, childhood maltreatment was
extremely prevalent. Over half of the inmates reported experi-
encing sexual abuse, 95% had been physically abused, and a
substantial proportion (75%) had experienced multiple forms
of victimization, including verbal, physical, and sexual abuse,
neglect, and having witnessed others being victimized.
Additional evidence supports the notion that being a
bystander is associated with later aggression. In addition to
early childhood physical abuse, witnessing a parent suffer
domestic abuse predicted later perpetration of intimate partner
aggression. It has also been found that there is a significant
association between children witnessing intimate partner
aggression between their parents at home and bullying others
at school. This association could be explained by Bandura's
Social Learning Modeling Theory, which states that people,
particularly children, learn certain behaviors by watching the
adults in their lives model them. In other words, children who
see themselves or others being victimized may learn to victim-
ize, be it via bullying or other aggressive acts.
Not all children who witness or experience such aggression
will later exhibit these same or similar violent behaviors, how-
ever. The pathway between the witnessing/experiencing of
aggression and later perpetration of violent acts is not certain
and, thus, the witnessing and/or experiencing of aggression
does not guarantee that such people will fall prey to this
'cycle of aggression' and perpetrate violent acts themselves.
Yet, as a whole, evidence seems to suggest that witnessing
and/or experiencing aggression, particularly multiple forms
of aggression, may be risk factors for the later perpetration of
violent acts.
Conclusion
Bullying is a prevalent phenomenon that causes harm to those
who perpetrate acts of bullying, experience acts of bullying,
or witness bullying occurring. While bullying is frequently
studied in middle childhood and in the context of schools,
bullying occurs in a variety of contexts across the life span,
from preschool to adulthood. Bullying also shares a close
relationship with sexual harassment, intimate partner aggres-
sion, and childhood maltreatment. Such aggressive acts have
been theorized to occur within a vicious cycle in which those
who experience aggression are more likely to perpetrate acts of
aggression in the future. In short, bullying can and does affect
the lives of countless children and adults.
422 Bullying
See also: Aggression; Jealousy; Self-Esteem; Suicide; Violence.
Further Reading
Baillien E, Neyens I, De Witte H, and De Cuyper N (2009) A qualitative study on the
development of workplace bullying: Toward a three way model. Journal ot
Community & Applied Social Psychology 19: 1-16. doi:10.1002/casp. 977.
Bandura A (1977) Social Learning Theory. New York: General Learning Press.
Beran T and Li Q (2005) Cyber-harassment: A study of a new method for an old
behavior. Journal of Educational Computing Research 32(3): 265-277.
doi:10.21 90/8YQM-B04H-PG4D-BLLH.
Card NA, Sawalani GM, Stucky BD, and Little TD (2008) Direct and indirect aggression
during childhood and adolescence: A meta-analytic review of gender differences,
intercorrelations, and relations to maladjustment. Child Development
79: 1185-1229. doi:1 0.1111 /j.1 467-8624.2008.01 1 84.x.
Connolly J, Pepler D, Craig W, and Taradash A (2000) Dating experiences of bullies
in early adolescence. Child Maltreatment 5: 299-310. doi:10.1177/
1077559500005005002.
Crick NR and Grotpeter JK (1995) Relational aggression, gender, and social-
psychological adjustment. Child Development 66: 710-722. doi:10.2307/1 131945.
Espelage D and Holt M (2007) Dating violence & sexual harassment across the
bully-victim continuum among middle and high school students. Journal of
Youth and Adolescence 36: 799-811. doi:1 0.1 007/si 0964-006-91 09-7.
Espelage D and Swearer S (2003) Research on school bullying and victimization:
What have we learned and where do we go from here? School Psychology Review
32: 365-383.
Jimerson SR, Swearer SM, and Espelage DL (2010) Handbook of Bullying in Schools:
An International Perspective. New York: Routledge.
Kowalski RM and Limber SP (2007) Electronic bullying among middle school students.
The Journal of Adolescent Health 41: S22-S30. doi:1 0.1 01 6/j.jadohealth
2007.08.017.
Nansel T, Overpeck M, Pilla R, Ruan W, Simons-Morton B, and Scheidt P (2001)
Bullying behaviors among US youth: Prevalence and association with psychosocial
adjustment. Journal of the American Medical Association 285: 2094-2100.
doi:1 0.1 001/jama.285. 16.2094.
Olafsson RF and Johannsdottir HL (2004) Coping with bullying in the workplace:
The effect of gender and type of bullying. British Journal of Guidance & Counseling
32: 319-333. doi:1 0.1 080/0306988041 0001 723549.
Olweus D (1993) Victimization by peers: Antecedents and long-term outcomes.
In: Rubin KH and Asendorpf JB (eds.) Social Withdrawal, Inhibition, and Shyness
in Childhood, pp. 315-341. Hillsdale, NJ: Erlbaum.
Patchin JW and Hinduja S (2006) Bullies move beyond the schoolyard: A preliminary
look at cyberbullying. Youth Violence and Juvenile Justice 4(2): 148-169.
doi:1 0.1 1 77/1541 204006286288.
Swearer S, Grills A, Haye K, and Cary P (2004) Internalizing problems in students
involved in bullying and victimization: Implications for intervention. Bullying in
American Schools: A Social-Ecological Perspective on Prevention and Intervention,
pp. 63-83. Mahwah, NJ: Lawrence Erlbaum.
Swearer SM (2008) Relational aggression: Not just a female issue. Journal of School
Psychology 46: 611-616. doi:10.1016/j.jsp. 2008.08.001.
Swearer SM and Doll B (2001) Bullying in schools: An ecological framework. Journal
of Emotional Abuse 2: 7-23.
Swearer SM, Espelage DL, Vaillancourt T, and Hymel S (2010) What can be done about
school bullying? Linking research to educational practice. Educational Researcher
39: 38-47. doi:1 0.31 02/001 31 89X09357622.
VandenBos GR (2007) APA Dictionary of Psychology. Washington, DC: American
Psychological Association.
Ybarra M and Mitchell K (2007) Prevalence and frequency of Internet harassment
instigation: Implications for adolescent health. The Journal of Adolescent Health
41: 189-195. doi:10.1016/j.jadohealth.2007.03.005.
Relevant Websites
http://brnet.unl.edu - Bullying Research Network.
www.bullying.org - bullying.org.
www.education.com/bullying - Bullying at school and online.
http://www.ncsvprp.org - Consortium to Prevent School Violence.
http://www.safeyouth.gov - Striving to Reduce Youth Violence Everywhere.
Caffeine
W Revelle, D M Condon, and J Wilt, Northwestern University, Evanston, IL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Adenosine A neuromodulator associated with
regulating the degree of alertness in central nervous
system functioning. At adenosine receptors inhibit
excitatory neurotransmitters, while A2 receptors
facilitate neurotransmission by inhibiting At
receptors.
Affect The feeling state associated with an emotion.
Affect is typically used in terms of direction of approach
(positive affect) and avoidance (negative affect). It is
most typically measured by self-reports of positive
and negative affect.
Arousal A construct used to organize the common
behavioral effects of exercise, stimulant drugs, sleep
deprivation (negatively), time of day, and time on task.
Arousal is associated with feelings of energy and tension. It is
better thought of in terms of the two independent
dimensions of energetic arousal and tense arousal. It
typically involves increases in heart rate, blood pressure, skin
conductance, and other autonomic indicators.
Caffeinism Chronic psychological dishess, including sleep
and anxiety disorders, associated with high levels of caffeine
consumption.
Dose response The shape of the relationship between
increasing dosage levels (measured in mg of caffeine/kg of
body weight) with physical or psychological responses.
Energetic arousal One of the two dimensions of arousal,
associated with self-reports of feeling energetic, lively, alert
versus sleepy, tired, and drowsy. It is correlated with but
separable from positive affect.
Negative affect A second unipolar dimension of affect. It is
associated with self-reports of feeling sad, blue, unhappy,
and depressed.
Positive affect One of the two unipolar dimensions of
affect. It is associated with self-reports of feeling happy,
pleased, delighted, and satisfied.
Tense arousal A second dimension of arousal, associated
with self-reports of feeling scared, tense, nervous, afraid
versus calm, relaxed, and at ease. It is correlated with but
separable from negative affect.
Vigilance task It is a monotonous task, typically requiring a
rapid response to the detection of an infrequently occurring
event. Most vigilance tasks require sustained performance
over a long period.
Withdrawal Symptoms associated with an abrupt cessation
of consumption of a drug (e.g., caffeine).
Introduction
Caffeine and its fellow xanthines, theophylline and theobro-
mine, are the most consumed drugs in the world, surpassing
alcohol and nicotine for popularity and universality. Caffeine,
normally associated with coffee, is also a major ingredient of tea,
as is theophylline, and is even found in chocolate, a source
of theobromine. Caffeine may also be found in the kola nut
(chewed in Africa), yerba mate (taken with hot water in South
America), and guarana seeds (primarily added to soft drinks in
Brazil but also included in various 'natural food supplements').
Moderate doses of caffeine enhance mood, alertness, and physi-
cal endurance, but at the cost of increased tolerance with use and
withdrawal symptoms of severe headache, fatigue, and dys-
phoria. The affective, behavioral, and cognitive consequences
of caffeine consumption are generally positive for moderate
doses, but can lead to clinical diagnoses of anxiety disorder, as
well as to physical tremor, twitching, and cognitive impairment
at higher levels. Besides the obvious sources of coffee and tea,
caffeine is also included as an active ingredient of colas, energy
drinks, and energy bars, sold in over-the-counter pill or lozenge
form, and is added to many over-the-counter and prescription
drugs to alleviate other side effects.
Perhaps because of its common usage and lack of apparent
danger, caffeine is generally not even considered a drug.
Indeed, its very ubiquity makes it very much a part of modern
society and hard to think of as a psychoactive drug. However,
given that its effects are similar to those of many other psy-
choactive drugs, if caffeine were to be discovered today, it is
quite likely that it would be illegal or at least seriously regu-
lated. Indeed, unsuccessful attempts at regulating caffeine
have occurred since coffee's first introduction into Arabia and
Europe.
Coffee came to Europe and then the rest of the world from
Arabia (probably starting in Yemen) or Ethiopia where the
legendary Ethiopian goat herder, Kaldi, saw his goats chew
on the red berries of a bush and then show great energy. Trying
the berries himself is said to have led him to the joys of coffee.
More likely is that warriors of Ethiopia used a mixture of
ground coffee beans and animal fat to provide nourishment
and energy while conducting raids on their neighbors. This
mixture spread to Arabia and then by the late sixteenth century
to the rest of the world. However, tea had been consumed in
India and China for millennia before being introduced to
Europeans searching for spices. The guarana seed was a source
of caffeine in pre-Columbian South America and remains a
major source of caffeine in South America and in 'health food'
stores. The introduction of coffee and tea into Europe as
sources of caffeine had a revolutionary impact.
The ubiquity of caffeine and its cousins as drinks and
candies around the world testifies to the ease of preparation
and its perceived benefits for thinking clearly and feeling alert.
Some attribute the European Enlightenment and subsequent
Industrial Revolution to the introduction of coffee to Europe
423
424 Caffeine
in the early seventeenth century and the resulting replacement
of beer by coffee or tea in workers' and intellectuals' diet. (The
boiling of water necessary for tea and coffee purified the drink-
ing water, making it a safe alternative to beer for breakfast.
Workers could thus work longer and intellectuals argue with
greater logic than had they started their day with beer.) Joining
the classic association with writers and musicians who praised
(and used) the powers of coffee (cf. Balzac, Johnson, Bach,
Mozart), it is difficult today to think of software developers
or college students without a nearby espresso maker, cans of
cola, or more powerful forms of caffeine ('energy drinks').
A wonderful statement about the powers of caffeine frequently
is attributed to the Hungarian mathematician, Erdos, although
probably was made by his colleague, Renyi, who claimed
that 'A mathematician is a machine for turning coffee into
theorems.'
A negative consequence of the introduction of coffee and
tea to Europe and then to the rest of the world was similar to
the later introduction of electric lights: By allowing people to
stay awake when tired or when it is dark, caffeine and the
electric light have modified our sleep patterns in a manner
that leads to sleep deprivation and insomnia. A society without
caffeine or electricity is bound to the natural rhythms of day-
light and spends more time sleeping than does one with caf-
feine and artificial lighting.
Sources and Consumption of Caffeine
Caffeine is typically consumed by drinking coffee or tea,
although it is also found in chocolate, many soft drinks, and
a variety of over-the-counter medicines. The amount of caf-
feine in these various preparations differs drastically with more
variance in tea (at least a factor of 10 between the weakest and
the strongest tea in one study) than in coffee (only a factor of 4
between the weakest and the strongest coffee). What makes
estimates of caffeine consumption difficult is that the standard
'coffee cup' unit of 5oz (150 ml) is probably used less often
than a 'coffee mug' of 8 oz (237 ml) or the 'Grande' size (16 oz
or 473 ml) of some coffee chains. A further difficulty is that
coffee differs in the way it is prepared. In a comparison of
coffees, home-brewed filtered coffee tends to be stronger than
does instant coffee, perhaps because the process of making the
latter is more efficient in extracting caffeine, as well as some of
the other bitter flavors than found in a cup of freshly brewed
and filtered coffee. (Table 1)
In the United States, 95% of the adult population consumes
caffeine in some form, with an average daily consumption
increasing with age to the late 50s and then declining slightly
(Figure 1). Typically taken with breakfast, the serving of coffee
or tea to visitors is an expected custom for many. In Europe and
North America, the social aspect of caffeine consumption
began with the establishment of coffee houses in seventeenth
century Britain and continues to this day with the popularity of
various coffee shop chains. Given caffeine's prevalence in mod-
ern society, it is hard to believe that four centuries ago, coffee
and tea were considered revolutionary, and there was a move
to ban coffee houses as sources of rebellion.
There are vast individual differences in the consumption
of caffeine, probably due to self-titration, in that people
self-administer as much caffeine as they find to yield the posi-
tive effects on mood and alertness but not enough to produce
some of the negative side effects of tension. As is true of most
psychological variables, there are reliable individual differ-
ences in caffeine consumption and these differences are mod-
erately heritable (h 2 > 0.4) with a higher heritability for heavy
use ( h 2 > 0.7). In addition to individual differences within
countries, the consumption of caffeine varies a great deal
between nations. While the average intake in the United States
is roughly 2-3mgkg~\ this varies drastically, with the top
10% of the population consuming as much as 5mgkg~\ In
the United Kingdom, the per capita consumption is about
4 mgkg -1 , with the top 10% consuming as much as 7 mgkg -1 ,
while in Scandinavia, the average is about 7.5 mg kg -1 , with
the top 10% consuming almost 15 mgkg -1 !
Although given GRAS status (generally recognized as safe)
by the US Food and Drug Administration as a supplement to
cola-like beverages, in 2009, the FDA announced that it did not
view it as GRAS as an additive to alcoholic beverages. This led
to some controversy, as a variety of alcohol-producing and -
marketing companies complained that this would limit their
sale of caffeinated alcoholic beverages (CABs). The US Center
for Disease Control has issued warnings about the use of CABs,
as well as the mixing of 'energy drinks' containing caffeine with
alcoholic beverages. The warning is that the stimulating effects
of the caffeine when combined with the disinhibiting effects of
alcohol can lead to binge drinking and increase the likelihood
of driving under the influence of alcohol or to unsafe or
undesired sexual behavior.
Physiology of Caffeine
The full chemical name for caffeine is 1,3,7-trimethylxanthine,
and its chemical formula is C 8 H 10 N 4 O 2 . Caffeine is a member
of the purine family of compounds, meaning that it has a dou-
ble-ringed, crystalline organic base, C 5 H 4 N 4 . When consumed
orally, caffeine is rapidly and completely absorbed into the
bloodstream through the gastrointestinal tract. Individual differ-
ences in the amount of time it takes to achieve peak plasma levels
in the bloodstream can range from 15 to 120 min, with the
majority of individuals reaching peak plasma levels approxi-
mately about 30-60 min after ingestion. Caffeine is broken
down extensively by the liver into three primary metabolites,
paraxanthine (84%), theobromine (12%), and theophylline
(4%), each of which have their own effects on the body. The
liver further breaks down these three metabolites into xanthine
by removal of methyl groups. Xanthine is either excreted in urine
or reused by the body. The average half-life of caffeine, the time it
takes for half of the caffeine consumed to be eliminated, is 2.5-
4.5 h. The half-life may vary depending on a number of factors; it
is much longer in individuals with liver disease, reduced by up to
50% in smokers, may be doubled in women taking contracep-
tives, and increases throughout pregnancy (reaching a maximum
time of 1 5 h during the last trimester of pregnancy) . The half-life
also depends on the amount of caffeine consumed. The pharma-
cokinetics for ~70-100mg of caffeine are linear, but the
clearance of caffeine is significantly reduced and its elimination
half-life is prolonged at higher doses of 250-500 mg indicating
nonlinearity. Caffeine has no nutritional value.
Caffeine 425
Table 1
Common sources of caffeine
Source
Serving
Dose/serving
Dose per ml
Coffee & tea
Espresso, generic
30 ml
30-90 mg a
1.00-3.00 mg ml
Coffee, generic, drip
237 ml
96-288 mg a
0.41-1 .22 mg ml
Coffee, generic, percolated
237 ml
64-272 mg a
0.27-1. 15 mg ml
Coffee, generic instant
237 ml
27-1 92 mg a
0.11-0.81 mg ml
Starbucks brewed coffee (Grande)
473 ml
320 mg
0.68 mg ml -1
Dunkin’ donuts medium coffee
473 ml
206 mg
0.50 mgmr 1
Starbucks vanilla latte (Grande)
473 ml
150 mg
0.32 mgmr 1
Black tea, 1 teabag brewed 3 min
237 ml
25-1 10 mg*
0.1 1-0.46 mg ml
Starbucks chai tea latte (Grande)
473 ml
100 mg
0.21 mgmr 1
Oolong tea, 1 teabag brewed 3 min
237 ml
1 2-55 mg*
0.05-0.23 mg ml
Snapple-flavored iced teas
473 ml
42 mg
0.09 mgmr 1
Green tea, 1 teabag brewed 3 min
237 ml
8-30 mg*
0.03-0.1 3 mg ml
Arizona iced tea, black
473 ml
32 mg
0.07 mg mr 1
Coffee, generic decaffeinated
237 ml
3-12 mg a
0.01-0.05 mg ml
Arizona iced tea, green
Soft drinks
473 ml
15 mg
0.03 mgmr 1
Jolt Cola
355 ml
140 mg
0.39 mgmr 1
Mountain Dew
355 ml
54 mg
0.15 mgmr 1
Diet Coke
355 ml
47 mg
0.13 mgmr 1
Dr. Pepper
355 ml
41 mg
0.12 mgmr 1
Pepsi
355 ml
38 mg
0.11 mg mr 1
Coca-Cola
355 ml
35 mg
0.10 mgmr 1
Diet Pepsi
355 ml
36 mg
0.10 mgmr 1
Guarana sodas
Energy drinks
355 ml
30 mg
0.08 mgmr 1
AMP Energy
237 ml
80 mg
0.34 mgmr 1
Monster Energy
473 ml
160mg
0.34 mgmr 1
Red Bull
Chocolate
250 ml
80 mg
0.32 mgmr 1
Hershey’s special dark chocolate bar
41 g
31 mg
Hershey’s unsweetened baking chocolate
14 g
15 mg
Hershey’s milk chocolate bar
43 g
9 mg
Hershey’s chocolate milk
237 ml
4 mg
0.02 mgmr 1
Starbucks coffee ice cream
Over-the-counter medications
101 g
40 mg
NoDoz (maximum strength)
1 tablet
200 mg
Excedrin (extra strength)
2 tablets
130 mg
Anacin (maximum strength)
Other
2 tablets
64 mg
AMP caffeinated gum
2 pieces
80 mg
Enerjets lozenges
1 lozenge
75 mg
GU energy gel
32 g pack
20 mg
Adapted from the Center for Science in the Public Interest, the USDA National Nutrient Database for Standard Reference, Release 22 (2009), the International Coffee Organization, and
company Web sites.
"Factors affecting the range of caffeine In coffee include the type and origin of the coffee bean, roasting method, grinding method, and brewing method.
^Factors affecting the range of caffeine in tea include the type, origin, and age of the tea plant, the location of the leaf on the plant, the method of steeping (loose leaf, bagged, etc.), and
the length of steeping time (where longer steeping times release more caffeine). 1 fluid oz = 30 ml; 1 oz weight = 28 g.
Not only does caffeine rapidly disperse to all cells in the
body, its chemical structure also allows it to easily cross the
blood-brain barrier. Once in the brain, the principal mode of
action is as a nonselective antagonist of adenosine receptors.
The caffeine molecule is structurally similar to adenosine and
binds to adenosine receptors on the surface of cells without
activating them, therefore acting as a competitive inhibitor or
'antagonist' of adenosine. Caffeine's physiological and psy-
chostimulant effects are thought to derive largely from inhibit-
ing the effects of adenosine, particularly at the A1 and A2a
adenosine receptors. The A1 receptors are located in all parts
of the brain with the heaviest concentration in the hippocam-
pus, cerebral cortex, cerebellum, and certain thalamic nuclei.
The A2a receptors are located in the dopamine-rich areas of the
brain, including the striatum, nucleus accumbens, and olfac-
tory tubercle. Adenosine transmission normally decreases the
neuronal firing rate and inhibits both synaptic transmission
and the release of most neurotransmitters; it promotes sleepi-
ness, dilates blood vessels, reduces the contractions of the
stomach and intestines, slows the reaction to stress, and lowers
426 Caffeine
Figure 1 Caffeine consumption increases by a factor of 21 with age and differs between men and women. However, when considering body weight,
the sex difference is diminished, and the increase with age is less than a factor of 5.
the heart rate, blood pressure, and body temperature. After
caffeine connects to the A1 and A2a receptors, an adenosine
blockage forms, thus reversing the effects of adenosine. Caffeine
is classified as a psychostimulant because it produces heightened
central nervous system (CNS) arousal, which increases the dom-
inance of the sympathetic branch of the autonomic nervous
system. The result of this shift is to increase the levels of activity
in physiological systems that mobilize the body for aggressive or
defensive activity, including the cardiovascular system, the respi-
ratory apparatus, and the digestive system. Under continued
stress, the pituitary gland is instructed by the hypothalamus to
increase the release of hormones that cause the adrenal glands
to produce epinephrine and cortisol. These hormones cause
increased heart rate, dilation of the bronchi, a release of sugar
into the bloodstream from the liver, increased blood flow to the
muscles, and a tightening of muscles for action.
Caffeine also affects dopaminergic transmission. By inhib-
iting adenosine A2a receptors, caffeine reduces the negative
modulatory effects of adenosine receptors on dopamine D2r
receptors, thus causing potentiation of dopaminergic neuro
transmission. This mechanism is thought to be related to the
ability of caffeine to induce a positive mood state that includes
mild euphoria. However, in contrast to the drugs of abuse that
selectively lead to a release of dopamine in the shell of the
nucleus accumbens, caffeine increases dopamine release in the
caudate nucleus. Dopamine release in the caudate nucleus
relates to the stimulatory properties of caffeine on locomotor
activity. Only at doses not likely to be consumed by humans
does caffeine cause release of dopamine in the nucleus accum-
bens; however, this dose is associated with nonspecific activa-
tion in the brain and aversive effects such as anxiety. Repeated
administration of caffeine quickly leads to tolerance of the phys-
iological effects previously described; tolerance refers to an
acquired change in responsiveness after repeated exposure to a
drug. Usually developing within a few days, tolerance to caf-
feine's effects on blood pressure, heart rate, diuresis, plasma
epinephrine and norepinephrine levels, renin activity, tension,
anxiety, jitteriness, and nervousness has also been demonstrated.
Tolerance to sleep disruption can develop within 1 week with
doses as low as two cups of coffee per day. Nonhuman animals
also develop a tolerance to caffeine-induced locomotor stimula-
tion, cerebral electrical activity, and reinforcement thresholds for
electrical brain stimulation. The alerting and attentional focus
effects of the drug do, however, continue.
Withdrawal
Physiological dependence is a state induced by repeated drug
use that results in a withdrawal syndrome when the drug is
discontinued or an antagonist is administered. It is important
to distinguish a withdrawal syndrome from a rebound phe-
nomenon. Withdrawal comprises a number of signs and symp-
toms not present during administration of the drug, whereas
rebound refers to a single sign or symptom that is the reverse
of the drug effect. Additionally, withdrawal most often occurs
after discontinuation of repeated drug administration; re-
bound can occur after single administrations of a drug. The
most oft-reported symptoms of caffeine withdrawal are head-
aches, fatigue, weakness, drowsiness, impaired concentration,
work difficulty, depression, anxiety, irritability, increased mus-
cle tension, decreased energy and activeness, decreased alert-
ness, drowsiness, and irritability. Onset of symptoms can
appear within only 3-6 h after abstinence and can last for
one week, although the incidence or severity of the symptoms
increases with increases in daily dose; symptoms can appear
with doses as low as lOOmgday -1 . It is important to note,
however, that the intensity of withdrawal symptomatology is
subject to substantial individual differences, based, in part, on
genetic predispositions. Some people experience quite extreme
withdrawal symptoms, while others experience almost none.
Although there is an overall, general withdrawal factor,
withdrawal symptoms can be grouped into lower-level factors
Caffeine 427
of fatigue/headache, dysphoric mood, and flu-like symptoms.
The effects of withdrawal are most clearly dose-dependent for
the first two sets of symptoms: compared to light caffeine
consumers (less than 100 mg day -1 ), heavy consumers (greater
than 200 mg day -1 ) are more than four times as likely to report
headache and fatigue as well as more than three times as likely
to report dysphoric mood.
The effect of acute withdrawal from caffeine is seen com-
monly in headache and fatigue. Indeed, office workers who
consume high doses during the week and then do not consume
caffeine on the weekend find that the resulting weekend head-
aches are easily treated with pain relievers that contain caffeine.
Perhaps one cause of postoperative headache is the preopera-
tive withdrawal from caffeine rather than the acute aftereffects
of anesthesia.
Some of the physiological underpinnings of common
withdrawal effects have just recently been discovered. Acute
caffeine abstinence results in increases in blood flow velocity
in middle and anterior cerebral arteries and decreases in
variability of blood flow velocity in the middle cerebral artery.
These effects suggest that a vascular mechanism causes the
caffeine withdrawal symptom of headache. Acute caffeine
abstinence also produces significant changes in electrical activ-
ity in the brain corresponding with increased drowsiness and
decreased alertness.
Withdrawal effects complicate the study of caffeine effects
on mood and performance. Because most participants nor-
mally consume caffeine, it has been proposed that the benefi-
cial effects of caffeine on mood, behavior, and cognitive
performance are actually due to antiwithdrawal effects. That
is, caffeine does not elevate mood or improve alertness, but
rather withdrawal leads to tension and sleepiness. The logic of
this position is that most double-blind caffeine studies are
done after 12-24 h of caffeine withdrawal (e.g., participants
are requested not to consume any caffeine after 8 pm prior to a
morning study), and thus, participants are in acute caffeine
withdrawal. Evidence against this proposal comes from com-
paring predrug administration measures of mood and perfor-
mance for heavy and light caffeine consumers as well as the
observation that both nonconsumers and normal consumers
of caffeine report increased alertness following caffeine
consumption when compared to a placebo. Further evidence
against the withdrawal hypothesis comes from the pattern of
cognitive performance effects of caffeine as they interact with
personality and the time of day.
Affective, Behavioral, and Cognitive Effects of Caffeine
Caffeine is most commonly consumed because it improves
alertness, attentional focus, and mood and fights off the effects
of fatigue, though it is also used by athletes because of its
effects on motor speed, power, and endurance. However, as
with most things, moderation is important, for too much
caffeine will mimic the symptoms of anxiety, induce hand
tremor, and hinder performance in a variety of ways. Though
the cognitive, affective, and behavioral effects of caffeine are
addressed in turn in the following sections, it is important to
acknowledge that the distinct causal mechanisms for these
effects are often complicated by the fact that caffeine affects
several physiological systems simultaneously. In addition to
the central nervous system, each of the cardiovascular, muscu-
lar, pulmonary, hormonal, and metabolic systems are affected
by the presence of caffeine in the bloodstream.
Affective Effects of Caffeine
Coffee and tea are part of the morning routines for many
around the world because they help people to feel more
awake. In low to moderate doses, caffeine consumption
enhances mood, increases levels of self-reported alertness,
and decreases self-reported fatigue. While some effects of caf-
feine consumption decrease with increasing tolerance, this is
generally not the case with enhancements to mood and mental
alertness, though considerable evidence suggesting an interac-
tion effect of caffeine and expectancy may play a role in the
preservation of these effects.
The effect of caffeine on affect needs to be considered in
terms of at least four distinct but correlated constructs. Ener-
getic arousal (EA), indicated by such terms as alert, energetic,
and wide awake as contrasted to tired, sleepy, or drowsy, is
increased with caffeine. This is, after all, why coffee and tea are
consumed! Orthogonal to EA is Tense arousal (TA), which is
indicated by words such as tense, anxious, nervous, or afraid as
contrasted with calm, relaxed, or at ease. State levels of TA are
seen as a response to stressful situations, while more stable
individual differences (trait levels) are associated with anxiety
and neuroticism. TA increases with caffeine, particularly at
higher doses (i.e., >4 mg kg -1 ). Positive affect (PA), although
highly correlated with energetic arousal and indexed by words
such as cheerful, pleased, and happy, is an interactive effect of
caffeine and situational cue for happiness. Without such cues,
caffeine has a small positive effect on PA. The dimension of
negative affect (NA) as indicated by unhappy, depressed, and
blue, is highly positively correlated with TA and slightly nega-
tively correlated with PA. NA is less affected by caffeine than it
is by situational manipulations of affect (e.g., sad or depressing
movies or other negative mood inductions).
In higher doses, and among individuals with lower toler-
ances, caffeine also increases 'jitteriness, ' nervousness, and gen-
eral anxiety, and these side effects are more common among
clinically anxious individuals. Indeed, caffeinism is a recog-
nized anxiety disorder associated with high levels of caffeine
intake. The current diagnostic and statistical manual of mental
disorders (DSM-IV-TR) lists four caffeine-related psychiatric
disorders: caffeine intoxication, caffeine-induced anxiety disor-
der, caffeine-induced sleep disorder, and other caffeine disor-
ders not otherwise specified. Caffeine intoxication is shown by
having recently consumed more that 250 mg of caffeine and
showing at least five of the following symptoms: restlessness,
nervousness, excitement, insomnia, flushed face, diuresis, gas-
trointestinal disturbance, muscle twitching, rambling flow of
thought and speech, tachycardia or cardiac arrhythmia, periods
of inexhaustibility, or psychomotor agitation. For a diagnosis,
these symptoms must be causing clinically significant distress
or impairment of functioning. Caffeine-induced anxiety disor-
der includes symptoms of anxiety associated with high levels of
caffeine intake. Caffeine-induced sleep disorder is associated
with high caffeine consumption and sleep disorders not other-
wise explained. Because the symptoms of high levels of caffeine
428 Caffeine
consumption are similar to those of the anxiety disorders,
physicians are encouraged to inquire about the level of caffeine
intake for patients complaining about anxiety symptoms.
Behavioral Effects of Caffeine
Of course, many of these affective effects carry over into more
overtly physical behaviors. The most renowned behavioral
effect of caffeine is increased alertness. Performance on monot-
onous tasks that require detection of rare events (e.g., vigilance
tasks such as long distance truck driving or looking for weap-
ons in airport scanners) normally deteriorates over time. Mod-
erate doses of caffeine ameliorate this decrease. A more subtle,
but equally reliable effect is the speeding up of reaction time,
particularly when choice is not required.
In addition to these benefits, moderate doses of caffeine
enhance physical endurance by increasing both work output
and the time to exhaustion. Caffeine's effect on shorter-term
physical behavior is less well documented, perhaps because
shorter durations make effects more difficult to reliably detect.
Nevertheless, the effect of caffeine on physical exercise of short
duration suggests that peak power output, speed, and isoki-
netic strength are improved for very short bouts (lasting less
than 10 s). For longer bouts (greater than 15 s), which rely on
the glycolytic system, these same improvements are not main-
tained; in fact, some findings suggest a detrimental effect of
caffeine on power for bouts of 15 s to 3 min. These discrepan-
cies reflect the likelihood that the effects of caffeine result from
a combination of mechanisms at work simultaneously. In
addition to the central nervous system, these mechanisms
result from changes to the cardiovascular, muscular, pulmo-
nary, hormonal, and metabolic systems.
The effects of caffeine on physical performance have led to
its widespread use among elite athletes and performers, even
though the benefits are not well established across activities.
With respect to elite athletic performance, it is important to
note that very small variations in performance are the differ-
ence between being on the victory stand and being one of the
crowd. The unit of analysis is thus not the percent change in
performance so much as change as a percent of the coefficient
of variation in performance, that is, the size of the effect in
terms of the standard deviation rather than in terms of the raw
performance.
In an effort to maintain parity, the International Olympic
Committee classified caffeine as a doping agent in 1984 by
setting an acceptable threshold level of 15 pgml -1 in urinary
samples (the level was reduced to 12pgml _1 the following
year). In 2004, however, this prohibition was revoked when
doping for most international sporting events (including the
Olympics) fell under the administration of the world anti-
doping agency (WAD A). Under the current WADA policy, the
concentration of caffeine in the bloodstream is monitored
through urine samples for signs of potential abuse, though
the threshold level constituting abuse has not been officially
declared. In the past, threshold levels have been subject to
criticism on the basis that they fail to take into account high
interindividual variability in terms of caffeine metabolization
(several studies suggest a 15-20-fold range across individuals)
as well as the fact that most performance-enhancing benefits
of caffeine have been documented at levels consistent with
moderate daily use. Estimates vary dramatically depending on
many factors, but peak urinary caffeine concentrations of 12 pg
mf 1 would be expected to occur 105 min after a mean caffeine
intake of at least 10.5 mgkg -1 of body weight.
Cognitive Effects of Caffeine
The effects of caffeine on cognitive performance are not as
clear-cut as they are for feelings of alertness. They are complex,
for they depend upon the type of task, the situational
demands, and characteristics of the individual. To make it
more complicated, there is reliable evidence for systematic
interactions with the various combinations of person, situa-
tion, and task variables.
To understand these effects, it is useful to organize a num-
ber of cognitive tasks along three dimensions: the require-
ments for sustained attention, the requirements for working
or immediate memory, and the requirements for integrating
long-term memories with immediate memory. Tasks can be
high or low on each of these three dimensions. The positive
effects of caffeine on the first (sustained attention) are quite
clear, whereas effects on the latter two dimensions are more
complicated.
Sustained attention tasks such as long distance truck
driving, security scanning at airports, or other examples of
vigilance that require detection of rare events in the presence
of many repetitive but nontarget signals are correlated with
general alertness. Performance on these tasks is hindered by
manipulations that lead to lower alertness: alcohol, sleep dep-
rivation, the time of day (the optimal time depends upon the
participant), or the time on task. In particular, performance on
these tasks decays over time, with very good performance
upon starting with a (negatively accelerating) exponential
decay over time. Performance may be indexed by reaction
time or by accuracy. Reaction time increases and accuracy
diminishes over time. Caffeine as well as brief exercise, or
even an increase in the signal frequency inhibits this decay,
although effort instructions ('try harder' or 'do not go to sleep')
do not. The clearest evidence for caffeine effects on vigilance
are found following sleep deprivation and after a period of
time on the task. That is, caffeine helps the most when the
subject is most fatigued. These effects are associated with
increases in energetic arousal.
Although the effect of caffeine on vigilance is quite clear, the
pattern on working memory tasks is much less so. Part of the
confusion is that caffeine facilitates alertness which is, in turn,
related to detection of the material to be remembered. In
addition, caffeine enhances speed of performance which in
turn can reduce accuracy. Thus, in some studies, memory for
recent events is enhanced, probably because the detection was
better, rather than improvements in storage or retrieval. An
example of the complexity of the results may be seen in proof-
reading where caffeine hinders the detection of interword
errors (e.g., subject-verb agreement) which requires more
working memory than does the detection of intraword errors
(e.g., spelling or broken typography) which has a more com-
plicated relationship with caffeine (a small decrease in detec-
tion is associated with an increase in the speed of processing).
Information acquired when one is alert is recalled better
later (after at least a day) than is information acquired when
Caffeine 429
one is sleepy. This effect maybe shown using sleep deprivation,
the time of day, exercise, and variation in personality dimen-
sions associated with arousal (impulsivity and extraversion).
Similar effects are found with caffeine: caffeine taken when
learning new material facilitates long-term recall and recogni-
tion of that material.
For complex cognitive tasks, similar to the sort of ability
tests given for admission to graduate school, the effects of
caffeine are even more complicated and are an interactive
function of individual differences in impulsivity and the time
of day. In a series of studies using these complex tasks, it was
shown that caffeine facilitated the performance of subjects
thought to be less aroused (high impulsive participants in the
morning, low impulsive participants in the evening) but had
deleterious effects for those thought to be more highly aroused
(low impulsiveness in the morning, high impulsiveness in the
evening.)
The general conclusion for cognitive performance is that
caffeine facilitates performance on tasks that require (sustained)
attention, particularly among participants who are fatigued or
less aroused. The beneficial effects of caffeine are harder to detect
for alert subjects or for tasks that require more memory. For
especially complicated tasks with highly aroused participants,
caffeine can have a deleterious effect.
Caffeine as a Tool for Psychological Research
Much of the research on caffeine is just that: studies of the
effects of caffeine on various psychological variables. A less
recognized use of caffeine is as a tool for psychological
research. By increasing the ways in which a participant's
energetic and tense arousal may be manipulated, caffeine
allows the careful researcher to tease apart the effects of putative
arousal and motivational-related variables such as the time of
day, introversion-extraversion, impulsivity, exercise, or incen-
tives. For example, in the study of the association between
positive affect and energetic arousal, and negative affect and
tense arousal, the differential effect of caffeine on these vari-
ables as contrasted to more typical mood manipulations allows
for distinctions that would otherwise be difficult to achieve.
Conclusion
Caffeine, particularly in the form of coffee or tea, is used by a
majority of adults worldwide. The benefits on feelings of alert-
ness and positive affect, as well as on endurance and simple
cognitive performance are clear, particularly in situations that
would otherwise normally lead to fatigue or when sleep-
deprived. However, the benefits come with some cost, in that
performance on complex reasoning tasks can be hindered, and
that high doses lead to unpleasant levels of tension. Although
the effects of low to moderate doses seem positive, higher doses
can lead to discomfort and impaired cognitive performance.
Further Reading
Fredholm BB (2011) Notes on the history of caffeine use. In: Starke K, et al. (ed.)
Methylxanthines, vol. 200, pp. 1-9. Berlin: Springer.
James JE (2011) Caffeine. In: Johnson BA (ed.) Addiction Medicine, pp. 551-583.
New York: Springer.
Nehlig A (1999) Are we dependent upon coffee and caffeine? A review on human and
animal data. Neuroscience & Biobehaviorai Reviews 23(4): 563-576.
Career Development
S H Osipow The Ohio State University, Columbus, OH, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Career The succession of vocational activities through the
life span.
Negative reinforcer Environmental event that reduces a
positive response.
Occupational stereotype The image an occupation holds in
the view of society in general.
Positive reinforcer Environmental event that satisfies an
individual's needs.
Vocational maturity The degree to which an individuals
vocational activities are congruent with the expectations of
his/her life stage.
Vocational self-concept The individual's personal image in
terms of work.
How It Began
Career development involves a lifelong series of vocational
activities. It represents a series of work activities which may
or may not have continuity. It is important to recognize that
the work activities are a sum of specific tasks. The accumula-
tion of these tasks makes up what is called a job. The sum of
jobs engaged in represent a vocation. One's vocational his-
tory across the life span becomes a career. One can infer a
person's career only in retrospect. The process begins with a
decision about educational objectives. This results in a chain
of events which leads to vocational entrance. For some peo-
ple, the vocational activity that follows may have little vari-
ance. That is, the vocation chosen may not change much
over the course of a lifetime. For others, there may be con-
siderable change, as skills develop and opportunities present
themselves.
History of Career Development
The concept of career development is a relatively new one,
generally thought to have its origins in the early part of
the twentieth century. Prior to that time no one thought that
careers were developmental in nature. Careers, or rather jobs
and work, were thought to be discrete. In other words, for
upper class or professional workers, some training might
occur such as in law or medicine, after which the worker
would enter and remain in the vocation doing essentially the
same thing throughout the entire work life. For middle class
workers such as small business owners, jobs were largely inher-
ited. A family business, for example, a store, a farm, or a trade,
was usually passed along from parent to child. For lower class
workers there was little continuity from job to job. Work was
obtained where it could be found and when it was exhausted
jobs were sought elsewhere. While some transferability of skills
and experience might occur from one job to another, no one
thought that there was that significant continuity in work
tasks. The question of choosing a career was hardly considered.
Except for the professional level jobs, choice was not a factor to
be considered. Even in these limited circumstances individual
choices were likely to be somewhat constrained depending on
family background.
As the Industrial Revolution continued to develop, dis-
crete industries were developed. Except for the lowest skill
level jobs, some training and/or experience was needed for
entry and advancement. As industries became increasingly
differentiated, the degree to which individual choices were
possible increased. However, the choices were still con-
strained to a significant degree. Furthermore, what choices
existed were limited to males for the most part, since women
were not significant players in the work place outside of the
home. Except for careers in teaching or in nursing, women
did not have a similar range of choices available to them as
did men. Not until the late nineteenth century when office
technology created the typewriter and the telephone were
women seen in office jobs, and even then for a long time
men did those jobs because the work was considered to be
too hard for women.
In the early part of the twentieth century more attention
was given to the role that inherent abilities played in work
effectiveness, along with that individual interests might be
differentially satisfied in different kinds of work. Interests
and abilities were of increasing interest as methods to mea-
sure them were developed. As a result, the concept of fitting
people to jobs (e.g., round pegs for round holes) as Parsons
proposed, became a cornerstone in the creation of the concept
of career development. Institutions devoted to the 'test them
and tell them' approach to careers began to occur. However,
even then, the notion that there might be life stages or phases
which would significantly influence career directions did not
occur to those practicing career planning for men. Women
were still excluded for the most part from career planning,
again, except for nursing and teaching, typically seen to be
women's work.
World War I accelerated the attention paid to differential
psychology as applied to careers because the rapid mobiliza-
tion of the work force into a war machine made it necessary to
classify people (mostly men) according to their abilities in
order to assign them to tasks most closely allied to their poten-
tial. World War II further accelerated this process because the
number of candidates to be sorted was larger and the jobs more
diverse and specialized. Furthermore, the methods of assessing
people became more sophisticated.
With the advent of the 1920s and 1930s the measurement
of interests became more sophisticated and became an equal
430
Career Development 431
partner to the use of abilities in understanding the issues
underlying career choices and ultimate adjustment to work.
Life Stages and Career Theories
Using knowledge developed from differential psychology it
was but a small step to thinking about careers as associated
with life stages. Adding to differential psychology was the
knowledge based on psychometrics and developmental psy-
chology, and, thus, gradually the concept of career develop-
ment took shape. These developments largely took the form of
theoretical ideas about how people choose, implement, and
adjust to their career possibilities across the life span.
One of the first of these theories relied heavily on develop-
mental psychology. It focused on stages of development through
childhood, and adolescence into early adulthood, later extend-
ing into middle adulthood and into the retirement years. Career
development was seen to be a process of growth, stability, and
decline. The most influential advocate of this approach was
Donald E. Super. According to Super, the earliest life stage
critical to career development is exploration, during which
individuals become increasingly aware of their physical and
psychological environment in general and later as it applies
personally. This process leads to a series of vocational develop-
mental age related stages, each of which has tasks to be
mastered.
To the degree that these tasks, called vocational developmen-
tal tasks, are mastered, the individual is seen to be vocationally
mature. Later stages named by Super were crystallization (starting
at age 14-18), specification (18-21), implementation (21-24),
and stabilization (25-35). Less clearly described are the stages
after stabilization. To the degree that an individual successfully
negotiates the stages in an age appropriate way, he/she is seen to
be vocationally mature. The vocationally mature individual pre-
sumably has a smoother and relatively trouble free career pattern.
Adequate vocational development requires the individual to
fit his/her emerging self-concept with an increasing awareness
of the world of work. In order to include individual continuity
in career development, Super added the self-concept. The self-
concept involves the integration of experiences that give feed-
back to the individual in terms of his/her interests, abilities,
and values. The emerging self-concept leads to a more sophis-
ticated way of fitting round pegs into round holes, as suggested
by the early work of Parsons. In the approach proposed by
Super, the development of a person's self-concept provides
one-half of the equation while increasing knowledge about
the world of work provides the other half. Career choice is,
thus, the process of increasing the accuracy of self-concept
implementation, which in turn leads to career development.
Measuring Vocational Interests
Vocational interests and their measurement have played a
significant role in understanding the career development pro-
cess. Starting with the early work of E.K. Strong, the measure-
ment of interests has occupied the attention of investigators of
career development. The rationale for their importance is obvi-
ous: if one is able to implement one's career interests in work,
then the only missing variable is the assessment of one's abil-
ities. When interests and abilities do not coincide, the career
development process is derailed.
Holland’s Approach to Career Development
Another way to view the manner in which self-concept imple-
mentation occurs is the approach proposed by J.L. Holland to
understanding career development. Holland has described the
world of work as consisting of six occupational environments
(i.e., realistic, investigative, artistic, social, enterprising, and
conventional.) These environments correspond to six personal
types having the same names. As an individual matures, his/
her experiences lead to the development of a personality,
which can be described in terms of the six personal types. People
are seen to seek occupational environments that correspond to
their personal type. This idea is similar to the self-concept
implementation in work as described by Super. While Holland's
theory does not explicitly describe the way these personal types
occur, they do seem to have a developmental aspect.
Values and Career Development
The role of values in the career development process is another
significant variable. Values seem to be an intervening variable
having a shaping effect on interests. In other words, interests
first develop based on individual reactions to activities. There is
some reason to believe interests are shaped by success or failure
experiences. That is, as the individual gets feedback from the
environment about how his/her abilities interact with satisfac-
tion resulting from engaging in various activities, interests flow
from success experiences. At a later point in the process, the
notion that some activities have more intrinsic value than others
begins to play a role in the early expression of occupational
choices.
Life Stages
Developmental psychologists have postulated that human
development can be described in terms of stages of life. Specific
stages and their names vary according to the investigator.
Career psychologists have adopted some of the names and
stages and also have described stages in ways that apply specif-
ically to careers.
As noted earlier, the most prominent career development
stage theorist is Super. Knowledge of life stages can direct career
psychologists in useful ways. For instance, the career develop-
mental stage determines issues such as the content of the tasks
of career development that are to be faced, how they take shape,
and how they shift over time. Thus, an individual in the earliest
stage would largely be focused on information gathering and
integration, while at a later stage, say midlife, would be dealing
with the ability to maintain his/her career status.
The Theory of Work Adjustment
Rene Dawis and Lloyd Lofquist created a theory of work adjust-
ment which postulates that a hierarchy of personal needs
432 Career Development
develops as a function of the individual's interactions with the
environment. The outcome variable of interest in this theory is
the correspondence between work requirements and personal
requirements. The degree to which these are in accord results in
what the authors call work adjustment, which can be viewed as
a developmental process over the life span.
The theory offers a substantial number of psychometric
instruments designed to measure personal needs that are to
be potentially satisfied in work as well as measures to assess the
degree to which these needs are satisfied in different occupa-
tions. Where the individual's needs are satisfied in their occu-
pation, job tenure is predicted to be high. As the individual's
needs change through the life span, satisfaction in different
jobs changes, leading the individual to strive to alter his/her
occupation. People strive to maintain a correspondence
between their work needs and work requirements. The degree
to which these correspond is called work adjustment, which is
measured in terms of job tenure, satisfaction, and effectiveness.
Social, Economic, and Technological Influences on
Career Development
Not to be denied is the impact of society on how one's career
develops. This includes the state of the economy, social mores,
and technology. Thus, an individual from a simple, primitive
society would have fewer choices and less control over them
than an individual from a complex society.
Underlying the social context of choices is the state of the
economy. In a thriving economy the range of career choices
available exceeds that in a stagnant or declining economy. Not
only is the range of choice larger in an expanding economy, but
new industries are spawned, which, in turn, create new kinds of
work opportunities and eliminate other, older jobs. Thus, for
example, the invention of the cell phone has dramatically
changed the communications industry, lobs for telephone
operators have changed. Not only are there fewer of them,
but the nature of the work itself has changed. Office technology
has had other significant effects on jobs. There are fewer secre-
taries, and those jobs that remain are different than they were
formerly. There is less need for shorthand skills. The invention
of the personal computer has reduced the number of typists,
since workers can do their own typing. An examination of the
total range of occupations demonstrates how technology has
created new jobs in the health professions, among others.
Social conditions interact with the individual's life stage.
The state of the economy when one first enters the labor
market, affects the range of possible initial decisions. This
range, in turn, has a long-term effect on what follows in later
life, even when the economy may have changed significantly
Another example of how technology has changed career devel-
opment is the existence of the automobile. Before the automo-
bile, people were limited in the range of their choices to those
occupations which could be performed within a short distance
from their homes. After the car came into widespread use, peo-
ple were able to travel extensively and so they could live at some
distance from their place of employment. This made it possible
for more drastic changes in one's career than possible earlier.
Another example of how technology has effected career
development can be seen in how the computer has and
continues to create new kinds of jobs, and job settings. For
the first time since the industrial revolution, the computer
enables people to work at home, or to work at long distances
from their primary place of employment.
Social changes have affected careers in a multitude of other
ways. Women's entrance into the labor force in significant
numbers changed existing jobs by creating new careers (such
as child care) in large numbers, the need for married couples to
accommodate to each other's careers, and because two income
families resulted in greater affluence, more and different con-
sumer goods on which to spend this new money. Issues that
occur throughout one's career are subjectto various interventions.
Career counseling, an entirely new profession, has been devel-
oped to help people deal with the problems described below.
Issues Involved in Choice
Numerous events can cause difficulty in determining an
individual's initial career choice. Among those are personal,
familial, and society barriers to implementing a person's career
choice.
Barriers Blocking Entrance to Desired Vocations
People may encounter barriers to their entrance into desired
occupations. These barriers may include lack of appropriate
abilities, unrealistic interests, lack of resources for training, social
discrimination, and lack of family support, to name but a few.
Clearly, some of the barriers are external to the individual, such
as social, economic, and new technologies.
Other issues may interfere with advancement. These include
poor performance, limited job opportunities, interpersonal
barriers, and social discrimination, for example.
Issues that are involved in premature or untimely voca-
tional termination are sometimes within the individual and
others external. For example, one reason for termination may
include failure to meet employers' goals, physical or mental
limitation, employment in obsolete industries, and reaching
retirement age.
Comparison of Theories
Examination of the various theories reveals some major simi-
larities of their major concepts. For example, Super's notion of
self-concept implementation in work choice has some of the
overtones of Holland's concept of correspondence between an
individual's personal type and the type of the work environment
experienced. And both the Holland and Super approaches share
similarities with the Dawis and Lofquist notion of satisfaction/
satisfactoriness of jobs as related to the degree to which the work
setting reinforces an individual's primary psychological needs.
This trend toward similarity of concept and outcomes suggests
that the theoretical approaches, different as they may seem
superficially, are converging.
See also: Vocational Choice; Work Efficiency and Motivation.
Career Development 433
Further Reading
Bordin ES (1984) Psychodynamic model of career choice and satisfaction. In: Brown D
and Brooks L (eds.) Career Choice and Development. San Francisco, CA:
Jossey-Bass.
Campbell RE and Cellini JV (1981) A diagnostic taxonomy of career development
problems. Journal of Vocational Behavior 19: 175-190.
Crites JO (1973) Career Maturity Inventory. Monterey, CA: CTB/McG raw- Hill.
Dawis R and Lofquist L (1984) A Psychological Theory ol Work Adjustment.
Minneapolis: University of Minnesota.
Gottfredson LS (1981) Circumscription and compromise: A developmental theory of
occupational aspirations. Journal of Counseling Psychology 28: 545-579.
Holland JL (1997) Making Vocational Choices , 3rd edn. Odessa, FL: Psychological
Assessment Resources.
Mitchell AM, Jones GB, and Krumboltz JD (eds.) (1979) Social Learning Theory and
Career Decision Making. Cranston, Rl: Carroll.
Osipow SH (1990) Convergence in theories of career choice and development.
Journal of Vocational Behavior 30: 122-134.
Osipow SH and Fitzgerald L (1996) Theories of Career Development. 4th edn. Needham
Heights, MA: Allyn & Bacon.
Parsons F (1909) Choosing a Vocation. Boston: Houghton-Mifflin.
Roe A (1957) Early determinants of vocational choice. Journal ol Counseling
Psychology 4: 212-217.
Savickas ML and Lent RW (1994) Convergence in Career Development Theories.
Palo Alto, CA: CPP.
Super DE (1980) A life-span, life-space approach to career development. Journal oi
Vocational Behavior 16: 282-298.
Super DE, Starishevsky R, Matlin N, and Jordaan JP (1963) Career Development:
A Sell-Concept Theory. New York: College Entrance Examination Board.
Tiedeman DV and O'Hara RP (1963) Career Development: Choice and Adjustment.
New York: College Entrance Examination Board.
Catecholamines and Behavior
L A Trevino, K Lorduy, M Natishyn, A L Dougall, and A Baum 1 ', The University of Texas at Arlington, Arlington, TX, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Catecholamines A class of neurohormones that are
involved in various central and peripheral processes;
includes epinephrine, norepinephrine, and dopamine.
Dopamine A catecholamine that is an important
neurotransmitter in the central nervous system involved in
movement, mood, and learning.
Epinephrine A catecholamine that is a neurotransmitter
and hormone that is produced from activation of
the sympathoadrenal-medullary axis with its primary
effects in the periphery including increased heart rate,
dilation of blood vessels in the skeletal muscular system,
opening of airways, and constriction of blood vessels
in the skin.
Learning Involves establishing a conditioned response to a
stimulus that did not illicit the behavior prior to a series
of interactions providing a conditioned reinforcement
and/or punishment.
Norepinephrine A catecholamine that acts as a
neurotransmitter and hormone in the periphery, and is
produced from activation of the sympathoadrenal-
medullary axis. In the central nervous system,
norepinephrine acts as a neurotransmitter affecting many
behaviors such as mood, learning, and arousal.
Stress Is experienced when an organism perceives harm,
loss, or threat to their homeostasis or well-being. Activation
of the sympathoadrenal-medullary axis and release of
catecholamines along with activation of the hypothalamus-
pituitary-adrenal axis and release of glucocorticoids are the
primary physiological responses associated with stress and
drive many of the other physiological, cognitive, and
psychological changes that occur.
Introduction
There has been a substantial focus on the relationship between
catecholamines and behavior. Researchers have examined a
number of topics, including catecholaminergic pathways in
the brain that promote stereotypical behavior, motor abilities,
operant learning, psychopathology, and other higher-order
behavior. As newer methods have become available, we have
been able to understand behavior at a more basic level. Tech-
nology has allowed research to explore the functions of cate-
cholamines, how they are synthesized, how they influence the
function of particular areas in the brain to promote behavior,
and how imbalances in particular regions produce pathology
(e.g., Parkinson's disease, depression, schizophrenia). Com-
prehension of the relationship between catecholamines and
behavior is best facilitated using a step-wise approach. We
start with basic knowledge regarding catecholamines, and
then discuss how they are measured centrally and peripherally.
The role catecholamines play in stress, mood, learning and
reward, and movement is also discussed.
What are Catecholamines?
The most prevalent catecholamines are dopamine (DA), nor-
epinephrine (NE), and epinephrine (E). Dopamine acts as a
neurotransmitter in the central nervous system (CNS) and is
a precursor for NE and E. Epinephrine and NE are the primary
catecholamines responsible for peripheral changes. Norepi-
nephrine also acts as a neurotransmitter in the CNS. Before
discussing the widespread effects of catecholamines on behav-
r Deceased.
ior, the types of catecholamines and how they are synthesized
and transmitted, as well as their central pathways, are
described.
Dopamine
Dopamine is synthesized from a tyrosine molecule into
L-dihydroxyphenylalanine (l-DOPA) by tyrosine hydroxylase,
which is a rate-limiting enzyme that determines the overall rate
of DA synthesis and the subsequent synthesis of NE and E.
(Figure 1) Tyrosine hydroxylase activity is regulated by (1) the
quantity of catecholamines present, such that high levels of
catecholamines will inhibit the rate of tyrosine production,
and (2) the rate of neural activity. l-DOPA is then converted
into DA by the enzyme, L-aromatic amino acid decarboxylase.
The location of synthesis occurs in the presynaptic terminals
(i.e., the sender) of central dopaminergic neurons and begins
with the activation of these neurons through action potentials.
Dopamine is released into the presynaptic cleft following the
influx of calcium ions into the presynaptic terminal and
the synthesis of dopamine from tyrosine. Once released into
the presynaptic cleft, DA activates dopaminergic receptors
on the postsynaptic terminals (i.e., the receiver). Areas of high
dopamine activity are (a) the arcuate nucleus projecting to the
infundibulum of the hypophysis (pituitary gland), (b) the ven-
tral tegmentumal area (VTA) projecting to the limbic system,
and (c) the substantia nigra that projects to the striatum. These
differing projections are implicated in various behaviors such
as homeostasis, motivation, and movement (these behaviors
are further discussed later in this article).
Projections from the arcuate nucleus through the tubero hy-
pophyseal to the adenohypophysis (anterior pituitary gland)
comprise the tuberoinfundibular DA pathway. This system
serves the regulation of releasing and inhibiting factors
434
Catecholamines and Behavior 435
Figure 1 All catecholamines are derived from the precursor tyrosine.
Its conversion by tyrosine hydroxylase is the rate-limiting factor in
catecholamine synthesis and is highly regulated by catecholamine levels.
(hormones) that in turn synthesize their target hormones in
the adenohypophysis and release them into the bloodstream.
The hypothalamus exerts downstream effects from these hor-
monal pathways that include the regulation of food and water
intake and thermoregulation. Damage or dysregulation of the
tuberoinfundibular DA pathway therefore has serious conse-
quences. For example, prolactin, one of the hormones regulated
by this pathway, is responsible for lactation in nursing females.
The production of prolactin is inhibited by DA, and a condition
called hyperprolactinemia (too much prolactin) manifests when
this inhibitory mechanism is disrupted. Symptoms of hyperpro-
lactinemia for both sexes include inappropriate lactation, abnor-
mal menstruation, and sexual side effects.
Dopaminergic projections originating from the VTA to cor-
tical regions yield the ventral DA pathway. Additionally, there
are two closely related subsidiaries of this pathway, the meso-
limbic and mesocortical pathways. The mesolimbic pathway
originates in the VTA and makes projections to the nucleus
accumbens, septum, amygdala, and the hippocampus. The
amygdala is involved in the emotional evaluation of a stimulus,
whereas the medial prefrontal cortex modulates the mesolimbic
pathway and facilitates both reward- and punishment-related
motivation. Without adequate NE synthesis in the medial pre-
frontal cortex, dopaminergic transmission in the nucleus accum-
bens is abolished. This finding is supported by many studies that
have demonstrated the suppressive effects of NE depletion on
learned aversion behaviors and reversal learning. It is well estab-
lished that the dopaminergic projections from the nucleus
accumbens are responsible for reinforcement by activating
reward and punishment neural regions. It has more recently
been discovered that adrenergic projections from the medial
prefrontal cortex are a necessary forerunner in mediating this
function. The mesocortical pathway connects the VTA to the
frontal lobes, particularly the dorsolateral prefrontal cortices.
These pathways are implicated in cognition, motivation, and
reward circuitry. Particularly, psychiatric disorders, such as bipo-
lar disorder and schizophrenia, have been associated with dysre-
gulation of the ventral DA pathway.
The nigrostriatal or dorsal DA pathway originates from the
substantia nigra to the striatum (both regions that comprise
the basal ganglia). This pathway controls movement and dys-
function results in pathological conditions such as Parkinson's
disease. Furthermore, medications, such as antipsychotics (e.g.,
haloperidol or thorazine) can inhibit DA pathways and pro-
duce side effects such as tardive dyskinesia (Parkinson's-like
symptoms).
Dopamine serves both inhibitory and excitatory functions
in its pathways. For example, in the dopaminergic nigrostriatal
pathway, there are both inhibitory and excitatory synapses
with y-aminobutyric acid (GABA) neurons: an inhibitory effect
on the striatal GABA-ergic neurons of the lateral segment of the
globus pallidus and an excitatory effect on the striatal GABA-
ergic neurons of the medial segment of the globus pallidus and
the pars reticulate of the substantia nigra. Essentially, these
integrated networks provide a neural substrate for complex
behaviors in organisms.
Norepinephrine
Norepinephrine acts as a neurotransmitter and as a hormone.
The synthesis of NE occurs when DA is converted to NE by
dopamine (3-hydroxylase. This synthesis occurs in noradrener-
gic neurons in the central and sympathetic nervous systems
(SNS), as well as in the cells in adrenal medulla. Peripherally,
the main sources of NE are the sympathetic ganglia and the
adrenal medulla.
There is a dorsal and a ventral NE pathway in the CNS. The
dorsal NE pathway originates in the locus coeruleus (LC) and
projects to the hippocampus and neocortex. The LC also
has projections to the medulla oblongata that project to
the SNS. The dorsal pathway is involved in the activation
of the SNS following exposure to stressful stimuli. Conversely,
the ventral NE pathway originates in the lateral tegmental field
and projects to the diencephalon, particularly the hypothala-
mus. The major function of the central NE systems is to regulate
arousal and modulate reward circuitry. In particular, these
pathways constitute critical neural circuitry in reward and posi-
tive reinforcement. Many antidepressant drugs thus target
noradrenergic projections to alleviate symptoms of depression.
Additionally, nicotine stimulates significant increases in cate-
cholamine activity in noradrenergic projections, contributing
to its mood elevation properties.
Epinephrine
Epinephrine is synthesized from NE by the enzyme, pheny-
lethanloamine N-methyltransferase (PNMT). Synthesis of E
occurs primarily in the adrenal medulla and is released in
conjunction with NE following activation of the SNS. Eighty
percent of catecholamines released from the adrenal medulla
following activation of the SNS are E; however, only 10% of
436 Catecholamines and Behavior
circulating catecholamines are E, indicating that additional
sources of catecholamines lie outside the adrenal medulla.
While the concentration of PNMT is highest in the adrenals,
production of E still occurs in the rostral medulla oblongata.
Consequently, E is the least ubiquitous catecholamine in the
brain, and its neurochemical circuitry is isolated to the medulla
oblongata. Nevertheless, it is important to consider that it does
in fact exist in the CNS. Functionally, this region of the brain-
stem is particularly important in autonomic nervous system
modulation, heart rate, and respiration. Glucocorticoids are
often released in conjunction with catecholamines during
stress responses. Glucocorticoids increase the expression of
PNMT, thereby increasing the production of E.
Measurement of Catecholamines
Measurements of catecholamines can be used to determine the
role they may play in behavior. Several techniques are used for
measuring catecholamines, depending largely on whether cen-
tral or peripheral measures are desired. Animal models are
commonly used to directly measure catecholamines in the
CNS. Brain dissection methods and tissue punching allow for
rapid dissection of brain nuclei along major nerve terminal
fields; however, these methods for measuring catecholamines
and their neural pathways in the human brain are limited.
Positron emission tomography (PET) is a current tool used to
detect the radiofrequencies of neurotransmitters or the molec-
ular weights of chemical substances in the human brain
through the use of highly selective radiotracers. Otherwise,
the detection of catecholamines in humans is muddled by
the lack of available methods for distinguishing catecholamin-
ergic function in the brain from that in the periphery.
Catecholamines, especially in humans, are more com-
monly measured in the periphery. Several common techniques
are used for detection, including enzyme-linked immunosor-
bent assays (ELISA), radioimmunoassays (RIA), and high-
performance liquid chromatography (HPLC). ELISA and RIA
methods involve binding isolates of catecholamines to anti-
bodies and use either color change (ELISA) or radioactive
tracers (RIA) to detect levels. HPLC involves isolating catecho-
lamines from proteins using electrospray ionization time-
of-flight mass spectrometry and detection typically using an
ultraviolet detector. Although HPLC is very sensitive and selec-
tive, it is also complicated and time-consuming. In compari-
son, ELISA is also sensitive and can be readily performed,
especially by using commercially available kits.
Catecholamines and Stress
Catecholamines are an integral component of the stress
response, in particular the release of NE and E from
sympathoadrenal-medulliary (SAM) axis activation. From an
evolutionary perspective, the stress response is adapted out of
the necessity for additional resources (energy) to be present
in order to better prepare an individual to attend to a threaten-
ing stimulus or stressor. Following exposure to a stressful stim-
ulus, the paraventricular nucleus (PVN) of the hypothalamus
is activated. Activation of the PVN, in turn, activates two
neuroendocrine systems that release hormones into the
periphery. The hypothalamic-pituitary-adrenal (HPA) axis is
activated and releases corticosteroids from the adrenal cortex.
Additionally, the PVN has projections to the LC, that then
projects to the medulla oblongata, and ultimately to the SNS.
Once the SNS is activated, NE is released from the sympathetic
terminals that innervate certain vital organs, such as the heart
and lungs. Additionally, activation of the SNS also causes the
release of NE and E from the adrenal medulla. Circulating NE
and E then have effects on the same vital organs, innervated
by the SNS. The cells that comprise the organs in each system
have |3-adrenergic receptors or a-adrenergic receptors that bind
selectively to NE or E, producing a myriad of effects, depending
on the receptor subtypes. Although it is beyond the scope of
this article to describe each of the subtypes, a general descrip-
tion of these SNS effects is provided subsequently.
When NE is released from the sympathetic ganglia and both
NE and E are released from the adrenal medulla, these circulat-
ing levels of catecholamines can have profound effects on
many important systems. For example, when the SNS is acti-
vated following sttess, the release of these catecholamines
results in increased heart rate and strength of cardiac contrac-
tions. Additionally, there is an increase in blood pressure by
increased cardiac output and vasoconstriction of small arter-
ioles. This occurs when NE binds to P-adrenergic receptors
that are located on the cardiac muscle. This effect is further
reinforced by E binding to these same receptors. The release
of these catecholamines also leads to bronchodilation via
P-adrenergic receptors to increase respiration. Glandular activ-
ity during SNS activation is greatly decreased with the excep-
tion of sweat glands. Increases in sweat gland activity follow
activation of the SNS via a-adrenergic receptors. Additionally,
during the sttess response, NE and E influence smooth muscle
of the gastrointestinal and urogenital tracts. The smooth mus-
cles in these tracts will either contract due to activation of
oc-adrenergic receptors or relax due to activation of P-adrenergic
receptors. Metabolic changes occur due to activation of the
SNS, such that glucose and glycogen synthesis are favored
in the liver, and lipid oxidation is favored as a primary
function to provide energy to cells. The SNS activation also
causes dilation of the pupils, increased saliva production,
and inhibited digestion and sexual reflexes. Collectively,
these physiological changes prepare an organism to fight
the threatening stressor or flee from impending danger.
These adjustments are adaptive and motivate individuals
to function (physically and psychologically) effectively as
they interact in a changing environment.
Catecholamines and Mood
Depression
Norepinephrine has a significant impact on neural activity,
mainly through the extensive connections of the LC to various
brain regions involved in mood, arousal, and endocrine func-
tions. Epinephrine is involved in the peripheral response
to SNS activation that triggers the 'fight or flight' response.
Dopamine is involved in the reward pathway and movement.
Studies have identified changes in the function of these neuro-
transmitters in patients with depression and anxiety disorders.
Catecholamines and Behavior 437
Additionally, serotonin has been shown to be involved in the
pathophysiology of depression. There is evidence that in
depressed patients, altered autoreceptor functions impair NE
release, and that the response of second messenger systems to
NE binding is blunted. In addition, some patients demonstrate
blunted responses to exogenous activation of the SNS. These
changes are seen in peripheral tissues, but some have also been
identified in the frontal cortex of patients who committed
suicide. Unfortunately, evidence of these changes is inconsis-
tent in the literature, and there appears to be a distinction
between patients with depression alone versus those with
depression and anxiety or panic disorder. Preclinical studies
have attempted to replicate these findings in animal models
of depression that involve chronic mild and/or unpredictable
stress. Animals subjected to chronic stress demonstrate changes
in adrenergic receptor levels and functionality similar to those
seen in depressed patients, lending further credence to the role
of NE in depression.
Genomic studies have identified variants in some of the
genes involved in catecholamine synthesis and metabolism
that are associated with depression. Polymorphisms exist for
the genes encoding tyrosine, hydroxylase, and dopamine beta-
hydroxylase, and catechol-O-methyltransferase (COMT) as well
as the genes encoding D 2 and D 4 dopamine receptors. From this,
it is clear that genetic vulnerabilities exist for depression and
that future treatments could target specific deficiencies - a
model currently in use for chemotherapy drugs.
Several classes of drugs are used to treat depression: mono-
amine oxidase inhibitors, tricyclic antidepressants (TCA), selec-
tive serotonin reuptake inhibitors (SSRI), receptor antagonists,
serotonin-norepinephrine reuptake inhibitors (SNRI), and
norepinephrine-dopamine reuptake inhibitors (NDRI). Mono-
amine oxidase inhibitors, such as phenelzine, inhibit the action
of monoamine oxidase, which is an enzyme that breaks down
monoamines (NE, E, and DA) following release into the synap-
tic cleft. Tricyclic antidepressants like amitriptyline work by
blocking several neurotransmitter receptors, such as adrenergic,
noradrenergic, and serotonergic, to name a few. The SSRIs, such
as sertraline, work by blocking serotonin reuptake following
release from the presynaptic terminal of the neurons. The SNRIs
(i.e., duloxetine) and NDRIs (i.e., bupropion) work in similar
fashion to those of the SSRIs by blocking serotonin and NE, and
NE and DA, respectively.
Studies comparing antidepressants that target monoamine
functions have demonstrated similar efficacies for drugs that
target NE and/or DA versus those that are serotonin-specific.
Although older versions of these drugs were hindered by side
effects (mainly due to peripheral adrenergic activity and toxic-
ity due to monoamine oxidase deficiency), current treatments
are much safer and demonstrate fewer side effects. The only
concern when using antidepressants that target catecholamine
activity is that some patients may experience increased anxiety
as a result of enhanced NE binding.
Additional treatments are used for depression, including
electroconvulsive therapy (ECT) and deep brain stimulation
(DBS). Both of these therapies are invasive to the patients, so
they are used as a last resort when pharmacological and talk
therapies have not worked. Electroconvulsive therapy involves
inducing a seizure in a patient through the use of electricity
and requires several sessions in order to fully alleviate
depressive symptoms. The course of ECT begins with marked
improvements in appetite and drive that is believed to be
caused by an increase in DA receptor function. Following
additional sessions of ECT, patients experience an increase in
energy and attention, which is believed to be the result of an
increase in synaptic NE. The final sessions of ECT see improve-
ments in cognition and resolution of anxiety that is associated
with depression. This is believed to occur through increased
serotonin function. Deep brain stimulation involves the
implantation of an electrode into the brain and stimulates
that area with low-frequency electrical stimulation. This proce-
dure is a new treatment for depression, so the areas that are
stimulated are still being mapped. However, there has been
some success in relieving depressive symptoms by stimulating
Brodmann's Area 25. It is believed that this therapy works by
inhibiting this area of the brain, which causes increased activa-
tion in other areas of the brain that have been implicated in the
cause of depressive symptoms.
Through manipulation of the various neurotransmitter sys-
tems in the brain, alleviation of depressive symptoms is
achieved. However, it is believed that certain neurotransmitters
are responsible for the alleviation of certain symptoms. For
example, manipulation of the NE system may be related to
improvements in alertness, energy, and interest in life. Thera-
pies targeted to the DA system show improvements in atten-
tion, motivation, pleasure, and reward. Additionally, negative
affect is improved by changes in the NE and serotonin systems,
whereas loss of positive affect is improved by manipulations of
the NE and DA.
Current research is examining the role of HPA axis dysregu-
lation in depression, and novel targets for pharmacotherapy
have been identified in preliminary studies. Dysregulation of
the HPA axis leads to excess catecholamine release and could
cause the pathological changes in catecholamine functions
seen in depressed individuals. Altered catecholamine activity
and HPA axis dysregulation have both been found in studies of
suicide victims. However, catecholamine markers have been
found to differ between suicide victims and patients with
depression who are not suicidal. This suggests that altered
HPA axis and catecholamine functionality is a risk factor for
suicide in patients with depression but not an indicator of
depression alone. Continued research into the relationship
between catecholamines and mood disorders may shed light
on the underlying mechanisms that trigger the development of
these conditions.
Schizophrenia
Schizophrenia is a psychotic disorder that involves positive
symptoms, such as hallucinations, delusions, and disorganized
thoughts, and negative symptoms, including lack of speech,
cognitive deficits, and depressive symptoms. Dopamine has
been identified as one of the neurotransmitter systems that
contribute to this debilitating disorder. The different dopami-
nergic pathways contribute in different ways to both positive
and negative symptoms. For example, increased dopaminergic
activity in the mesolimbic pathway is thought to underlie the
hallucinations, delusions, and disordered thoughts that are
the positive symptoms. Additionally, decreased dopaminergic
activity in the mesocortical pathway and function in the
438 Catecholamines and Behavior
prefrontal cortex have been associated with negative symp-
toms. Memory deficits that are common in schizophrenic
patients may also be associated with decreases in D] receptor
transmission in the dorsolateral prefrontal cortex. With avail-
able technology and strides in the creation of animal models,
additional neurotransmitter systems, such as NE, serotonin,
GABA, and glutamate, have been implicated in maintenance of
schizophrenic symptoms.
It has been well established that an increase in DA in critical
sites in the brain play a major role in psychosis. However, there is
a growing interest in examining the role of NE dysfunction.
Recently, it has been found that individuals with schizophrenia
whose psychotic symptoms returned while on medication pro-
duced more dopamine-P-hydroxylase (the enzyme that synthe-
sizes NE from DA) than those that did not relapse. Also, increases
in NE were strongly correlated with increases in both negative
and positive symptoms in schizophrenic patients. Research
demonstrates that when NE is not synthesized proportionately
with DA, the emotional valence is increased. The disproportion
between DA and NE is thought to explain the extreme fear
experienced in schizophrenic individuals.
Treatments for schizophrenia fall into two categories of
drugs: typical and atypical antipsychotics. Typical antipsychotics,
such as Chlorpromazine, are used to treat positive symptoms,
and can exacerbate negative symptoms. Typical antipsychotics
are similar to DA and bind to D 2 receptors in the mesocortical
and nigrostriatal dopaminergic pathways in the CNS. Atypical
antipsychotics are more effective in eliminating negative symp-
toms; however, they do not have as high affinity for D 2 receptors
as typical antipsychotics. There has been some debate regarding
how to effectively treat negative symptoms of schizophrenia.
Recently, there have been some questions on the advantages
of treating these symptoms with atypical antipsychotic drugs
versus neuroleptics. These classes of drugs are not solely specific
to manipulation of dopaminergic pathways, but also improve
imbalances in the other neurotransmitters, such as serotonin,
glutamate, and GABA, that are associated with schizophrenic
symptoms.
The DA hypothesis of schizophrenia is further supported by
the fact that normal individuals will experience some psychotic
symptoms that mimic those involved in schizophrenia when
they take DA agonists such as cocaine and amphetamines.
Also, when the DA agonist, l-DOPA, is administered in Parkin-
sonian patients at too high a level, it can potentially induce
schizophrenic-like symptoms, including hallucinations, para-
noia, and the stereotypic behaviors. In another study, the ability
to modulate auditory filtering, a hallmark of schizophrenia was
lost when amphetamines were administered in an animal study.
Further, there have been studies that demonstrate the use of
monoamine-P inhibitors (mirtazapine) to inhibit DA that
could potentially help in treating the negative symptoms of this
disease. Together, this evidence supports the role of catechol-
aminergic function in the pathogenesis of schizophrenia.
Learning and Reward
Learning is associated with acquiring new skills and behaviors
through a series of stimulus-response associations. Recall of
these associations is important for the expression of future
behaviors and adaptations that are important for survival,
such as locating high-quality food sources and avoiding areas
where predators reside. The process of differentiating reward-
ing stimuli from those that have negative consequences aids in
the ability to make predictions about the surrounding environ-
ment and serves as a foundation for goal-directed behavior.
Catecholamines are intrinsically involved in learning as well as
reward and motivation, as demonstrated by studies employing
genetic knockouts, drugs that modify catecholamine synthesis,
metabolism, and functionality, and behavioral paradigms
designed to evaluate reward and aversion. Alterations in nor-
adrenergic and dopaminergic pathways produce changes on
behavior. For example, damage to the nigrostriatal areas
reduces eating and drinking behaviors, while damage to the
mesolimbic dopaminergic pathway causes deficits in movement
that would facilitate these goal-directed behaviors. The central
DA pathways facilitate learning and reward goal-directed behav-
ior, such as seeking out food, water, or dmgs, as demonstrated by
numerous studies that have observed increases in consumptive
behaviors with the administration of DA agonists. Additionally,
DA antagonists reduce or block the rewarding nature of food,
water, cocaine, and amphetamines.
Both noradrenergic and dopaminergic neurons fire when
paired with a novel or unpredictable stimulus, and their acti-
vation is inhibited by a predictable stimulus. Levels of these
catecholamines are also dose-dependent for learned behavior,
in that both high and low doses of NE impair learning. There-
fore, there is an optimal range to facilitate learning. The major
difference between NE and DA in learning appears to be that
NE is needed to start learning or changing of behavior. Its
continued release is not needed to sustain the expression of a
behavior, but DA is necessary to maintain the continued
expression of such behaviors. In contrast, behavioral suppres-
sion occurs when the DA or NE functioning is impaired.
Other neurotransmitters appear to exert effects on the cate-
cholamine pathways in learning, such as glutamate, GABA, and
even serotonin. Glutamate is highly important in learning, as
long-term potentiation is necessary for memory formation in
the hippocampus. Many of the dopaminergic projections
involved in learning interface with glutamatergic projections
to the hippocampus, and both NE and DA terminals impli-
cated in learning are regulated by glutamate receptor activity.
Serotonin can also moderate catecholamine-dependent behav-
ioral learning by changing the dose-dependence of catechola-
mines to initiate a learned behavior. Finally, drugs such as
cocaine and amphetamine activate these same pathways for
reward and learning. For example, amphetamine has a similar
chemical structure to that of NE and produces similar effects in
the reward and learning pathways.
Projections from the basal ganglia to the hypothalamus
play a role in conditioning responses by reinforcing behaviors
through stimulation of reward centers in the brain. The endog-
enous reward system yields a highly positive experience as
evidenced by numerous studies that demonstrate that animals
will voluntarily stimulate areas of their brain involved in
reward by pressing a lever that applies an electric current to
electrodes implanted on neurons in reward centers in the
brain. The use of self-stimulation paradigms revealed that elec-
trical stimulation associated with activation of central dopami-
nergic pathways is rewarding, despite the fact that no intrinsic
Catecholamines and Behavior 439
drive (such as hunger or thirst) is present. Additionally, admin-
istration of DA agonists increases self-stimulation and DA
antagonists decrease it. Under normal circumstances, an activ-
ity associated with the reduction of a biological drive state
would activate these same pathways, reinforcing the action as
well as environmental cues associated with the reward.
Drugs of addiction directly or indirectly act upon the cenUal
dopaminergic pathways to produce a pleasurable sensation
that is inherently rewarding. The response is so sttong that
the administration of drugs such as cocaine enhances self-
stimulation and reduces the stimulus intensity required to
reinforce this behavior. Unfortunately, repeated activation of
the reward pathway by cocaine or amphetamines alters the
postsynaptic response to dopamine release, and eventually,
tolerance develops to the initial dose of drug, and the original
euphoria can only be achieved through increased consump-
tion. Eventually, the brain adapts to the continual presence of
the drug, and addiction develops as an individual finds it
necessary to continue using in order to achieve normalcy
instead of euphoria.
Movement
The basal ganglia is an area of the brain responsible for the
organization, execution, and timing of voluntary movements,
such as writing on a chalkboard. The basal ganglia consist of
several different nuclei, including the caudate nucleus, puta-
men, globus pallidus, substantia nigra, and the subthalamic
nucleus. The circuitry of these different nuclei is responsible
for the execution of voluntary movement. (Figure 2) The sub-
stantia nigra has projections to the putamen via two different
dopaminergic pathways. The direct pathway that uses the Dj
dopamine receptors facilitates movement, whereas the indirect
pathway that uses the D 2 dopamine receptors, inhibits move-
ment. The direct pathway begins when the putamen receives
excitatory dopaminergic projections from the substantia nigra.
The putamen then sends inhibitory GABA projections to the
internal globus pallidus that sends inhibitory GABA-ergic pro-
jections to the thalamus. In the indirect pathway, the substan-
tia nigra sends inhibitory dopaminergic projections to the
putamen. The putamen sends inhibitory GABA-ergic projec-
tions to the external globus pallidus. The external globus palli-
dus sends inhibitory projections to either the internal globus
pallidus or the subthalamic nucleus. The subthalamic nucleus
sends excitatory glutamatergic projections back to the internal
globus pallidus. The internal globus pallidus sends inhibitory
projections to the thalamus. When projections from the inter-
nal globus pallidus reach the thalamus, the result is inhibition
of thalamic neurons. Both of these pathways from the subtha-
lamic nucleus are vital in the coordination of voluntary
movement.
Parkinson's disease is a debilitating disorder that is marked
by characteristic tremors at rest, difficulty initiating movement,
and rigidity in movement. Additionally, there can be some
emotional and cognitive deficits associated with the disease.
Parkinson's disease is characterized by a depletion of dopami-
nergic neurons in the substantia nigra that projects to the
putamen. Depletion of these neurons causes deficits in both
the direct and indirect pathways. Activation of the dopaminer-
gic neurons that contain D] receptors that project to the puta-
men from the substantia nigra causes a reduction in neuronal
activity to the internal globus pallidus. This ultimately leads to
increased neuronal activity in the projections from the internal
globus pallidus to the thalamus. When this occurs, there is
less inhibition of thalamic neurons. In the indirect pathway,
the inhibition of dopaminergic neurons from the substantia
nigra is greatly reduced, leading to increased neuronal activity
to the external globus pallidus. The increase in neuronal activ-
ity of the external globus pallidus leads to decreased neuronal
activity to the internal globus pallidus and the subthalamic
nucleus. The reduction in neuronal activity ultimately leads
to increased neuronal activity to the thalamus. The increases
in thalamic activity following destruction of the dopaminergic
neurons in the substantia nigra would lead to the characteristic
rigidity that is the hallmark of Parkinson's disease.
Putamen
Figure 2 Neural circuitry of movement. This is simplified to include the dopaminergic pathways discussed in the text. Solid lines reflect excitatory
activity and dashed lines reflect inhibition. SN, substantia nigra; STN, subthalamic nuclei; GPE, external globus pallidus; GPI, internal globus
pallidus; and D1/D2 refer to specific dopamine receptor types.
440 Catecholamines and Behavior
In order to treat Parkinsonian symptoms, several treatments
have been targeted to improve depleted DA levels in the basal
ganglia. l-DOPA, which is the precursor to dopamine, is com-
monly used in Parkinsonian patients to regain motor control.
l-DOPA influences the ability of damaged dopaminergic neu-
rons to metabolize DA along the nigrostriatal path, a pathway
largely involved in motor function. Recently, another DA ago-
nist, pramipexole has been experimentally challenged as an
adjunct to l-DOPA in order to improve motor deficits along
with the depressive behaviors (e.g., dysphoria, social withdraw,
cognitive deficits) that accompany the dopaminergic dysfunc-
tion associated with Parkinson's disease. Although l-DOPA is a
common method of treatment for Parkinsonian symptoms,
administration of this drug at too high a level can potentially
induce schizophrenic-like symptoms, including hallucina-
tions, paranoia, and the stereotypic behaviors. Deep brain
stimulation of the basal ganglia has shown remarkable and
immediate improvements in symptoms associated with Par-
kinson's disease.
Summary
The catecholamines, DA, NE, and E, are essential not only for
survival during times of crisis but are also involved in learning,
motivation, and reward processes that contribute to an organ-
ism's adaptive prowess. The actions of NE and E as part of the
body's stress response system trigger an immediate response to
threatening stimuli through increased arousal, priming of the
cardiovascular system, and changes in blood flow that favor
skeletal muscles over the organs of digestion. Mood is greatly
affected by NE and DA, and disruptions to these systems
can lead to depression, anxiety disorders, or schizophrenia.
Dopamine's role in motivation and reward enables an organ-
ism to identify pleasurable events and also learn to avoid
aversive conditions, a distinction that is vital for long-term
survival. In addition, the delicate balance of DA and NE in
the CNS is responsible for voluntary movement, and any dis-
ruption to these neurotransmitters can lead to disorders of
movement as seen in Parkinson's disease. Continued research
into the factors that influence both central and peripheral
catecholamines and disease states is essential for the creation
of novel therapies that could prevent dysfunction and deleteri-
ous consequences.
See also: Associative Learning; Bipolar Disorder; Classical
Conditioning; Depression; Hormones and Behavior; Operant
Conditioning; Schizophrenia.
Further Reading
Cooper JR, Bloom FE, and Roth RH (2003) Norepinephrine and epinephrine.
In: The Biochemical Basis of Neuropharmacology, 8th edn., pp. 181-223.
New York: Oxford University Press.
Cooper JR, Bloom FE, and Roth RH (2003) Dopamine. In: The Biochemical Basis of
Neuropharmacology, 8th edn., pp. 225-270. New York: Oxford University Press.
DeLong MR (2000) The basal ganglia. In: Kandel ER, Schwartz JH, and Jessell TM
(eds.) Principles of Neural Science, 4th edn., pp. 853-867. New York: McGraw-Hill.
Guyton AC and Hall JE (2006) Contributions of the cerebellum and basal ganglia to
overall motor control. In: Textbook of Medical Physiology, 11th edn., pp. 698-713.
Philadelphia, PA: Elsevier.
Hadley ME and Levine JE (2007) Catecholamines and the sympathoadrenal system.
In: Endocrinology , 6th edn., pp. 315-336. Upper Saddle River, NJ: Pearson
Princeton Hall.
Harley C (2004) Norepinephrine and dopamine as learning signals. Neural Plasticity
11: 191-204.
Johnson E0, Kamilaris TC, Chrousos GP, and Gold PW (1992) Mechanisms of stress:
A dynamic overview of hormonal and behavioral homeostasis. Neuroscience and
Biobehaviorai Reviews 16: 115-130.
Lyne J, Kelly BD, and O'Connor WT (2004) Schizophrenia: A review of
neuropharmacology. Irish Journal of Medical Science 173(3): 155-159.
Nutt DJ (2008) Relationship of neurotransmitters to the symptoms of major
depressive disorder. The Journal of Clinical Psychiatry 69{supp\emen\ El): 4-7.
Ressler KJ and Nemeroff CB (1999) Role of norepinephrine in the pathophysiology
and treatment of mood disorders. Biological Psychiatry 46: 1219-1233.
Central Auditory Processing Disorders
G A Stefanatos and A T DeMarco, Temple University, Philadelphia, PA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Amusia A selective impairment of the ability to
appreciate music related to damage to the central nervous
system.
Attention deficit hyperactivity disorder
A neurodevelopmental disorder characterized
by inattentiveness, hyperactivity, and
impulsivity.
Auditory agnosia A collection of disorders associated with
loss of the ability to recognize sounds that is not attributable
to sensory difficulties.
Auditory evoked potentials A series of minute electrical
potentials that reflect the neurophysiologic activity
generated in the auditory system in the course of processing
an auditory stimulus.
Auditory processing disorder (APD) Disorders of auditory
processing in children resulting from functional
compromise of the central auditory processes underlying the
analysis of auditory information.
Central deafness A broad term referring to conditions in
which the detection of sound is impaired due to dysfunction
of the central auditory nervous system.
Cortical deafness A relatively rare condition usually
associated with bilateral temporal lobe damage resulting in
functional deafness.
Developmental dyslexia or reading disorder (RD)
A neurodevelopmental disorder characterized by specific
deficiencies in decoding written language.
Dichotic listening A task that evaluates a listener's ability to
process competing signals by delivering different
information simultaneously to each ear.
Hyperacusia Heightened sensitivity to auditory stimulation
associated with an aversion to or intolerance for certain
sounds at their typical sound levels.
Planum temporale An area on the dorsal surface of the
superior temporal gyrus posterior to Heschl's gyrus thought
to play a role in processing the rapid spectrotemporal cues in
sound, including speech.
Pure word deafness A subtype of auditory agnosia
characterized by severe difficulties in understanding spoken
language with intact speech production, nonauditory
language, and relative preservation of nonverbal sound
recognition and identification.
Specific language impairment (SLI) A neurodevelopmental
disorder characterized by problems in understanding or
producing language in the absence of causal factors such as
hearing impairment, mental retardation, neurological deficits,
primary emotional disorder, or social deprivation.
Tinnitus The perception of sound in the absence of an
external acoustic stimulus.
Introduction
The complexity of the central auditory nervous system (CANS)
is unrivaled by other human sensory systems. Comprised of
multiple parallel pathways consisting of many intricately
interconnected nuclei, the system is capable of processing infor-
mation with remarkable speed and efficiency. While informa-
tion processing in other sensory modalities usually occurs on a
timescale of milliseconds, the CANS is capable of coding time-
based acoustic cues on the order of microseconds. Detailed
knowledge of the function of the CANS has been slow to emerge
and correspondingly there is also limited understanding of the
central auditory processing disorders (CAPDs) that arise when
its operation is impaired by disease, damage, or maldevelop-
ment. Nevertheless, significant progress has been made in recent
years both in recognizing these disorders and in developing a
workable framework for their diagnosis and treatment.
The term CAPD encompasses a broad range of deficits that
occur when neural processes underlying the analysis of audi-
tory information by the brain are functionally compromised.
Currently, it is defined by deficiencies in one or more of the
following auditory behaviors: sound localization and laterali-
zation, auditory discrimination, auditory pattern recognition,
auditory temporal processing (e.g., resolution, integration,
masking, ordering), and performance with competing or
degraded acoustic signals. Given the paramount role of hearing
in everyday human communication, CAPD often comes to
attention as difficulties in processing speech.
We begin this overview of CAPD with a brief synopsis of
the organization and function of the CANS. This is followed
by a summary of acquired syndromes that arise with lesions to
the CANS. We then discuss idiopathic and developmental
forms of CAPD and conclude with a discussion of treatment
approaches.
Neural Architecture of Central Auditory Processing
The three primary functions of the auditory system are to detect
sounds, localize their source, and determine their identity and
meaningfulness. For the brain to interpret what we hear, sound
vibrations must first be transduced into patterns of neuronal
excitation. This is accomplished by ~3500 inner hair cells
(IHCs) (see Figure 1(a)) that are organized in a single row
on the ribbon-like basilar membrane that runs down the
cochlea of the inner ear. Analogous to the arrangement of the
strings on a piano, there is an orderly relationship between the
frequency of a sound and the location of IHCs on the basilar
441
442 Central Auditory Processing Disorders
(b) Brainstem (c)
Coronal section
^ Medial
geniculate
nucleus
secondary
^auditory
cortex
B Inferior
colliculus
Lateral
lemniscus
Cochlea
Superior
* olive
Inner
hair cell
Axons forming
Primary
auditory
f cortex
Ventral
cochlear
nucleus
Organ of Corti
Outer
membrane
Caudal
midbrain
Auditory radiations
Medulla
-i
nucleus
Tectorial
membrane
Rostral
midbrain
Pons
Figure 1 (a) Schematic drawing of a cross-section of the cochlea showing inner and outer hair cell on the basilar membrane. The fine filaments
protruding upward from inner hair cells toward the tectorial membrane represent clusters of cilia of varying length. Movement of a bundle back and forth
causes cycles of depolarization and hyperpolarization of the hair cell. These occur at a preferred time during the stimulus wave cycle resulting in pulsatile
release of neurotransmitter and phase locking of the postsynaptic potentials of spiral ganglion cells, (b) Schematic diagram of the major nuclei of the
ascending pathways. Colored lines represent the chief ipsilateral and contralateral pathways from the left cochlea to the auditory cortex. The same
pathways from the right cochlea are not depicted here to allow a clear view of the anatomy of the brain stem. Similarly, no attempt is made to illustrate
subdivisions of nuclei and connections within regions that transmit information both ipsilateral and contralateral to the ear of stimulation, (c) Coronal
view of the brain illustrating the location Heschl’s gyrus and secondary auditory cortex on the dorsal surface of the superior posterior temporal lobe.
Courtesy of Patricia Hearons.
membrane that a sound maximally stimulates. This tonotopic
organization enables the cochlea to act as a frequency analyzer,
essentially decomposing spectrally complex sounds into con-
stituent frequencies.
Stimulated IHCs release neurotransmitters that are detected
by receptors on axons of nearby inner ear spiral ganglion cells,
causing action potentials to be propagated along these axons
toward the brainstem. Approximately 30000 of these fibers
comprise the bulk (95%) of the eighth cranial nerve, also
known as the vestibulocochlear nerve (VN). Given that each
spiral ganglion nerve cell axon carries action potentials origi-
nating from stimulation of a single IHC, the VN is also tono-
topically organized. Indeed, tonotopy is preserved up to and
including auditory cortex.
The frequency of a stimulus is coded not only by the loca-
tion of the stimulated IHC on the basilar membrane, but also
in the timing of the intervals between spikes. The firing rate of
IHCs encode the intensity of sound which, up to a certain
sound level (~50 dB), translates to an increase in the discharge
rate and the number of spiral ganglion cells firing. Thus, critical
information on the frequency, intensity, and timing of a stim-
ulus is coded in activity of these afferents to the brainstem.
Ascending Auditory Pathways
VN fibers enter the brainstem at the posterior pontomedullary
junction, bifurcate, and synapse with neurons of the cochlear
nucleus (CN), the gateway to the CANS (see Figure 1(b)).
Neurons of the CN begin the process of extracting and coding
features of acoustic input, setting the stage for more refined
processing of acoustic features at progressively higher levels of
the CANS.
From the CN, coded impulses ascend the CANS in multi-
ple, parallel, and hierarchically organized processing pathways
that carry information regarding different aspects of sound
to the auditory cortex. Figure 1 (b) provides a highly simplified
schematic representation of the flow of information. In reality,
the system is exceedingly complex, involving at least nine nuclei
in the ascending system. Each major division is characterized by
a distinctive composition of cell types functionally specialized
for performing particular computations on sensory input.
As shown in Figure 1(b), ascending fibers leave the CN and
decussate, sending information to both the ipsilateral and
contralateral superior olivary complex (SOC). Since each
SOC receives input from both ears, critical binaural analyses
Central Auditory Processing Disorders 443
occur here relating to sound localization. Neurons in the
lateral superior olive (LSO) process interaural sound level
differences (ILDs), a cue used in the spatial localization of
high-frequency sounds. By contrast, cells in the medial supe-
rior olive (MSO) assist in sound localization by detecting
microsecond level disparities in the arrival time of sound at
the two ears (intra-aural timing differences, ITDs). Overall,
approximately five nuclei of the ascending CANS are involved
in some aspect of sound-source localization.
Nearly all of the ascending auditory fibers converge in the
central nucleus of the inferior colliculus (ICC) in the midbrain.
The IC plays a critical role in integrating spectral and temporal
information in order to localize sound sources and code dyna-
mically varying stimulus attributes, such as changes in fre-
quency and amplitude.
Two nuclei, the external nucleus of the IC (ICX) and the
dorsal cortex of the IC (DCIC), comprise components of the
so-called nonclassical or extralemniscal auditory system. This
system receives its primary input from the ICC and follows a
parallel course to the classical auditory pathway up to the
thalamus, synapsing with the dorsal and medial parts of the
medial geniculate body (MGB). The pathway then diverges
from the classical pathway, which projects to primary auditory
cortex (PAC). Instead, the extralemniscal system projects to
secondary and association cortices and to regions not directly
contacted by the classical auditory pathway, such as the baso-
lateral nucleus of the amygdala and the cerebellum. Neurons
of the nonclassical system differ from neurons of the classical
auditory pathways in that they are not as sharply tuned to
stimulus features and they respond to input from the auditory,
visual, and somatosensory systems.
In the classical auditory system, the main target for projec-
tions of ICC neurons is the ventral nucleus of the MGB (MGv).
From the MGv, projections go on to the PAC via the auditory
radiations. These connections allow for extensive convergence
and divergence of information. A single cortical locus can
receive inputs from large regions of the thalamus and con-
versely individual axons can establish synaptic contacts span-
ning up to 7 mm of cortical surface.
Cells in MGv demonstrate greater selectivity of responses to
stimulus features than neurons lower in the system. Many
appear unresponsive to simple acoustic stimuli but show vig-
orous responses to specific parameters of complex time-varying
stimuli (e.g., direction or rate of frequency change). The
responses of neurons in MGv, to some degree, are tuned by
cortical influences (directly or through the thalamic reticular
nucleus) aimed at optimizing the processing of dynamic
aspects of auditory input used to identify, localize, and ascribe
meaning to sounds.
Auditory Cortex
Many computational transformations related to the processing
of an auditory stimulus are completed before activation
reaches the PAC, located on Heschl's gyrus in the superior
temporal lobes (see Figure 1(c)). Nevertheless, further proces-
sing and integration occurs at the cortical level, evidenced by
sensitivity of various cortical fields to a range of basic stimulus
attributes such as frequency (tonotopic organization), band-
width, intensity level, spatial location, and movement or
motion. Some response properties of PAC neurons are inher-
ited from thalamic inputs while others undergo significant
transformation or are generated intracortically. While most
neurons in PAC are binaurally influenced, 10-20% show no
evidence of binaural interaction.
Surrounding the PAC are so-called belt and parabelt
regions, which are responsible for secondary and tertiary levels
of cortical processing, respectively. While PAC responds to
all types of auditory stimuli, secondary and association audi-
tory cortex in anterolateral and ventral auditory areas show
particular sensitivity to spectrally and temporally complex
sounds, including speech. These areas form part of a hierarchi-
cally organized processing stream involved in auditory object
and speech perception. This auditory analog of the so-called
'what' pathway in vision includes connections from anterior
belt areas directly to ventrolateral prefrontal cortex (PFC). By
contrast, a posterodorsal stream, analogous to the 'where'
pathway in visual system, is involved in the perception of
auditory space and motion as well as aspects of processing
speech and language. This pathway extends from PAC to the
posterior superior temporal gyrus, inferior parietal cortex, and
eventually projects to dorsolateral PFC.
Experience plays an important role in determining the
functional circuitry of the auditory cortex. Processing in the
auditory cortex is dynamic, with reorganization occurring to
optimize the apprehension of relevant stimuli. This plasticity is
reflected in cortical tonotopic maps that can adaptively change
in response to reinforcement, or fill in areas that may be
otherwise deprived of input from cochlear lesions.
Ontogenetically, synaptogenesis of PAC peaks at 50 months
of age and myelination of thalamocortical projections con-
tinues until ~ 6 years of age. Hence, auditory cortex function
continues to evolve throughout early to middle childhood,
maturing late relative to other sensory cortex. Many cortically
mediated aspects of central auditory function are dependent on
processing by secondary and association cortex and their intra-
and interhemispheric connections. Consequently, some func-
tions of the CANS do not fully mature until adolescence.
The auditory cortices of humans possess some hemispheric
specialization of function. Auditory processes distributed in
the left cerebral hemisphere appear to be particularly adept at
processing rapid dynamic changes (up to a few tens of milli-
seconds) in the spectrotemporal form of sound, whether or not
stimuli are speech. The right hemisphere appears to preferen-
tially process stimulus features that vary more slowly or require
integration over longer periods of time (e.g., melody, prosodic
contours in speech, emotions in voice).
Efferent Auditory Pathways
The CANS contains an extensive network of structures and
pathways that allow higher order auditory processing centers
to exert 'top-down' influences over the activity of lower nuclei.
The auditory corticofugal system (ACS) takes the form of des-
cending cortical projections that reach every major element
in the auditory system, extending as far as the cochlear hair
cells. This system forms important feedback loops designed
ultimately to adjust and improve input to the cortex for signal
processing. The ACS temporarily or permanently influences the
filtering, tuning, and response plasticity of the CANS.
444 Central Auditory Processing Disorders
CAPD: Diagnostic Considerations
Individuals diagnosed with CAPD present with a variety of
symptoms that can range in severity from a lack of responsive-
ness to sound (cortical deafness) to a constellation of more
subtle problems reflecting impaired listening skills or difficul-
ties discriminating slight differences in sounds. The latter pro-
blems include difficulties attending to and remembering orally
presented information, following multistep directions, and
perceiving speech when it is presented in a noisy environment
(e.g., classroom, cocktail party), at a rapid rate, or when signals
are degraded (e.g., indistinctly articulated, transmitted via
intercoms, public address systems, cell phones).
In considering the diagnosis, it is necessary to establish that
a peripheral hearing loss cannot account for a presenting prob-
lem. Quantification of peripheral hearing sensitivity typically
entails standard audiometric testing, such as pure tone and
speech audiometry. This may be combined with immittance
audiometry, otoacoustic emissions and possibly brainstem
auditory evoked potentials. It is also necessary to evaluate
potential 'top-down' contributions to auditory processing pro-
blems such as difficulties with attention, memory, linguistic
processing, or other specific cognitive domains. The diagnosis
of CAPD therefore often requires multidisciplinary input.
Three complementary approaches are used to identify
the presence of deficits consistent with CAPD. The most com-
mon initial approach is behavioral assessment, which exam-
ines a listener's performance on structured auditory tests that
evaluate several dimensions of processing commonly impaired
individuals with CAPD. Behavioral assessments can be supple-
mented, when necessary, with electrophysiological testing
measures and/or neuroimaging techniques. The essentials of
these approaches are discussed in the following sections.
Behavioral Assessment
Behavioral assessments utilize psychoacoustic measures and
norm-referenced, performance-based tests to identify deficien-
cies in auditory skills linked to CAPD. There is no standard
approach to the assessment of CAPD, although some guidelines
have been established. The tests can be conceptualized as falling
into five broad categories of auditory behavior. A number of the
measures commonly used in assessing CAPD are listed in Table 1
along with the auditory skill to which they broadly relate. It is not
possible to precisely match the behaviors associated with CAPD
with specific measures in a one to one correspondence, in part
due to the multifactorial nature of task demands.
The choice of diagnostic measures is driven by a careful
review of the patient's history and presenting problems, but
also by the characteristics of the tests themselves, such as the
quality of the normative data and the discriminant validity.
Details of the specific nature of the presenting problem(s),
onset, progression, consistency, and the circumstances of man-
ifestation help to define the scope of an evaluation. In addi-
tion, it is necessary to consider variables such as age, gender,
medical and psychological history, and the presence of coex-
isting conditions or concerns.
Implicit in the overall approach is the notion that different
patterns of auditory processing impairment can emerge depend-
ing on the location, extent, and nature of compromise of the
CANS. Pathophysiology of the brainstem frequently results in
problems of binaural convergence and integration. As a conse-
quence, conditions such as multiple sclerosis or brainstem stroke
are associated with highly abnormal performance on tests exam-
ining sensitivity to ILD or ITD, due to both the need for integra-
tion and the microsecond level timing requirements for accurate
ITD perception. Interestingly, sound localization can also be
impaired in conditions such as albinism where the development
of decussed pathways to the SOC is atypical.
By contrast, patients with unilateral lesions often but
not invariably retain some ability to process either ILD or
ITD measures, although their performance may differ from
normal controls. Left hemisphere lesions can result in particu-
lar difficulty with ITD judgments. However, unilateral patients
are more likely to have difficulty localizing sounds in the
vertical plane or when spatial localization cues must be
integrated with mental representations of space (as in free-
field localization). They may have particular difficulty in loca-
lizing sound sources in the side of space contralateral to their
lesions. This is most strikingly evident following right parietal
lobe damage which acutely can result in auditory neglect.
Individuals with neglect may be unresponsive to voices and
environmental noises emanating from the left side of space or
they may demonstrate alloacusis, systematically localizing
these sounds to the right side of space.
Auditory discrimination is a fundamental skill necessary for
a listener to consistently assign meaningful properties to an
acoustic object. Auditory discrimination is typically tested by
prompting the listener to decide whether two stimuli are the
same or different, or by matching an auditory stimulus to one
of several pictures of acoustically confusable objects. Stimuli
may be contrasted in a number of domains including pitch,
duration, intensity, and phonetic cues.
Dysfunction of the CANS often results in difficulty discrim-
inating brief auditory temporal events in sound. These difficul-
ties commonly become evident in poor speech perception,
because phonemic discrimination and identification requires
resolution of brief or rapidly changing acoustic cues in the
speech signal such as consonantal bursts, formant frequency
transitions, and voice onset time. These cues occur on the order
of milliseconds to tens of milliseconds.
Discriminating the duration of particular cues is essen-
tial for recognition of voicing contrasts in stop consonants
(e.g., /p / vs. /b/, /t/ vs. /d/, /k/ vs. /g/), the timing of silent
bursts in fricatives and vowel length. Resolution of rapid
changes in frequency is essential for distinguishing acoustic
cues related to the place of articulation in stop consonants
and manner of articulation in stops and glides. Problems dis-
criminating these temporal cues lead to errors in discrimina-
tion even though speech audiometry thresholds are normal.
Basic temporal resolution is commonly measured using
psychoacoustic tasks such as click fusion, click differentiation,
or gap detection. In click fusion, very brief (< 1 ms) spectrally
broadband stimuli (clicks) are presented in rapid succession
separated by a short intervening period of silence. When the
silent interstimulus interval is extremely brief, the temporal
boundary between each click is insufficient to result in separate
percepts, so listeners fuse two clicks and report hearing only a
single click. Extending the duration of the interstimulus inter-
val beyond a threshold value results in the perception of two
Central Auditory Processing Disorders 445
Table 1 Summary of auditory behaviors and assessments for CAPD
Auditory behavior Assessment measure
1. Localization and lateralization
2. Temporal processing
Resolution, integration,
discrimination, temporal
ordering
3. Auditory discrimination
4. Processing low redundancy
or degraded auditory signals
(separation, closure)
Time compressed,
filtered, interrupted,
competing with noise
5. Processing auditory signal
with competing acoustic
signals (binaural competition)
Consonant-vowel
syllables, digits, words,
and sentences
6. Auditory pattern recognition
Interaural intensity difference, interaural time difference (e.g., Kaga et at, 1997 )
Click fusion, click differentiation (e.g., Lackner and Teuber, 1973; Stefanatos et al., 2005 )
Auditory fusion threshold test (McCrosky and Keith, 1996)
Within and between channel gap detection (e.g., Phillips, 1999; Stefanatos et al., 2007 )
Random gap detection test (Keith, 2000)
Gaps in Noise Test (Musiek et al., 2005)
The Duration Pattern Test (DPT) (Musiek et al, 1990)
Pitch Pattern Sequence Test (PPS) (Pinheiro.l 977)
Newcastle Auditory Battery (Griffths et al., 2001)
Phoneme discrimination (Wepman and Reynolds, 1986, Benton et al., 1983)
Test of Auditory Perceptual Skills (TAPS-R) (Gardner, 1997) [Word Discrimination]
Environmental sound discrimination (e.g., Schnider et al., 1994 )
Multiple Auditory Processing Assessment (MAPA) [Monaural-Selective Auditory Attention Test (M-SAAT)]
(Schow et al., 2007)
MAPA [Speech in Noise for Children and Adults (SiNCA)] (Schow et at, 2007)
Test for Auditory Processing Disorders in Children (SCAN-C) [Filtered Words, Auditory Figure Ground] (Keith, 2000)
Performance Intensity for Phonetically Balanced Words (PIPB) (Boothroyd, 2008)
Speech in Noise (SPIN) (Bilger et al., 1984)
Bamford-Kowel-Bench speech in noise test (BKB SIN) (Nyquette, 2005)
Willeford Battery [Competing speech, filtered speech, binaural fusion, and rapidly alternating speech]
(Willeford,1 977)
Compressed Speech Test (Beasley et al., 1972)
Synthetic Sentence Identification (SSI-ICM) (Speaks and Jerger, 1965)
MAPA [Dichotic Digits Test, Competing Sentences] (Schow et al., 2007)
SCAN-C [Competing Words] (Keith, 2000)
Hugdahl Dichotic Listening Test (Hugdahl and Andersson, 1986)
Staggered Spondaic Words (SSW) (Katz, 1968)
MAPA [Pitch Pattern Test, Auditory Fusion Test-Revised, Duration Patterns Test, Tap Test] (Schow et at, 2007)
Frequency (Pitch) Pattern Test (Musiek, 1994).
Montreal Battery of Evaluation of Amusia (MBEA) (Peretz et al., 2003)
Note: "research articles, [italics] designates subtests.
Lackner JR and Teuber HL (1973) Alterations in auditory fusion thresholds after cerebral injury in man. Neuropsychologia 11(4): 409-415.
McCroskey RL and Keith RW (1996) Auditory Fusion Test -Revised (AFT-R). St. Louis: Auditec.
Keith RW (2000) Random Gap Detection Test. St Louis: Auditec.
Stefanatos GA, Braitman LE, and Madigan S (2007) Fine grain temporal analysis In aphasia: Evidence from auditory gap detection. Neuropsychologia 45: 1127-1133.
Musiek FE, Shinn JB, Jirsa R, Bamiou D, Baran J, and Zaida E (2005) GIN (Gaps-ln-Noise) Test performance in subjects with confirmed central auditory nervous system involvement.
Ear and Hearing 26(6): 608-61 8.
Musiek FE, Baran JA, and Pinheiro M (1990) Duration pattern recognition in normal subjects and patients with cerebral and cochlear lesions. Audiology 29: 304-313.
Pinheiro M (1977) Tests of central auditory function in children with learning disabilities. In: Keith RW (ed.) Central Auditory Dysfunction, pp 223-256. New York:
Grune & Stratton.
Griffiths TD, Dean JL, Woods W, Rees A, Green GGR (2001) The Newcastle Auditory Battery (NAB): A temporal and spatial test battery for use on individual naive subjects. Hearing
Research 154: 165-169.
Wepman JM and Reynolds WM (1987) Auditory Discrimination Test, 2nd ed. Los Angeles, CA: Western Psychological Services.
Schnider A, Benson DF, Alexander DN, and Schnider-Klaus A (1994) Non-verbal environmental sound recognition after unilateral hemispheric stroke. Brain 117(2): 281-287.
Katz J (1968) The SSW Test: An Interim Report. Speech and Hearing Disorders 33: 132-146.
Musiek FL (1994) Frequency (pitch) and duration pattern tests. Journal of the American Academy of Audiology 5(4): 265-268.
Schow RL, Seikel JA, Brackett JE, and Whitaker MM (2007). Multiple Auditory Processing Assessment (MAPA) Test Manual Version 1.0. St. Louis, M0: AUDITEC.
Keith RW (2000) SCAN-C: A Test for Auditory Processing in Children. San Antonio, TX: The Psychological Corporation.
Jerger J and Speaks C (1965) Method for measurement of speech identification (A). Journal of the Acoustical Society of America 37(6): 1205.
Boothroyd A (2008) The performance/intensity function: An underused resource. Ear and Hearing 29: 479-491.
Bilger RC, Nuetzel JM, Rabinowitz WM, and Rzeczkowski C (1984) Standardization of a test of speech perception in noise. Journal of Speech and Hearing Research 27: 32-48.
Beasley DS, Schwimmer S, and Rintelmann WF (1972) Intelligibility of time- compressed CNC monosyllables. Journal of Speech and Hearing Research 15: 340-350.
Nyquette PA (2005) BKB-SIN Speech-in-Noise test. Elk Grove Village, IL: Etymotic Research.
Willeford JA (1977) Differential diagnosis of central auditory dysfunction. In: Bradford L (ed.) Audiology: An Audio Journal for Continuing Education, p. 2. New York:
Grune and Stratton.
Hugdahl K and Andersson L (1986) The ‘forced-attention paradigm’ in dichotic listening to CV-syllables: A comparison between adults and children. Cortex 22: 417-432.
Gardner M (1997) Test of Auditory Perceptual Skills (TAPS-R). Hydesville, CA: Psychological and Educational Publications.
Kaga K, Shindo M, and Tanaka Y (1997) Central auditory information processing in patients with bilateral auditory cortex lesions. Acta Otolaryngolica (Supplement)
532: 77-82.
446 Central Auditory Processing Disorders
distinct clicks. For normal hearing adults, this threshold is in
the order of 2-3 ms. By comparison, several studies have
shown that patients with bitemporal lesions require prolonged
periods of silence (15-300 ms) between two clicks before they
are able to reliably discriminate one from two clicks. In rare
instances, temporal lobe lesions can result in palinacousia, in
which the percepts of sounds continue or recur beyond their
cessation, leading to an echoing effect.
Gap detection is a similar but potentially more refined
measure in which listeners must identify the presence or
absence of brief gaps of silence inserted between two sounds
(e.g., tones, noise bursts) lasting tens to hundreds of millise-
conds. This paradigm allows the examination of the effects of
gap duration as well as the influence of the frequency and
duration of the sounds that precede and follow the gap (called
markers). When the frequency characteristics of markers differ
sufficiently, determination of whether or not a gap is present
requires integration across frequency-specific channels. This
so-called central gap detection task requires longer gap dura-
tions for detection of the intervening silent interval.
Gap detection is an auditory capacity that can be impaired
with bilateral temporal dysfunction but also in individuals
with aphasia secondary to left hemisphere stroke. Interestingly,
impaired gap detection performance in children can be predic-
tive of poor language development. Other temporal tasks sen-
sitive to cortical lesions include the ability to judge the
temporal order of rapid tonal sequences and the direction of
rapid frequency sweeps.
Individuals presenting with possible CAPD commonly
complain of difficulties with listening to speech in noisy envir-
onments. This is often a specific instance of more general
problems detecting and processing a sound when it occurs in
the presence of competing sounds. Formal assessment may
involve the determination of masking level differences (thresh-
old for tone detection in the presence of noise). Alternatively, a
more direct approach would entail the presentation of speech
(e.g., words) embedded in noise or measured amounts of
background 'cocktail party' conversation. Interestingly, the
use of a linguistically relevant background stimulus such as
cocktail party chatter is more effective a masker than speech
spectrum noise, white noise, or narrow band noise.
Many speech-in-noise or 'auditory figure-ground' tests of
central auditory processing utilize low-redundancy monaurally
presented stimuli. The low redundancy limits a listener's ability
to utilize compensatory mechanisms or strategies, while mon-
aural presentation allows comparisons of the performance of
the left and right ear. Comparing the ears can sometimes be
informative since cortical lesions tend to impair performance
in the ear contralateral to the lesion more so than the ipsilat-
eral ear. However, relatively depressed performance can be
observed in the ear ipsilateral or contralateral to subcortical
lesions. While these tests are sensitive to functional compro-
mise of the CANS, they are generally limited in their localiza-
tion value. Other low redundancy monaural stimuli include
low pass frequency-filtered speech, time-compressed speech,
and time-expanded speech.
While auditory figure-ground measures assess the ability to
hear target sounds in the presence of competing noise pre-
sented within the same ear, other measures assess the ability
to isolate and separate different streams of information
presented to both ears. Dichotic tasks evaluate a listener's
ability to process competing signals by delivering different
information concurrently to each ear. A dichotic words test,
for example, may present a different word simultaneously to
the left and right ear and require a listener to report both words
if he/she can (free recall), just the stronger percept, or just the
word from a predesignated ear.
In the vast majority of neurologically healthy right-handed
individuals, free recall of verbal stimuli is better in the right ear
than the left, reflecting hemispheric specialization of the con-
tralateral (left) hemisphere for language processing. Unilateral
temporal lobe damage commonly results in difficulties proces-
sing speech presented to the ear contralateral to the lesion.
Indeed, patients may demonstrate extinction of the ear contra-
lateral to the lesion, failing to recall most if not all of the
stimuli presented to that ear. A related task, the Staggered
Spondaic Word test, entails presentation of temporally over-
lapping bisyllabic compound words presented to the two ears
(e.g., hotdog in one ear, baseball in the other, with only dog
and base presented simultaneously).
One of the higher level demands on the auditory system
involves auditory pattern recognition. Because of the complexity
of tests of temporal pattern recognition, they are highly sensitive
to a wide variety of cerebral lesions, including temporal, tem-
poroparietal, frontoparietal, insular cortex as well as diffuse
cerebral damage. These tasks are less sensitive to damage to the
brainstem. Due to their differential sensitivity, such factorially
complex tests are often useful for screening purposes.
Like speech processing, musical ability may be selectively
impaired in neurological disease. The recent development of
assessment measures specifically designed to examine the
breakdown of musical ability provides an important expan-
sion of the repertoire of measures used to evaluate pattern
processing.
Electrophysiological Assessment
Electrophysiological measures examine the electrical potentials
produced by the central auditory system in response to audi-
tory stimulation. Auditory brainstem responses (ABR) have
become a mainstay of the electrophysiological assessment of
central auditory function. These far-field electrical potentials,
recorded with electrodes placed on the vertex of the head and
the mastoids, reflect neural activity generated by the auditory
system in a 10-ms period following presentation of a stimulus
(usually clicks) over headphones. The earliest components
(Waves I and II) of this multiwave response occur within the
first 3 ms and are thought to represent electrical activity gener-
ated in the distal and proximal portions of the VN, respectively.
Wave III arises at about 4-5 ms from second-order neuron
activity in or near the CN, while Wave IV likely reflects third-
order neurons located in the vicinity of the SOC with possible
contributions from the CN and the nucleus of the lateral
lemniscus. Finally, Wave V, which follows Wave III by about
2 ms, is believed to originate in the vicinity of the IC, although
it too likely has contributions from multiple anatomic struc-
tures. The actual site of generation of Waves VI and VII remains
unclear, but they are believed to be thalamic in origin.
The procedure can disclose dysfunction of brainstem audi-
tory nuclei and their afferents. For example, the existence of
Central Auditory Processing Disorders 447
eight cranial nerve tumors can be detected with ABRs in ~ 90%
of cases. Specifically, Wave I is present but subsequent compo-
nents (Waves III or V) are absent ( 10-20% of cases) or delayed.
A prolonged interval between Wave I and Wave III is also
common (40-60% of cases).
Later potentials in the range of 20-70 ms post stimulus
comprise the so-called middle-latency responses. These poten-
tials reflect a combination of diencephalic and cortical genera-
tors. They are small potentials that have received variable use in
clinical settings.
Cortical dysfunction can be assessed using long-latency
auditory event-related potentials (ERPs) such as mismatched
negativity (MMN). This complex, evident between 150 and
250 ms poststimulus, reflects automatically generated preatten-
tive cortical responses to a detected change in a stimulus. A late
positive wave occurring at 300 ms, called the P300, can be an
excellent neurophysiological marker of attention to auditory
events and processes related to updating of working memory.
Auditory steady-state responses provide a useful method of
assessing the processing of rapidly changing auditory features
such as amplitude and frequency modulations.
Structural and Functional Neuroimaging
Recent advances in magnetic resonance imaging permit the
acquisition of very high resolution structural images of the
brain and brainstem which correspondingly has improved
the capacity to detect small lesions. Contrast agents such as
gadolinium can further enhance detection of small lesions or
masses (e.g., tumors) which can be missed with electrophysio-
logical procedures alone or MRI without contrast. However,
small lesions can still be missed with structural neuroimaging.
In the past decade, there has also been a dramatic increase
in the use of functional magnetic resonance imaging (fMRI)
which can examine metabolic changes in oxygen utilization
in different areas of the brain related to neural activity pro-
duced in the performance of auditory tasks. Recent advances
in the production of nonmagnetic transducers (e.g., electro-
static, piezoelectric) allows good quality sound in the
MRI environment and sparse sampling sequences allow pre-
sentation of stimuli at a time when noise generated by the
scanner during image acquisition does not interfere with
perception.
Central Deafness
Central deafness (CD) is a broad term referring to conditions
in which partial or complete loss of hearing occurs due to
damage or dysfunction of the CANS. The model for this class
of disorders is cortical deafness, a relatively rare condition that
typically results from bilateral lesions of auditory cortex or
auditory radiations. The impairments associated with cortical
deafness are summarized in Table 2. Patients initially show
substantial elevations of hearing thresholds when assessed
with pure tone and speech audiometry, yet have normal or
near normal function of the middle and inner ear as indexed
by tympanograms, stapedius reflex, and otoacoustic emissions.
Brainstem auditory evoked potentials also commonly reveal
evidence of normal peripheral hearing, while middle-latency
and cortical evoked responses are either variable, abnormal or
absent. These cases emphasize the importance of objective
electrophysiological assessments to accompany standard
behavioral audiometry.
Deficits in hearing sensitivity are accompanied by profound
difficulties in understanding or responding consistently to
both speech and environmental sounds. While appearing
deaf, some subcortically mediated reflex responses, such as
orienting toward a sudden loud sound may be preserved.
Hearing sensitivity sometimes improves over time, but gener-
ally remains clinically elevated. In some cases, the speech of
affected individuals acquires the inflection and prosodic char-
acteristics typically associated with profound peripheral
hearing loss.
The most severe and persistent forms of cortical deafness
arise from subcortical white matter lesions, involving the tha-
lamocortical auditory radiations. These strategically placed
lesions result in total or near total interruption of thalamocor-
tical projection fibers from the medial geniculate bodies to
primary and secondary auditory cortex (likely disrupting both
classical and nonclassical auditory pathways).
Bitemporal lesions can result from a variety of circum-
stances but when caused by stroke, the history often reveals
consecutive cerebrovascular events. An initial unilateral tem-
poral lobe lesion may result in a selective cortical auditory
processing disorder (APD) such as an auditory agnosia for
environmental sounds, word sound deafness (WSD), or apha-
sia (discussed in later sections of this article). Cortical deafness
Table 2 Summary of impairment in classical syndromes of CAPD
Function
Cortical
deafness
Word sound
deafness
Word meaning
deafness
Nonverbal auditory
agnosia
Receptive
amusia
Speech comprehension
V ✓ ✓
✓ ✓ y/
V ✓ y/
®
®
Speech repetition
✓ ✓
y/ y/ y/
®
®
®
Writing to dictation
✓ ✓
y/ y/ y/
®
®
®
Spontaneous speech
y/ y/
®
®
®
®
Reading comprehension
S >
®
®
®
s
Spontaneous writing
S
<3
®
®
®
Recognition of familiar nonspeech sounds
Recognition of melody
✓ ✓
V ✓ ✓
<3
®
y/ y/
y/ y/ y/
y/ y/
y/ y/ y/
Hearing sensitivity
V y/ y/
®
®
®
®
Severity of impairment: S, none/mild; * ✓ , moderate; ✓ </ , Severe. Expanded from Bauer RM and McDonald CR (2003) Auditory agnosia and amusia. In: Farah MJ and
Feinberg TE (eds.) Patient-Based Approaches to Cognitive Neuroscience , pp. 257-270. Cambridge, MA: MIT.
448 Central Auditory Processing Disorders
emerges when a second lesion occurs some time later and
causes damage to PAC or subcortical afferents in the contralat-
eral hemisphere. The syndrome can also emerge in children
and has been described in association with a severe closed head
injury and herpes encephalitis.
CD has also been described in individuals with subcortical
lesions involving the MGB and the IC. Subcortical forms of CD
can occur from a variety of etiologies including stroke. This
generally entails a single event involving the vertebrobasilar
artery system which perfuses the pons, midbrain, and thala-
mus, rather than the consecutive middle cerebral artery strokes
implicated in cortical deafness.
Since virtually all ascending and descending auditory path-
ways synapse in the IC, bilateral lesions that result in complete
destruction of the ICs should, in theory, result in complete
deafness, since such a lesion presumably isolates the auditory
cortex from information encoded at the ear. However, avail-
able clinical reports present a mixed picture. In two case
reports, IC lesions resulted in complete deafness, while in
three other reports, there was mild elevation of pure tone
thresholds, but profound impairment of word recognition.
In these cases, pure tone perception may have been mediated
by the extralemniscal auditory brainstem pathway, whereas
functional integrity of the IC is critical for speech processing,
given its role in processing frequency modulation and other
dynamic acoustic features.
General Auditory Agnosia
The term agnosia is derived from the Greek words for 'not
knowing.' Auditory agnosia refers to a collection of disorders
associated with the loss of the ability to recognize auditory
objects that cannot be attributed to sensory difficulties. Audi-
tory agnosia can be fairly selective, affecting particular classes
of auditory input. Specifically, auditory agnosia can preferen-
tially disrupt the processing of verbal information, music or
environmental sounds. These specific forms of the disorders
are discussed below. The term general auditory agnosia refers
to a condition in which a neurologically based auditory recog-
nition disorder is evident to both speech and nonverbal stimuli.
Unlike cortical deafness, the processing impairment is not
associated with clinically significant audiometric thresholds.
Verbal Auditory Agnosia
Word Sound Deafness
Word deafness is characterized by severe difficulties in under-
standing and reproducing spoken language, with otherwise
intact speech production and nonauditory language compre-
hension (e.g., reading). In the context of marked deficits in
processing speech sounds, words, phrases, and sentences, there
is relative preservation of musical appreciation and the recog-
nition of environmental sounds. The deficiencies in decoding
language occur at an early stage of analysis and are relatively
specific to the auditory modality. Given this, the condition has
more recently been called WSD to distinguish it from other
forms of verbal auditory agnosia, such as word meaning deaf-
ness (discussed below).
Phenomenological descriptions have implicated a range of
disturbances of perceptual processing. In some cases, particu-
larly those with bitemporal lesions, the dissolution of auditory
perceptual processing is sufficient in severity to impede recog-
nition of the acoustic characteristics of the human voice. Speech
is described in such terms as 'a noise,' 'a garbled sound,' such as
'wind in the trees' or 'the rustling of leaves.' In others, spoken
communications are recognized as speech, but a breakdown
occurs in the process of mapping speech sounds to appropriate
stored mental representations. Discourse sounds like 'jabber-
ing,' 'a foreign language' or simply does not 'register.' There are
frequent suggestions that perceptual resources are incapable of
keeping up with the rate at which speech is produced: 'words
just run together' or 'come too quickly.'
Some studies have noted systematic error patterns in speech
discrimination and identification that implicate problems at
phonetic levels of analysis, a stage of auditory processing spe-
cific to speech. Problems particularly impede the discrimina-
tion of stop consonants (e.g., /p/, /b/, /t/, /d/, /k/, /g/) while
differentiation of vowels is relatively spared. This pattern of
phonemic imperception is common in WSD but not necessar-
ily specific to it.
Several reports have suggested that the symptoms charac-
teristic of WSD are secondary to fundamental problems in
basic aspects of auditory temporal processing. Problems
noted in WSD include deficient intensity-duration functions
and poor resolution of temporally distinct auditory events
(click fusion, gap detection). In addition, individuals with
WSD have difficulty detecting frequency modulations in
sound and in judging the temporal order of sequential audi-
tory stimuli. Based on such findings, some have argued that
WSD is associated with general limitations in auditory tempo-
ral resolution that are particularly detrimental to language
comprehension because they impede the ability to perceive
brief spectrotemporal cues in speech necessary for the deriva-
tion of linguistic meaning.
There is disagreement regarding the anatomical basis of the
syndrome. The majority of cases reported in the literature
demonstrate bilateral temporal lobe lesions, particularly
involving the mid portion of the superior temporal gyrus.
However, WSD was first described in patients considered to
have deep lesions confined to the left posterior temporal
regions and a number of case reports have confirmed that a
strategically placed unilateral left hemisphere lesion is suffi-
cient to result in WSD. Unilateral lesions can impede the
analysis of formant frequency modulation in speech sounds
without causing impairment on basic temporal resolution
tasks such as click fusion. WSD can also occur developmen-
tally, and is the most common pattern of impairment asso-
ciated with the Landau-Kleffner syndrome, a rare epileptic
syndrome in childhood.
Word Meaning Deafness
Individuals with word meaning deafness have severe difficulties
in understanding auditory language, yet can repeat and write to
dictation sentences that they cannot comprehend. Curiously,
they can also read and understand those sentences that they
could write to dictation but not comprehend. The preservation
of repetition and writing to dictation distinguishes this
Central Auditory Processing Disorders 449
condition from WSD and suggests that the analysis of phono-
logical information is relatively intact. Rather, the breakdown in
language comprehension appears to occur at the level of lexical
access, possibly secondary to a failure to match the intact pho-
nological code with corresponding representations in the lexi-
con or as a 'post access' failure of lexical items to activate
corresponding representations in the semantic system.
Phonagnosia
Phonagnosia refers to disorders of familiar voice recognition.
These individuals experience problems identifying coworkers,
friends, and relatives by the sound of their voice. This deficit
was first noted in patients with right hemisphere lesions who
had difficulty recognizing the voice of famous individuals. A
follow-up study revealed that temporal lobe damage of either
hemisphere impaired the ability to discriminate famous voices,
while deficits in discriminating and recognizing familiar voices
was correlated with right parietal lobe damage.
Problems with familiar voice recognition typically occur in
the context of other disorders of auditory processing, such as a
broader auditory agnosia, amusia, or word deafness. However,
some case reports suggest that it can occur in the context of
unimpaired recognition of environmental sounds. Phonagnosia
and amusia may be more closely tied, as both depend on audi-
tory mechanisms required to distinguish the timbre of a sound. A
developmental variant of the disorder has been described.
Nonverbal Auditory Agnosia
Sensory Amusia
Receptive amusia is associated with deficits in the ability to
appreciate various characteristics of music such as pitch, har-
mony, and melody. While the disorder can occur as a fairly
specific impairment dissociated from the perception of speech
and environmental sounds, it is more commonly seen in the
context of a broader nonverbal auditory agnosia. It has also
been reported in cases of aphasia, suggesting that the left and
right cerebral hemispheres contribute differentially to the mul-
ticomponent processes involved in music appreciation.
The right cerebral hemisphere appears to play an important
role in the processing of some aspects of music such as the
appreciation of melody. The left cerebral hemisphere appears
particularly important for the processing of sequential, tempo-
rally organized information included in a series of musical notes.
The left hemisphere's role may therefore increase with greater
analytic analysis or with more attention to temporal patterns
such as timing and rhythm. For example, there are reports of
professional musicians who suffer from an amusia subsequent to
a left temporoparietal stroke. These individuals may show nor-
mal performance on all aspects of melody processing but lose
the ability to discriminate or produce rhythms. This arrhythmia
is specific to auditory input and cannot be attributed to the
impairment of supramodel temporal processing.
Other clinical studies also suggest a distinction between
time-dependent processes governing melody recognition,
such as the appreciation of rhythm or duration and the more
instantaneous features of music, such as the pitch, harmony,
timbre, and intensity. For example, double dissociations have
been described between pitch perception and the processing of
temporal sequences. Similarly, the ability to appreciate expres-
sive patterns of music indexed by pitch, timbre, and intensity
(musical connotation) can be dissociated from the ability to
appreciate the significance of lyrics (musical denotation). It has
been further suggested that timbre-dependent skills such as
recognition of specific voices and musical instruments can be
distinguished from pitch-dependent skills, such as recognition
of tunes.
A developmental form of amusia occurs in ~4% of the
population. These individuals have difficulty discriminating
pitch changes in melodies that cannot be explained by periph-
eral sensory impairment. The disorder, referred to as tone
deafness or congenital amusia (CA), is frequently manifested
by an individual's inability to carry a musical melody but
seems unrelated to problems with musical time.
There is compelling evidence suggesting that CA is based in
a congenital musical pitch disorder. The disorder is thought to
have a genetic contribution insofar as in the families of indivi-
duals with CA, 39% of first-degree relatives have the disorder
whereas only 3% have it in control families.
Neuroanatomical studies have suggested that CA is asso-
ciated with abnormal gray and white matter in auditory cortex
and the inferior frontal cortex. Neurophysiological evidence
derived from recording auditory ERPs, has suggested that
these individuals are capable of near-normal processing of
musical pitch incongruities, suggesting that, similar to the
visual phenomenon of blindsight, subjects with CA can pro-
cess stimuli yet not have conscious access to that information.
Their behavioral deficits therefore appear related to a limited
awareness of pitch representations and a disconnection of
these representations with their knowledge of pitch.
Environmental Sound Agnosia
Some clinical reports suggest that auditory agnosia can be
restricted to environmental sounds. One case report with con-
secutive hemorrhagic lesions (left then right) sustained bilateral
damage in the region of the putamen extending to the thala-
mocortical auditory radiations. Following the second infarct,
the patient experienced a brief period of cortical deafness asso-
ciated with bilateral sensorineural hearing loss of 55-100 dB
at 1 kHz, in the context of normal brainstem-evoked responses.
At this time, she could not distinguish broad categories of
sounds, such as speech, music, or environmental sounds. How-
ever, this resolved over a 3-month period and was followed
by a reportedly selective auditory agnosia for environmental
sounds. Errors appeared discriminative rather than associative
in nature. Processing of speech was relatively intact at this
point. Unfortunately, neither this report nor previous reports
directly compared environmental sound recognition to the
perception of music. Consequently, they serve mainly to illus-
trate that environmental sound recognition can be impaired
with relatively intact perception of speech.
Cortical Motion Deafness
A recent study described a patient with a selective deficit for the
reception and discrimination of auditory motion. The disorder
emerged following resection of the right anterior temporal lobe
450 Central Auditory Processing Disorders
and posterior temporal gyrus. While the patient could correctly
identify and localize auditory objects in space, he failed to
reveal normal responses indicative of the ability to process
auditory movement.
Idiopathic Acquired CAPD
Hyperacusis
Hyperacusia is a term used to refer to raised sensitivity to
auditory stimulation often associated with an aversion to or
intolerance for certain sounds at their typical sound levels. A
person with severe hyperacusis may have difficulty enduring
the sound of everyday objects, such as a vacuum cleaner, a hair
dryer or a food blender, indicating that they are unpleasantly
loud when they are easily tolerated by others. While hypera-
cusis can occur with peripheral damage, it can also result from
dysfunction or damage to the CANS. It is thought that efferent
mechanisms from the SOC to the cochlea involving the audi-
tory nerve (olivocochlear bundle) may be functioning abnor-
mally in a proportion in these individuals.
It has been suggested that hyperacusis can be a symptom of
a serotonin dysfunction, leading to suggestions that selective
serotonin reuptake inhibitors (SSRI) may be helpful in treating
the disorder. Though some case studies report improvement of
hyperacusis from SSRIs, it has yet to be robustly clinically
validated. Other etiologies associated with hyperacusis include
exposure to certain medications or drugs, head trauma, and
migraine. It is commonly described in children with autistic
spectrum disorders and William's Syndrome. Finally, hypera-
cusia tends to accompany tinnitus.
Tinnitus
Tinnitus is the perception of sound in the absence of an exter-
nal acoustic stimulus. It is often described as persistent ringing
or buzzing in the ear and is commonly associated with pres-
bycusis, a hearing loss due to aging. Indeed by 70 years of age,
over 1 0% of all adults report episodes of tinnitus. Of those who
experience tinnitus, 79% report hearing sounds like tones,
while 21% report hearing a hissing or rushing noise. It is also
a symptom of Meniere's disease and is common in individuals
with tumors (acoustic neuroma or vestibular Schwannoma)
growing in the space between the pons and cerebellum.
Tinnitus may additionally occur as a result of exposure to
loud noise (e.g., artillery fire, rock concerts) or after the use
of certain antibiotics.
In ~60% of tinnitus cases, there is no known cause.
Because of the association between cochlear pathology and
tinnitus, it has been assumed that tinnitus is caused by hyper-
activity of cochlear hair cells. It has often been compared to
chronic pain, specifically phantom limb pain. However, evi-
dence such as the persistence of tinnitus after cochlear ablation
and a strong central effect of ototoxic medication suggests a
central origin for persisting tinnitus.
Recent research has implicated the nontraditional extralem-
niscal auditory pathway in the disorder. Because the thalamic
nuclei (medial and dorsal nuclei of the MGB) of the extralem-
niscal system project directly to the basolateral nucleus of the
amygdala, involvement of this pathway may be fundamental
to understanding why some individuals experience tinnitus as
a debilitating problem with affective symptoms.
Due to the commonalities between hyperacusis and tinni-
tus, SSRIs have also been proposed as a treatment for the latter.
However, a recent randomized placebo-controlled study found
that most subjects showed no consistent benefit from SSRIs.
Etiologic Heterogeneity/Diversity
Acquired forms of CAPD such as CD and the agnosias typically
represent the sequelae of ischemic or hemorrhagic cerebrovas-
cular accidents. Such syndromes can also result from a variety
of other conditions such as head injury, herpes simplex
encephalitis, tumor, and epilepsy. Less severe forms of CAPD
have been described in association with aphasia, encephalitis,
seizures disorders, brain tumors (e.g., cerebropontine angle
tumors), as well as degenerative disorders (e.g., Alzheimer's dis-
ease), demyelinating disorders (e.g., multiple sclerosis), metabolic
disorders (e.g., adrenoleukodysttophy), bacterial meningitis,
Lyme disease, and drug toxicity.
A similar range of severity can be seen in children. Classic
forms of either generalized auditory agnosia or verbal auditory
agnosia are typical features of Landau-Kleffner syndrome, a
relatively uncommon form of childhood epilepsy associated
with continuous spike and wave activity. However, most chil-
dren with CAPD have a developmental variant that is asso-
ciated with less florid disturbances. In some of these cases, the
disorder is specific to auditory processing, while in others
CAPD is comorbid with other developmental disorders, such
as attention deficit hyperactivity disorder (ADHD), specific
language impairment (SLI), dyslexia, and autism.
Below, we briefly review the concept of developmental
forms of CAPD as a distinct diagnosis and discuss its relation
to various comorbid conditions.
Developmental CAPD
APD as a Distinct Diagnostic Entity
Children with CAPD often first come to attention when they
experience difficulties in an academic setting, where success is
dependent on absorbing highly structured auditory information,
often in the presence of poor classroom acoustics and abun-
dant background noise. An estimated 2-7% of school-aged
children exhibit problems suggestive of the existence of a
developmental form of CAPD. These children have difficulty
attending to the teacher for an appropriate length of time and
in following, remembering, or understanding verbal instruc-
tions. They often need to have spoken information repeated
and are easily distracted or influenced by extraneous sounds
in their environment. Though these children are intelligent
and capable of attaining academic success, these deficits
impede their academic performance and have the potential
to be interpreted as a behavioral or motivational problem.
In 2000, a consensus statement by a group of 14 prominent
scientists and clinicians on the problems related to diagnosing
these disorders in school-aged children suggested relabeling
the problem 'auditory processing disorders' (APDs) in order
to 'keep with the goals of maintaining operational definitions,
Central Auditory Processing Disorders 451
avoid the imputation of anatomic loci, and (to) emphasiz(e)
the interactions of disorders at both peripheral and central
sites.' Many continue to utilize the qualifier 'Central,' if only
parenthetically, with recognition of equivalence between the
terms. Existing criteria that define APD are the same as those
discuss for CAPD in the introduction to this articles.
Numerous controversies surround the concept itself and
the pragmatic issues concerning its diagnosis. These issues
include the extent to which a deficit must be restricted to the
auditory domain to be called an APD and whether testing with
nonspeech stimuli, such as tones is sufficient to indicate an
APD, since evidence suggests that speech processing is compu-
tationally more taxing than other sounds, particularly with
respect to the temporal processing demands. Moreover, given
that speech processing is a special case of auditory processing,
there is disagreement over where an APD ends and linguistic or
neurocognitive impairment (e.g., memory, executive function)
begins. Finally, the concept of APD in children is complicated
further by comorbid presentation with developmental disor-
ders, such as autism, dyslexia, SLI, and ADHD.
Despite objective evidence of normal peripheral hearing
sensitivity, children with APD demonstrate a variety of deficits
in behavioral batteries measuring central auditory function. In
addition, their difficulties in remembering phonemes and
manipulating them can result in problems with learning to
read and spell. They may also have difficulty understanding
persons with unfamiliar accents and may stmggle in second
language learning. Auditory memory span and sequencing
auditory information can be problematic. Correspondingly,
they may also demonstrate poor sense of rhythm and musical
ability.
In a portion of children with CAPD, problems in central
auditory processing may be confirmed using electrophysiolog-
ical measures. While abnormal ABRs are observed in less than
10% of children with CAPD, as many as 42% show abnormal
auditory middle-latency responses or P300 responses.
It is uncommon for individuals with developmental forms
of CAPD to demonstrate an obvious structural lesion. Never-
theless, a recent case study is noteworthy insofar as it reported
some characteristic anomalies in a developmental case of CAPD.
The structural MRI revealed selective atrophy evident bilaterally
in auditory cortex and subcortex. Positron emission tomography
(PET), a functional neuroimaging procedure that can measure
neural activation by imaging the brain's use of glucose, found
bilateral hypometabolism in primary and secondary auditory
cortices including Heschl's gyri. These findings underscore the
potential future role that neuroimaging may play in the differen-
tial diagnosis of CAPD.
APD and Autism
Autism is a developmental disorder distinguished by
impairment of social interaction and communication, accom-
panied by restricted interests and repetitive behaviors. Reports
of auditory processing abnormalities are common, frequently
including an absence of orienting reflex to sound, a lack or
inconsistency in response to familiar sounds and noise, intol-
erance for certain sounds at their typical sound levels (hyper-
acusis), overselectivity when listening to multiple auditory
streams, difficulty listening in the presence of background
noise and deficits in perception of emotional content and
prosody in the human voice. In severe cases, the impairment
can be as severe as functional deafness.
Neurophysiological and neuroimaging studies have impli-
cated neurodevelopmental anomalies at multiple levels of the
CANS, from the superior olive to auditory association cortex.
Brainstems of children with autism are significantly smaller
than typically developing peers and although findings are
inconsistent, several studies have reported that auditory brain-
stem-evoked potentials show decreased amplitude and greater
latency.
Functional neuroimaging studies have shown diminished
activation of the auditory cortex when presented with nonver-
bal auditory stimuli. When listening to speech, a neurotypical
brain shows bilateral activation in the superior posterior tem-
poral lobes with a left-sided asymmetry involving posterior
language areas. Studies of individuals with autism have
shown a lack of activation in speech areas entirely or a reversal
of the usual left hemispheric asymmetry. While failing to show
activation of the superior temporal sulcus in response to the
human voice, these studies generally report normal patterns of
activation to nonvocal sounds.
APD and SLI
SLI is a neurodevelopmental disorder characterized by pro-
blems in understanding or producing language in the absence
of causal factors such as hearing impairment, mental retarda-
tion, neurological deficits, primary emotional disorder, or
social deprivation. A number of studies have suggested that
these children have particular difficulties in discriminating
speech sounds characterized by rapidly changing acoustic
information (e.g., stop consonants) as well as in judging
the temporal sequence of rapid presented complex tone pat-
terns. These findings have prompted suggestions that a basic
impairment of auditory temporal processing contributes to
their difficulties with language acquisition, presumably result-
ing in unstable phonemic processing, which may interfere with
encoding and producing spoken language.
Theories advocating a direct causal relationship of CAPD to
SLI have been criticized. Several studies have found that while
rapid auditory processing deficits are common in SLI, they are
not universal. There also seems to be little or no relationship
between the severity of a rapid auditory processing deficit
and the SLI.
While the causal relationship between auditory processing
difficulties and language problems remains in dispute, there
appears to be ample evidence that auditory processing diffi-
culties exist in a significant percentage of these children.
Indeed, neurophysiological evidence of these deficits has
been shown in the form of attenuated or absent MMN
responses to vowel sounds and tones. In cases with severe
receptive involvement, abnormal steady-state responses to
rapid frequency changes may be observed. Recent studies
have shown that impaired electrophysiological responses in
some children with SLI may approach typicality as the child
reaches adulthood. This eventual advance has led some to
interpret these electrophysiological findings as evidence for a
more general immature auditory processing system as a con-
tributory factor to SLI.
452 Central Auditory Processing Disorders
APD and Dyslexia
There is also ongoing debate about the relationship between
an auditory processing deficit and a specific reading disorder
(dyslexia). Again, it has been suggested that children with
dyslexia have low-level auditory and visual perceptual deficits,
possibly related to anomalies impacting magnocellular visual
and auditory systems. These deficits are thought to result in
subtle perceptual problems that interfere with the develop-
ment of phonology, reading, and spelling ability.
While numerous studies have provided evidence in support
of magnocellular deficits in the visual system, data supporting
comparable auditory system involvement are less compelling
partly due to methods of subject selection and inadequate
controls over potentially comorbid attentional problems.
However, postmortem cytoarchitectural studies of a small
number of brains of people with dyslexia have revealed smaller
left MGN and planum temporale than expected. In addition,
ectopias, and microgyria have been observed bilaterally in
perisylvian cortex (including auditory cortex), greater on the
left side. These findings suggest anomalies in cell migration
during neurodevelopment that would impact auditory func-
tion in addition to language and reading. Some association has
been observed between deficits in auditory temporal proces-
sing and problems with phonological processing, although
this appears to be a weak or indirect association.
APD and ADHD
ADHD and APDs share many features which make their
clinical presentation similar. Overlapping symptoms include
poor concentration, distractibility, fidgetiness, and poor
academic achievement. These similarities have led to the
question of whether APD and ADHD are in fact distinct dis-
orders. Though some authors have suggested that they are
variants of the same disorder, noting that they share some
prenatal and postnatal etiological factors, there is a general
consensus that the two disorders are dissociable and discrete
diagnostic entities.
The focus of the relationship between ADHD and APD
has therefore shifted to diagnostic differentiation. Several
approaches to differential diagnosis have been proposed, though
none have been satisfactorily validated at this point in time.
It has been suggested that a diagnosis be made only after com-
paring performance on a visual task with an auditory analogue,
such as visual and auditory continuous performance tests, which
examines a person's ability to sustain attention on a task. Others
have attempted to differentiate ADHD from APD using electro-
physiological measures with little success thus far. Finally, while
stimulant medication such as methylphenidate has been used to
control hyperactivity, impulsivity, and distractibility in ADHD,
there is little evidence to suggest it improves deficits specific to
CAPD.
Treatment
Because of the variability in presentation of CAPDs, interven-
tion should be personalized to address each individual's
specific deficits while building on their strengths. CAPDs may
be treated through environmental modification, teaching of
compensatory strategies or direct treatment of the deficit.
Intervention should be planned and executed by an inter-
disciplinary team.
Environmental modifications increase access to the acoustic
information itself or facilitate learning through modified
teaching methods. This can be achieved through assistive lis-
tening devices such as FM systems, removal of competing noise
or preferential seating in a classroom environment. Instruction
may be modified to include preteaching, reduced linguistic
complexity, slower rates of speaking, and ample use of repeti-
tion/rephrasing. In addition, supplementing auditory informa-
tion with visual presentation of information either in writing
or using gestures is often helpful.
Teaching an individual with CAPD to use compensatory
strategies is key to successful intervention in many cases. Com-
pensatory strategies might include teaching an individual to
ask for repetition of information if he/she did not comprehend
it, independently choose seating away from noise or near a
speaker, and to look at a speaker's mouth while listening.
Individuals often need to take a proactive approach to becom-
ing more aware of the impact of their deficits and the need to
manage potential problems in different situations.
Direct treatment for CAPDs involves auditory training. This
approach is designed to take advantage of brain plasticity to
effect changes in the circuitry of the system through appropriate
and often repeated practice. There is evidence indicating that
sufficiently frequent, intense, and challenging auditory training
can result in neurophysiologic and functional improvements in
auditory skills. Many formal auditory training programs are
available as commercial computer software packages.
See also: Academic Achievement; Aging and the Brain; Aphasia;
Attention Deficit Hyperactivity Disorder; Autism and Pervasive
Developmental Disorders; Behavior Measurement in Psychobiological
Research; The Brain; Brain and Behavior Relationships; Central
Nervous System; Clinical Assessment; Developmental
Psychopathology; Drugs, the Brain, and Behavior; Dyslexia; Event-
Related Potentials (ERPs); Language Development; Neuroimaging of
Dementia; Perceptual Development; Phonetics; Psychophysics;
Sentence Processing; Visual Neglect.
Further Reading
Bamiou D-E, Musiek FE, and Luxon LM (2001) Aetiology and clinical presentations of
auditory processing disorders: A review. Archives of Diseases in Children 85:
361-365.
Bauer RM and Zawacki T (2000) Auditory agnosia and amusia. In: Farah MJ and
Feinberg TE (eds.) Patient-Based Approaches to Cognitive Neuroscience,
pp. 97-106. Cambridge, MA: MIT.
Bellmann A, Meuli R, and Clarke S (2001) Two types of auditory neglect. Brain
124: 676-687.
Cacace A and McFarland D (eds.) (2009) Controversies in Central Auditory Processing
Disorder. San Diego, CA: Plural.
Griffiths TD, Rees A, and Green GGR (1999) Disorders of human complex sound
processing. Neurocase 5: 365-378.
Hausler R and Levine RA (2000) Auditory dysfunction in stroke. Acta Oto-Laryngologica
120: 689-703.
Jerger El and Musiek F (2000) Report on the consensus conference on the diagnosis of
auditory processing disorders in school-aged children. Journal of the American
Academy of Audiology 11: 467-474.
Central Auditory Processing Disorders 453
Johnstone PM (2008) Toward evidence-based practice in the auditory processing
disorder clinic. Perspectives on Audiology 4: 9-14.
Katz J and Tillery KL (2005) Can central auditory processing tests resist supramodal
influences? American Journal of Audiology 14: 124-127.
Konig R, Heil P, Budinger E, and Scheich H (eds.) (2005) The auditory cortex:
A synthesis of human and animat research. Mahwah, NJ: Erlbaum.
Langguth B, Hajak G, Kleinjung T, Cacace AT, and Mailer AR (eds.) (2007) Tinnitus:
Pathophysiology and Treatment. Progress in Brain Research. Amsterdam: Elsevier.
Martin JS and Jerger JF (2005) Some effects of aging on central auditory processing.
Journal of Rehabilitation Research and Development 42(supplement 2):
25-44.
Mailer AR (2006) Hearing: Anatomy, physiology, and disorders of the auditory system,
2nd edn. Amsterdam: Academic Press, Elsevier.
Moore DR (2007) Auditory processing disorders: Acquisition and treatment. Journal of
Communication Disorder 40: 295-304.
Musiek FE and Chermak GC (eds.) (2007) Handbook of Central Auditory Processing
Disorders, 2 vols. San Diego, CA: Plural.
Peretz I and Hyde KL (2003) What is specific to music processing? Insights from
congenital amusia. Trends in Cognitive Science 7: 362-367.
Stefanatos GA, Gershkoff A, and Madigan S (2005) On pure word deafness, temporal
processing and the left hemisphere. Journal of the International Neuropsychological
Society 11: 456-470.
Suga N (2008) Role of corticofugal feedback in hearing. Journal of Comparative
Physiology 194: 169-183.
Relevant Websites
http://www.asha.org/public/hearing/disorders/understand-apd-child.htm - American
Speech and Hearing Association.
http://www.ldonline.org/ld_indepth/process_deficit/capd_paton.html - Central
auditory processing disorders (CAPDs). Learning Disabilities OnLine.
http://kidshealth.org/parent/medical/ears/central_auditory.html - KidsHealth, ‘Central
Auditory Processing Disorder’.
http://www.ldonline.org/article/8056 - National Center for Learning Disabilities, ‘Fact
sheet: Auditory Processing Disorders’.
http://www.nidcd.nih.gov/health/voice/auditory.asp - National Institute on Deafness
and other Communication Disorders.
Central Nervous System
B P Daly, Drexel University, Philadelphia, PA, USA
D M Eichen and B Bailer, Temple University, Philadelphia, PA, USA
R T Brown, Wayne State University, Detroit, Ml, USA
C L Buchanan, The Children’s Hospital of Philadelphia, Philadelphia, PA, USA
© 2012 Elsevier Inc. All rights reserved.
This article is a revision of the previous edition article by Ronald T. Brown and Mary K. Morris, volume 1, pp. 537-549, © 1994, Elsevier Inc.
Glossary
Amnesia Partial or total loss of memory; anterograde
amnesia occurs when an individual has no memory
for events that happened after a traumatic event or
surgery, while retrograde amnesia is the inability to
recall earlier experiences, especially those preceding
a traumatic injury.
Aphasia Impairment in the ability to express or
comprehend language due to injury or disease of
the brain.
Central nervous system The brain and spinal cord.
Executive functions A set of cognitive abilities that control
and regulate other abilities and behaviors, including the
ability to evaluate a problem, plan a response, carry out the
plan, and assess the adequacy of the response within the
context of ongoing environmental cues.
Neurons The basic unit of the nervous system or the nerve
cell that sends and receives electrical signals over long
distances within the body. The neuron includes the cell body
(soma), dendrites, and an axon.
Neurotransmitters Chemicals that regulate nervous system
functioning. The chemicals are released from a nerve cell
which thereby transmits an impulse from a nerve cell to
another nerve, muscle, organ, or other tissue.
Synapse The synapse contains a small gap separating
neurons. Information from one neuron flows to another
neuron across a synapse; it is at the synapse that biochemical
transmission occurs.
The central nervous system (CNS) consists of the brain
(encephalon) and spinal cord (medulla spinalis). The brain is
located in the cranial cavity and is protected by the skull. The
spinal cord lies in the vertebral canal and is protected by the
vertebrae. The CNS is distinguished from the peripheral ner-
vous system, which is composed of the remaining network of
neurons throughout the body. First, we review the basic neuro-
anatomy of the CNS. Next, we examine both reflexive and
voluntary behaviors that are derived from the CNS. Specifi-
cally, behavioral domains that include sensation, perception,
motor activity, and language are described.
Neuroanatomy
Basic Structures
The CNS consists of neurons and glial cells. Neurons consist of
the nerve cell body and its processes - the dendrites and axons.
Neurons constitute about 10% of the cells in the CNS while
glial cells make up the remainder. Neuron cells conduct infor-
mation while glial cells both protect and provide a variety of
support information to neurons. The CNS maybe divided into
regions of gray matter that consist of densely packed glial cells,
neurons, and white matter. White matter consists of myelin-
ated fiber tracts that connect parts of the CNS. The long pro-
cesses of neurons, which are termed axons, are also located in
the white matter. Axons are responsible for carrying outgoing
messages from the cell.
The brain is surrounded by three membranes known as the
meninges, which are located beneath the cranium and verte-
bral column. The first membrane is a tough, outermost layer,
or the dura mater, which is closely attached to the inner surface
of the skull, beneath the subdural space. Epidural hematomas
are triggered by injury to the middle meningeal artery and are
formed between the dura and the skull. Subdural hematomas
form between the dura and arachnoid and are caused by rup-
ture of bridging veins. The arachnoid mater is the second
membrane, which is thin and delicate and resembles a cobweb,
and beneath this is the fluid-filled subarachnoid space that
is part of the cerebrospinal fluid system. The rupture of an
aneurysm of a cerebral artery can result in a subarachnoid
hemorrhage between the arachnoid and the pia. The third
layer, or pia mater, is the thin and transparent innermost
layer of the meninges and is attached to the brain and dips
down into the fissures or sulci. The pia mater forms the inner
boundary of the subarachnoid space.
Neural Tube and Ventricular System
The neural tube is an embryonic structure that is formed
when the lips of the neural groove fuse and is the basis for
the development of the entire CNS. This process generally
begins around the fourth week of development. Differential
growth along the walls of the neural tube occurs during
development and gives rise to the four primary subdivisions
of the embryonic brain: (1) the prosencephalon (forebrain),
(2) the mesencephalon (midbrain), (3) the rhombencephalon
(hindbrain), and (4) the spinal cord. As the fetus develops, the
prosencephalon goes on to develop into the telencephalon and
the diencephalon. The mesencephalon develops into the mid-
brain. The rhombencephalon develops into the metencepha-
lon (the pons and cerebellum) and the myelencephalon (the
454
Central Nervous System 455
medulla oblongata). The spinal cord develops from the most
caudal (toward the rear) portion of the neural tube.
The ventricular system develops from the cavity of the
neural tube in the embryo and consists of interconnected
fluid-filled cavities located in the core of the forebrain and
brainstem. There are four ventricles including two lateral ven-
tricles, the third ventricle, and the fourth ventricle. Each of the
ventricles is filled with cerebrospinal fluid and is lined
throughout the ependyma. The paired lateral ventricles com-
municate with the third ventricle through the interventricular
foramina. The third ventricle is connected to the fourth ventri-
cle by the cerebral aqueduct. At its caudal end, the aqueduct
opens into the fourth ventricle. Cerebrospinal fluid is a clear
and colorless fluid that acts as a buffer to protect the brain and
the spinal cord from mechanical trauma. It is present in the
ventricles as well as the central canal of the spinal cord and in
the subarachnoid space that surrounds the brain and spinal
cord. Each ventricle contains a choroid plexus that produces
the cerebrospinal fluid which is continually being produced
and reabsorbed as it circulates through this intricate system.
The cerebrospinal fluid is reabsorbed via the arachnoid villi
into the venous sinuses when the cerebrospinal fluid pressure
exceeds the venous pressure in the sinus.
Major Subdivisions of the Brain
Diencephalon
The diencephalon is located deep within the cerebral hemi-
spheres and encloses the third ventricle. The four major
subdivisions of the diencephalon include the thalamus, hypo-
thalamus, subthalamus, and the epithalamus. The primary
function of the thalamus, which is the largest division, is
the recognition of sensory stimuli and the relay of sensory
impulses (other than olfaction) to the cerebral cortex. The
thalamus is the chief relay station for all output to the cortex
that comes from the cerebellum and basal ganglia. The thala-
mus also plays an integral role in motor and limbic functions.
The thalamus contains a Y-shaped sheet of fibers termed the
internal medullary lamina that may be divided into three large
cell masses, anterior, mediodorsal, and lateral, each of which
contains several discrete nuclei that are interconnected with
specific cortical regions. The anterior and mediodorsal cell
masses are associated with the thalamic division of the limbic
system, and the lateral cell mass belongs to the sensory and
motor systems. Problems from damage to the thalamus
may include sensory abnormalities, visual-field deficits, and
behavioral changes.
The hypothalamus is located below the cerebral cortex at
the base of the diencephalon. The primary role of the hypo-
thalamus is the central regulation of homeostasis, which
occurs through interfacing with the endocrine system, the
autonomic nervous system, and the limbic system. The hypo-
thalamus controls the nervous system by synthesizing and
releasing neurohormones at regular intervals. Homeostatic
functions associated with the hypothalamus include energy
and temperature regulation, water balance, appetite, cardiovas-
cular functioning, sleep and circadian rhythms, reproduction,
emotional behavior, stress response, blood pressure, and res-
piration. Autonomic, emotional, or endocrine dysfunction is
associated with lesions in the hypothalamus.
The subthalamus and epithalamus are small divisions of
the diencephalon, with the subthalamus situated under the
optic thalamus, while the epithalamus is superior and poste-
rior to the thalamus. The subthalamus contains fiber tracts and
the subthalamic nucleus. The subthalamus and epithalamus
are involved in secretion of melatonin by the pineal gland, the
motor functions of the basal ganglia, and the control of the
circadian rhythms.
Telencephalon
The telecephalon is the rostral part of the embryonic forebrain
and consists of the cerebrum and the limbic system. The cere-
brum is the uppermost and largest part of the human brain and
is situated at the top of the cranial cavity. The cerebrum
includes the cerebral cortex, which is the outermost layer, and
is responsible for controlling complex motor and mental activ-
ity. The cerebral cortex includes the left and right cerebral
hemispheres, separated by the median longitudinal fissure, a
large crevice that runs from front to rear. The cerebral cortex,
which contains billions of neurons and multitudes of inter-
connections, is involved in higher order intelligence, sensory
perception, movement, language, memory, consciousness,
emotion, and personality. Although the left and right hemi-
spheres appear to be physically similar, they control different
types of processes and also are anatomically different.
The cerebral cortex has both gray and white matter. Gray
matter is located on the surface of the cerebral cortex, while a
large mass of white matter is located below the gray matter. The
white matter contains axonal fibers that provide input to and
output from the cortical matter. The fiber tracts are identified as
follows: (1) association fibers, which link one area of the cortex
to another area within the same hemisphere; (2) commissural
fibers, which link the two hemispheres; and finally, (3) projec-
tion fibers that connect the cerebral cortex with the subcortical
centers, the brain stem, and the spinal cord.
The cerebral cortex consists of four lobes: frontal, parietal,
temporal, and occipital. The frontal lobes are located superior
to the lateral sulcus and in front of the central sulcus (i.e., over
the eyes), and contain the motor cortex, which coordinates
voluntary motor activities. Specifically, activities of the volun-
tary muscles on the right side of the body are controlled by the
left part of the motor cortex, while voluntary activities on
the left side of the body are controlled by the right part of the
motor cortex. Motor intention is regulated by the dorsolateral
frontal lobe as increased activity is apparent when animals are
about to make movements. Damage to this area is associated
with motor or autonomic disturbances, including contralateral
limb akinesia, or failure to use one limb in order to respond to
a stimulus, not caused by inattention or weakness of the limb.
The frontal lobe is also involved in important mental activities
such as planning, association of ideas, emotion, and self-
awareness. Finally, the frontal lobe is also concerned with
personality, so lesions in the frontal lobe produce alterations
in character and behavior.
The parietal lobes lie posterior to the central sulcus, but
above the lateral sulcus on the lateral surface of the brain. The
parietal lobe contains the somatosensory cortex, which is
involved in body sensations such as pain, cold, heat, touch,
pressure, and body movement. Damage to the posterior frontal
or anterior parietal lobes can result in the loss of the ability to
456 Central Nervous System
express emotion by modulation of speech patterns resulting in
aprosody or aprosodia. Individuals with aprosodia speak in a
monotone regardless of the situation, but nonetheless experi-
ence normal emotional feelings. Lesions involving the parietal
lobe may also result in a disturbance or loss of function in
perception of size, shape, and texture (agnosia), as well as
writing and memory difficulties.
The temporal lobes are located behind the ears and below
the lateral sulcus. The functions of the temporal lobes are still
being defined, but they are primarily involved in hearing,
smell, and taste, with some involvement also occurring in
vision and memory. Lesions involving the temporal lobe may
produce disturbances in hearing and memory and in the regu-
lation of emotional behavior. The occipital lobes are located
at the extreme back of the head, and contain the visual cortex,
which is involved in visual perception. Damage to the cortical
visual areas of the occipital lobes may result in blindness.
The limbic system is a set of brain structures located at the
top of the brainstem and buried under the cortex. The struc-
tures of the limbic system are highly influential in complex
functions of emotional behavior such as stress, motivations,
emotional responses, and memory. Structures of the limbic
system include the following: amygdaloid complex (amyg-
dala), hippocampal formation (hippocampus, dentate gyrus,
subicular region), hypothalamus, thalamus, olfactory cortex,
basal ganglia, and septum. The amygdala is located deep in
the dorsomedial portion of the temporal lobe and is involved
in emotional responses, the detection of threat, and sensory
input about olfaction. Lesions in the amygdala can cause
behavioural outbursts such as rage or docility. The hippocam-
pus, located within the parahippocampal gyrus on the medial
surface of the temporal lobe, is the major regulator of the
limbic system. The hippocampus is involved in the formation
and encoding of memory, including emotionally charged
memories. Damage to the intrinsic structure of the hippocam-
pus, such as occurs with individuals with Alzheimer's disease,
results in impaired memory for the recall of facts and events
(i.e., declarative or episodic memory). Emotional functioning
is also associated with the cingulated gyrus, the septum, and
areas in the hypothalamus. The cingulate gyrus, situated above
the corpus collusum, is composed of a five-layered cortex as
well as the neocortex. The hypothalamus is responsible for
overseeing autonomic physiological and hormonal regulation.
The olfactory cortex, responsible for our sense of smell, is
located in the temporal lobe. The olfactory cortex is composed
of the pyriform cortex, periamygdaloid area, and part of the
entorhinal area. The olfactory cortex is concerned with con-
scious perception of olfactory stimuli.
Brainstem
The brainstem connects the spinal cord to the cerebral cortex
and consists of three major parts: (1) the midbrain (or mesen-
cephalon), (2) the pons (or metencephalon), and (3) the
medulla oblongata (or myelencephalon). The brainstem has
three main functions: (1) it is responsible for transmitting
information between the spinal cord and cerebrum; (2) it
controls consciousness as well as vital functions such as respi-
ration and cardiovascular functioning through its reflex cen-
ters; and (3) it contains many cranial nerve nuclei, responsible
for much of the movement in the face and neck.
The midbrain is the smallest part of the brainstem and is
located directly inferior to (underneath) the diencephalon and
superior to (above) the pons. The midbrain has two parts, the
tectum and the tegmentum, which are separated by the aque-
duct of Sylvius, or cerebral aqueduct. The tectum, located
dorsal (toward the back) to the aqueduct of Sylvius, is com-
prised of four rounded protrusions termed the corpora quad-
rigemina. The corpora quadrigemina is further divided into
two parts, the superior (upper) colliculi and the inferior
(lower) colliculi. The corpora quadrigemina controls the posi-
tion of the head and eyes in response to visual, auditory, and
somatic stimuli as well as transmits auditory signals to the
auditory center in the cerebral cortex. The tegmentum, located
ventral (toward the front or belly) to the aqueduct of Sylvius,
contains the red nucleus that controls motor function, the
oculomotor nerve, and the trochlear nerve which are both
involved in eye movement. Another important structure in
the midbrain is the substantia nigra that is involved in reward
seeking, addiction, and movement. Degeneration of the sub-
stantia nigra is implicated in Parkinson's disease. Finally, the
cerebral peduncles, located in the tegmentum, are bundles of
nerves that run from the cerebrum to the spinal cord and are
responsible for motor control.
The pons is located between the midbrain and the medulla
oblongata. It is ventral to (in front of) the cerebellum and is
responsible for relaying information between the cerebellum
and the cerebrum. Part of the reticular formation, which is
responsible for maintaining an awake state, is housed within
the pons. Lesions in the pons may cause a coma-like state
called Tocked-in syndrome' which is characterized by weak-
ness of the extremities and the face, with a limited level of
consciousness.
The medulla oblongata is the lowest structure of the brain
and connects to the spinal cord at the foramen magnum.
Because of its location, the medulla is responsible for transmit-
ting information from the spinal cord to the higher centers of
the brain. The medulla oblongata contains the respiratory,
cardiac, vomiting, and vasomotor centers. The medulla also
contains part of the reticular formation and controls vital
autonomic functions such as breathing, heart rate, and blood
pressure. As such, any damage to this portion of the brain may
result in respiratory failure and/or death.
Cerebellum
The cerebellum sits dorsal to the brainstem and directly poste-
rior to (or toward the back of) the pons and the medulla
oblongata and is connected to all three parts of the brainstem
through large bundles of nerve fibers called peduncles. The
peduncles connect the cerebellum to the rest of the nervous
system. Similar to the cerebrum, the cerebellum has two hemi-
spheres that are covered by a folded surface layer, called the
cortex. Both white and gray matter comprise the cortex, which
also houses deep nuclear structures. The cerebellum is primar-
ily responsible for motor functions, such as coordination and
stabilization of movement, and is involved with maintaining
balance. The cerebellum is able to fine-tune movements using
the extensive neural connections from the motor cortex of the
cerebrum to the spinocerebellar tract (located in the spinal
cord). These neural connections allow the cerebellum to gather
information regarding intended and actual movement and
Central Nervous System 457
make adjustments as needed. Individuals with cerebellar dis-
ease may demonstrate slurred speech, poor coordination, and
disrupted gait.
Spinal cord
The spinal cord is the lowermost part of the CNS and connects
with the brain through the medulla oblongata at the foramen
magnum. The spinal cord is shorter than the bony vertebral
canal that surrounds it and runs from the foramen magnum to
the second or third lumbar vertebra. The spinal cord consists of
inner gray matter, containing the nerve cell bodies, and outer
white matter, containing the axons of the nerves. In the center
of the spinal cord is a space called the central canal, which is
filled with cerebrospinal fluid.
The spinal cord has 3 1 segments of paired spinal nerves and
ends with a thin, fibrous thread called filum terminale, in the
coccyx. The 31 segments are divided into regions: (1) the top
8 segments comprise the cervical region such that Cl-8 nerves
exit the spinal cord in the cervical region or in the neck; (2) the
next 12 segments are part of the thoracic region such that Tl- 12
nerves exit the spinal cord in the thoracic region or the chest;
(3) the next 5 segments make up the sacral region such that
LI -5 nerves exit the spinal cord in the lumbar region or
the lower back; (4) the Sl-5 nerves exit the spinal cord in
the sacral region or near the pelvis; and finally, (5) the last
segment consists of the coccygeal nerves which exit the spinal
cord near the coccyx, or tailbone. The main purpose of the
spinal cord is to transmit and receive information between
the brain and the rest of the body (muscles and glands). The
spinal cord also has some of its own neural circuits that
are responsible for reflexes and rhythmic movements, such as
locomotion and respiration.
Spinal cord injuries affect the whole body; however, impair-
ments in sensation and/or motor ability typically occur below
the site of the injury. For this reason, some persons have altered
functioning in their legs and lower body only (paraplegic),
while others have altered functioning involving all four limbs
(tetraplegic). As such, spinal cord injuries are classified accord-
ing to the affected individual's degree of loss of motor and/or
sensory function.
Behavior
Sensation and Perception
All human contact with the outside world is contingent upon
sensation and perception. Sensation is the process of receiving
physical stimulation through specialized sensory organs. Per-
ception is the process of becoming aware of, modulating,
integrating, and interpreting sensory stimulation. Perception
represents the modification and organization of incoming sti-
muli. Specialized receptors convert sensory stimuli (e.g., light
waves) into neural activity in the form of either graded poten-
tials or action potentials, which is a process called transduc-
tion. The sensory systems include vision, audition (hearing),
taste, olfaction (smell), and somatic sensation (touch).
Visual system
The sensory organ of the visual system is the eye, which detects
and focuses stimuli. Light enters the eye through a transparent
outer covering known as the cornea. Subsequently, light travels
through the pupil, an adjustable opening in the iris (the col-
ored part of the eye), is focused by the lens, and finally falls
on the light-sensitive area of the eye, referred to as the retina.
Within the retina are two visual receptor cells: rods and cones.
Rods are visual receptors that are sensitive to low-intensity light
and are involved in night and peripheral vision; cones are
mainly found in the central region of the retina and permit
color vision.
Once the rods and cones are stimulated, they carry visual
impulses via the optic nerve, the large nerve that carries visual
impulses from the retina to the visual processing areas of the
brain, to the optic chiasm. At the optic chiasm, approximately
one-half of the fibers from each eye's optic nerve cross to the
other side of the brain. These fibers cross to the other side of
the brain from the nasal portions of each retina. As a result,
information from each eye reaches both sides of the brain,
with signals from the left visual field terminating in the right
hemisphere and signals from the right visual field being
sent to the left hemisphere. Finally, these optic tracts travel to
the thalamus and ultimately to the primary visual cortex of
the occipital lobe.
Auditory system
The ear is the sensory organ of the auditory system. From the
outer ear, sound waves enter the auditory canal and travel to
the thin membrane known as the eardrum (tympanic mem-
brane). Sound waves cause the eardrum to vibrate, which
creates movement in the three ossicles (tiny bones) of the
middle ear: the malleus, incus, and stapes. The ossicles amplify
vibrations and then transmit these higher-pressure sound
vibrations to the oval window, the membrane that leads from
the middle to the inner ear. The oval window then sends the
sound information in wave form through the fluid of the inner
ear to the cochlea. Attached to the basilar membrane of the
cochlea is the organ of Corti which contains auditory receptor
cells known as hair cells. The hair cells are compressed between
the basilar membrane and the tectorial membrane. Fluid
sound waves in the cochlea move these two membranes in
such a way that the hair cells are bent and distorted, thus
transforming the fluid waves into neural impulses in the adja-
cent auditory nerve. The impulses follow pathways from the
cochlea via a series of nuclei to the thalamus; finally, the signal
ends in the primary auditory cortex, which is located in the
temporal lobe. The auditory system is bilateral, meaning that
each ear projects to both sides of the brain.
Taste and smell
Taste and smell (olfaction) are referred to as chemical senses
because both transduce chemical signals into perception. The
primary receptor cells for taste are located in the taste buds
(gustatory calyculi), concentrated on the upper surface of the
tongue. The taste buds are sensitive to five basic tastes: bitter,
sweet, sour, salt, and umami (savory). Other tastes are consid-
ered to be a combination of these five qualities and/or a
mixture of taste and smell. In humans, signals from the taste
buds are conveyed to the brain through three different cranial
nerves (facial, glossopharyngeal, vagus) that, combined, form
the tractus solitarius in the medulla oblongata. From here, the
signals cross the midline and continue, via the thalamus, to the
458 Central Nervous System
insula and the somatosensory cortex (tongue area), ultimately
leading to the experience of taste. A second branch of the
tractus solitarius projects to the lateral hypothalamus and the
amygdala, areas that are involved in feeding.
In contrast to taste, there are hundreds of olfactory recep-
tors, each of which binds to a particular molecular feature.
The primary receptor cells for olfaction are located within the
olfactory epithelium, the membrane that lines the roof of
the nasal cavity. The axons of these olfactory receptor cells
synapse in the olfactory bulb. From there, the olfactory nerve
passes through the pyriform cortex and subsequently to the
thalamus and the orbital frontal cortex. Olfactory receptor cells
are not specific to single odors; instead, each odor molecule
possesses a variety of features that excite specific receptors in
varying degrees. In fact, it is believed that the combination of
multiple signals from different receptors is processed in the
olfactory system to form the experience of a particular smell.
Somatosensory system
The somatosensory system includes the following cutaneous or
skin senses: pressure (touch), pain (nociception), vibration
and temperature, position sense (proprioception), and body
movement (kinesthesis). The somatosensory system reacts to
diverse stimuli and skin sensations through the mechanical,
chemical, and thermal properties of the tissue in which the
sensory nerves terminate. The quality of sensation is also due
to variations in temporal and spatial patterns of neural dis-
charge in the sensory nerves. Axons that carry information
from cutaneous receptors gather together in nerves that enter
the spinal cord through the dorsal roots; these nerves are
organized into two systems in the spinal cord, the posterior
column-medial lemniscus pathway and the spinothalamic
tract. Both systems ascend to the posteroventral nuclear com-
plex of the thalamus, which, in turn, sends projections to the
somatosensory cortex. The posterior column-medial lemnis-
cus pathway is responsible for transmitting pressure and posi-
tion senses; the spinothalamic tract mediates the sensation of
pain and temperature. The arrangement of these projections in
the postcennal gyrus generally represents the body parts topo-
graphically, with the legs at the top and the hand at the bottom
of the gyrus.
Motor System
The motor system is the part of the CNS that is involved in
movement. The motor system consists of a number of subsys-
tems including the spinal cord, the brainstem, the cerebellum,
the basal ganglia, and the neocortex. These structures are all
closely interconnected and do not function in isolation. In fact,
some of these connections are arranged in parallel and hierar-
chical systems exist within the motor system. The brainstem
and cortex each influence the spinal cord autonomously for the
purpose of invoking movements, but each system is also
interconnected with the other system, thereby permitting
them to influence each other. The cerebellum is a crucial
coordinating center in the motor system; it directs balance
and posture. Along with the basal ganglia and motor cortex,
the cerebellum is essential to the performance of coordinated
and refined motor movements. Damage to the cerebellum can
result in ataxia, a lack of muscle coordination during voluntary
movements that can affect movement, speech, balance, and
gait. The basal ganglia represent another level of the motor
system. These substantial subcortical nuclei are important in
the initiation and control of voluntary movements, particularly
in the regulation of muscle tone. Damage to the basal ganglia
and the regulation and initiation of motor commands can
result in either decreased motor activity (e.g., akinesia) or
generation of involuntary movements. These disruptions are
most prominently observed in the tremors and rigidity of
Parkinson's disease or the uncontrollable limb movements
of Huntington's chorea.
The motor cortex follows two different major pathways to
descend into the spinal cord; these are referred to as the lateral
and ventromedial systems. These pathways allow each hemi-
sphere to control the trunk on the corresponding side of the
body and the arms and legs of the opposite side of the body.
Ultimately, both pathways synapse with motor neurons
located in the nuclei of the cranial nerves and in the ventral
horns of the spinal cord. The motor neurons are the final link
in the pathway, projecting directly into the muscles. The lateral
system projects from both the cortex and the brainstem in one
hemisphere to the spinal cord of the opposite side. The lateral
system contains the axons of the cortical neurons that are
responsible for the muscles and the extremities. Specifically,
the lateral system includes the corticospinal and rubrospinal
components. The corticospinal component connects directly
with motor neurons and is involved in the movement of distal
musculature (i.e., fingers and hands) on the contralateral side
of the body. The rubrospinal component connects with inter-
neurons and then with motor neurons that control the muscu-
lature of the shoulders, arms, and hands. The other major
pathway, the ventromedial system, projects to interneurons
of the spinal cord from the cortex and the brainstem. Unlike
the lateral system, the ventromedial system does not change
sides when it enters the spinal cord. This system descends
directly into the ventral and medial part of the spinal cord,
where it innervates motor neurons that control the muscula-
ture of the trunk.
For both motor and sensory cortex, a topographical rela-
tionship exists between body parts and locations within the
neocortex. This relationship is often demonstrated schemati-
cally in texts by cartoon men or homunculi drawn over the
motor and sensory areas to represent this type of organization.
The face and the hands of the homunculi are larger than other
body parts, as they are capable of finer perceptions and move-
ments. Thus, relative to other body areas, they require propor-
tionately more neocortex to be represented accurately.
Cerebral Lateralization
While both hemispheres play a role in most behavior, cere-
bral lateralization refers to the preferential use of one hemi-
sphere (i.e., left and right hemisphere) of the brain to
perform certain cognitive functions. Although on initial ex-
amination the left and right cerebral hemispheres appear to
be quite similar, a detailed examination reveals numerous
anatomic differences that are believed to support functional
asymmetries. For example, the superior temporal portion of
the planum temporale, a cortical region posterior to the
auditory cortex, is larger on the left side in the majority of
Central Nervous System 459
individuals and this difference in size is believed to reflect
left-hemisphere specialization for language.
Information about hemispheric contributions to cognitive
function and hemispheric specialization was initially obtained
from studies of split-brain patients whose corpus callosums
were severed as treatment for severe epilepsy. Additional evi-
dence from studies of patients with lateralized brain lesions
and normal individuals evaluated under specialized labora-
tory conditions also suggest differences. Specifically, for most
people, the left cerebral hemisphere is dominant in the control
of speech, language and analytic processing, and mathematics.
In contrast, the right hemisphere appears to be specialized in
spatial-relational processing, creativity, and facial recognition.
Although the two hemispheres may be specialized for different
types of stimuli, each hemisphere also might have a character-
istic processing style that is best suited for analyzing those
stimuli. For example, the left hemisphere may process infor-
mation in a serial, sequential manner, which is conducive to
the analysis of language. Alternatively, the right hemisphere
has been described as a parallel, holistic processor, more
appropriate for handling complex configurations.
Whereas the majority of individuals fit this functional
model of lateralization, individual variability is clearly present
and is generally related to factors such as handedness and
gender. Neuroimaging techniques recently provided support
for gender differences in language lateralization, with males
utilizing only the left hemisphere and a majority of females
activating both the right and left hemispheres on language-
related tasks. However, a majority of research supports that
some degree of functional asymmetry appears to be present
from birth. For example, newborns can accurately discriminate
speech from nonspeech sounds and demonstrate greater acti-
vation of the left hemisphere in response to human speech.
Language
Language is considered a function that is both lateralized and
localized in the brain. The association between damage to the
left hemisphere and language impairments is most widely
credited to the discoveries of French neurologist Paul Broca
and German neurologist Carl Wernicke. Broca found that
lesions in the left frontal lobe in an area now known as 'Broca's
area' affected speech production. Wernicke discovered that
damage to the left temporal lobe results in different language
impairment, specifically a difficulty in understanding lan-
guage. Collectively, these are early descriptions of aphasias, or
acquired language disorders that can be traced to damage of
specific areas of the brain. Aphasias are classified by different
patterns of impairment of language processes. Patients may
exhibit disorders of comprehension of speech or written lan-
guage, or they may exhibit deficits in the production of
language. These disorders may affect speech fluency, speech
repetition, syntax, grammar, word retrieval, or writing. For
example, in all cases when comprehension remains intact,
Broca's aphasia (motor/ expressive aphasia) is associated with
repetitive speech, phonemic paraphasias (one sound is incor-
rectly substituted for another), and disrupted grammar and
syntax. On the other hand, patients with Wernicke's aphasia
(sensory/receptive aphasia) demonstrate fluent speech with
adequate syntax and grammar but often include contrived or
inappropriate words; thus, problems are largely associated
with comprehension.
Studies of split-brain patients have assisted in the identifi-
cation of the independent function of each hemisphere in
language. Specifically, the right hemisphere is largely responsi-
ble for prosodic elements, or the emotional elements of lan-
guage. The right hemisphere also plays a role in rudimentary
speech and is largely responsible for spatial abilities. Research
on individuals who are deaf, who use sign language, reveals
that similar areas of the brain are involved in sign production.
This line of research also indicates that damage to the left-
hemisphere language area is more likely to produce signing
aphasia than damage to the same area in the right hemisphere.
As such, the lateralization of language is preserved in deaf
individuals.
In addition to clinical investigation of patients with brain
damage, the major brain regions involved in language have
also been identified by using cortical stimulation during brain
surgery and functional brain injury. Other areas associated
with language include the primary sensory and motor regions
of the face and the supplementary motor area.
See also: Aphasia: Glial Cells; Hippocampal Formation; Language
Development; Memory; Motor Control; Sense of Smell; Touch; Visual
Perception.
Further Reading
Blumenfeld H (2002) Neuroanatomy Through Clinical Cases. Sunderland, MA: Sinauer
Associates.
Brodal P (2004) The Central Nervous System: Structure and Function, 3rd edn.
New York: Oxford University Press.
England MA and Wakely J (2006) Color Atlas of the Brain and Spinal Cord, 2nd edn.
Philadelphia, PA: Mosby Elsevier.
Fix JD (2005) High-Yield Neuroanatomy. Philadelphia, PA: Lippincott Williams &
Wilkins.
Hendelman WJ (2006) Atlas of Functional Neuroanatomy, 2nd edn. Boca Raton, FL:
CRC Press, Taylor and Francis Group.
Lane TE, Carson M, Bergmann C, and Wyss-Coray T (2007) Central Nervous System
Diseases and Inflammation. New York: Springer.
Morgan JE and Ricker JH (eds.) (2008) Textbook of Clinical Neuropsychology.
New York: Taylor & Francis.
Nieuwenhuys R, Voogd J, and van Huijzen C (2007) The Human Central Nervous
System, 4th edn. New York: Springer.
Purves D, Augustine GJ, and Fitzpatrick D, et al. (eds.) (2008) Neurocience, 4th edn.
Sunderland, MA: Sinauer Associates.
Rubin M and Safdieh JE (2007) NetteYs Concise Neuroanatomy. Philadelphia, PA:
Saunders Elsevier.
Scheld WM, Whitley RJ, and Marra CM (2004) Infections of the Central Nervous
System. Philadelphia, PA: Lippincott Williams & Wilkins.
Child Abuse
M J Lawler and E B Talbot, University of South Dakota, Vermillion, SD, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Attachment Developed by John Bowlby and Mary
Ainsworth, it is an innate emotional and behavioral system
of protection for children that includes a child's attachment
system and a parent's protection system.
Child neglect It is the failure of a parent or caregiver to
provide for a child's basic emotional, medical, educational,
and physical needs. Forms of neglect include emotional,
medical, educational, and physical.
Child sexual abuse Sexual abuse involves an adult engaging
a child in sexually explicit conduct, including direct sexual
contact, simulation of sexual contact, or visual
representation of sexual contact. This may include rape,
molestation, prostitution, child pornography, incest, or
other forms of exploitation.
Emotional abuse Also called psychological abuse, it is a
pattern of parenting or caregiving that degrades a child or
impairs healthy development.
Human trafficking It is the recruitment, transportation,
transfer, harboring, or receipt of persons by means of
threat, or use of force or other forms of coercion,
of abduction, of fraud, of deception, of the abuse of
power or position of vulnerability, or the giving or
receiving of payments or benefits to achieve the
consent of a person having control over another person,
for the purpose of exploitation.
Physical abuse It is nonaccidental injury, inclusive of
all levels from bruises to death, as a result of harm
from a parent, caregiver, or adult responsible for
a child's care.
Introduction
Throughout history, most cultures prohibited the abuse of
children. Since the future of families, groups, and nations
depends on the success of children to move forward and bear
their own children to continue the species, most cultures have
stated or implicit values about protecting children from harm.
Children are generally valued for their representations of emo-
tional connections between parents or others and serve as
profound symbols of hope for the future as concrete protec-
tors of traditions, caregivers of elders, and producers of new
resources and subsequent generations. In addition, children
are the rejuvenators of families and communities, renewing
adults' senses of possibility.
Human children are altricial mammals, vulnerable at birth
and unable to survive outside the care and support of a care-
giving adult, usually a mother or parent. In this sense, there is
no individual infant or young child independent of a dyad
or family. On his or her own without a responsive and protec-
tive parent, caregiver, or community of caregivers, an infant
or young child will not survive. Thus, millennia of evolu-
tion created complex and strong caregiving and protection
systems for human children that are part of the fabric of all
communities.
Nonetheless, child abuse remains a serious social problem,
as children are some of the most vulnerable members of our
societies and prone to victimization by those stronger than
them or on whom they depend for care and protection -
adults. While abuse against children is often perpetrated by
nonfamily members, it is the exception rather than the rule.
Outside areas of conflict such as war zones, children are more
likely to be harmed, or put in harmful situations, by a family
member or someone close to them in their community.
In this article, we provide an overview of child abuse as it
pertains to intrafamilial child abuse or child maltreatment
perpetrated by a family member or a person close to a child.
This includes four general types of child abuse: physical abuse,
sexual abuse, neglect, and emotional abuse. However, we have
also included information on human trafficking, which may
be the result of intrafamilial or extrafamilial maltreatment, as it
is an area of child exploitation, affecting children and mothers,
that warrants further investigation and concern.
The theoretical framework used for this article is attachment
and parent-child relationships with the view that child abuse
is an extreme form of relationship-disruptions for children. As
such, child abuse risk factors and protective factors for chil-
dren, parents or caregivers, and community is reviewed with a
particular focus on the role of relationships in contributing to
maltreatment, ameliorating its traumatic impact, and prevent-
ing its occurrence. Child abuse treatment models reviewed here
focus on interventions that repair a child's capacity, and a
parent's or alternative caregiver's capacity, to enter into trusting
and emotionally responsive relationships.
Attachment
Young children rely on their mothers or other caregivers for
protection and guidance. Ideally, when a young child calls out
for help or appears to need assistance, a mother or other
caregiver will respond with support, nurturance, and protec-
tion. This emotional response is believed to be part of innate
emotional and behavioral systems of protection - a child's
attachment system and a parent's protection system.
Over time, the repeated expression of needs from a child
and appropriate responses of emotional and physical support
from a parent or other caregiver contribute to a child's feeling
of a safe haven with that parent or caregiver. A parent's consis-
tent signaling of emotional availability to a child affirms the
stability of a child's safe haven. With a secure attachment
460
Child Abuse 461
system between a child and a parent or caregiver, a child is able
to use the relationship as a secure base from which he or she
can explore the world and engage in other relationships based,
in part, on the initial attachment system. A child is able to seek
assurance and emotionally refuel as needed with a secure base,
a responsive parent or caregiver.
In essence, a secure attachment system can create a template
for future healthy and emotionally responsive relationships. It
is believed that internal working models of relationships are
based on early attachments systems between a child and a
parent or caregiver. When those attachment systems are secure,
derived from emotionally and physically responsive support
from a parent or caregiver, a child can integrate a secure inter-
nal working model of attachment, encouraging his or her
healthy development of autonomous emotional regulation
and adept social skills.
If a parent or caregiver is not responsive, emotionally un-
available, abusive, or neglectful, a child is likely to develop an
insecure attachment pattern and corresponding insecure inter-
nal working model of attachment. Child abuse and neglect can
be the result of or contribute to insecure attachment systems.
In the long term, children with insecure attachment sys-
tems are more likely to have social, cognitive, and mental
health deficits.
Attachment Patterns
Four attachment patterns can be predicted from a child's
attachment systems: secure, anxious, avoidant, and disorga-
nized. Secure attachment usually emerges from a relatively
normal functioning, nonabusive, child-parent relationship
and attachment system. Additionally, secure attachment
tends to correspond with stable physical, academic, and
social development, and a general absence of anxiety and
frustration.
Anxious attachments
Anxious, avoidant, and disorganized attachment patterns are
classified as forms of insecure attachment and are associated
with behavioral, relationship and mental health problems.
Anxious attachment is usually produced from anxious, uncer-
tain, and unreliable parenting from a caregiver who may have
experienced similar anxious attachment patterns with his or
her parents. The anxiety arises from a sense that a parent or
caregiver is not reliably there to meet a child's needs. Such
anxious attachments emerge from neglectful or inconsistent
caregivers, resulting in a child doubting support from others
and demanding more attention and reassurance.
Avoidant attachment
When a child's requests for comfort and nurturance are consis-
tently met with insensitivity and unresponsiveness, he or she is
likely to develop avoidant attachment patterns. A child with
avoidant attachment patterns may exhibit uncertainty and
anger resulting from a view of others as unhelpful, cold, or
uninterested when a child needed help or support. Avoidant
attachment patterns tend to be associated with people who
do not trust others and may not be able to fully consider the
needs of others.
Disorganized attachment
The most recently identified classification of attachment sys-
tems, disorganized attachment, arises from scenarios where a
child's attachment figure or parent is observably frightened or
frightening when a child needs comforting or reassurance.
Disorganized attachment involves frightening and violent
behaviors from parents or caregivers and is, therefore, more
common with families suffering from combined or distinct
problems of child abuse, domestic violence, and family insta-
bility. The severe long-term consequences for disorganized
attachment systems include later dissociative disorders, anxiety
disorders, and serious behavior problems.
Cultural Variations
Though there is general cultural pressure worldwide toward the
selection of secure attachment systems, more cross-cultural and
contextual attachment research is needed. Attachment theory
and research focuses on Western culture models of parent-
child relationships, but it continues to expand in consideration
of other cultural and subcultural variations of child caregiving.
Emerging models of social network support and plural caregiv-
ing referencing multiple attachment figures are in the process
of being tested and will warrant further consideration in future
reviews on attachment.
Types of Child Abuse
Most cultures recognize four major areas of child abuse: physi-
cal abuse, sexual abuse, neglect, and emotional abuse. These
categories are not fully inclusive of all harm perpetrated on
children, and most maltreated children are abused in various
forms and not exclusively in one type or another.
Physical Abuse
Physical abuse is nonaccidental injury, inclusive of all levels
from bruises to death, as a result of harm from a parent,
caregiver, or adult who is responsible for a child's care. The
exact detail of what constitutes physical abuse varies by juris-
diction or culture, but usually hinges on the intent of injury to
be classified as abuse. Physical discipline is not considered
abuse as long as it is culturally and physically reasonable and
does not cause bodily injury to a child. However, infants are an
exception, and physical discipline of an infant is usually
viewed as abuse.
Sexual Abuse
Sexual abuse involves an adult engaging a child in sexually
explicit conduct, including direct sexual contact, simulation of
sexual contact, or visual representation of sexual contact. The acts
of sexual abuse may include rape, molestation, prostitution,
child pornography, incest, or other forms of exploitation of a
child for the sexual gratification of an adult. Though there are
varying legal definitions for child sexual abuse across cultures
and jurisdictions, sexual abuse of a child can lead to trauma and
stigma for a child and his or her family. Long-term health of a
462 Child Abuse
sexually abused child usually depends on the support of the
family or community and protection from further harm.
Child Neglect
The usual hallmark of child neglect is the omission of care
rather than the commission of any particular act of maltreat-
ment. Neglect is the failure of a parent or caregiver to provide
for a child's basic emotional, medical, educational, and phys-
ical needs. Interpretation of child neglect depends on com-
munity standards and cultural values, as well as prevailing
resources within a community. For example, lack of indoor
plumbing or safe housing may be viewed as substandard living
arrangement in many communities, leading to a potential
assessment of neglect of children living in such conditions.
However, in a community that lacks resources and standard
housing does not always include indoor plumbing, child
neglect may not be an appropriate assessment based on these
particular elements.
Emotional neglect
Emotional neglect involves inattentiveness to a child's emo-
tional and development needs. This may take various forms,
including allowing independence inappropriate for a child's
development needs. For example, permitting a child to use
drugs or alcohol could be considered neglectful of a child's
emotional and physical development. Interpretations of emo-
tional neglect must also include cultural contexts, as plural or
collective caregiving in some cultures and communities may
allow for greater variation in emotional responsiveness based
on shared caregiving practices. More specifically, a parent's
potential lack of attentiveness to a child may be balanced by
caregiving and attachment systems with other caregivers, such
as grandparents, who are able to respond adequately to a
child's needs.
Physical neglect
Physical neglect means a parent or caregiver fails to provide
necessary shelter, food, or supervision. In these situations,
resources and poverty are important considerations, as a parent
or caregiver doing his or her best with meager resources to care
for children is not physical neglect. In fact, poverty itself should
never be considered physical neglect. Rather, physical neglect
involves a risk to a child's health and safety and may involve
parents failing to use the resources available to them to support
children's needs.
Examples of physical neglect include
• Unsanitary living conditions: Home is unsafe or set near
unhealthy or unsanitary conditions, which may cause sick-
ness or harm to a child, and a parent or caregiver does not
respond to a child's needs.
• Unstable living arrangements: A parent or caregiver does not
have a stable home (e.g., homeless), and the instability
presents a danger to the health and well-being of a child.
• Abandonment: A child is left without adequate care and
supervision or without a plan for the child's care, leaving
a child at risk of harm.
• Exposure to drugs and alcohol: Parents' or caregivers' addic-
tion to drugs and alcohol presents a host of risk factors for
children. Children exposed to drugs or alcohol prenatally
have special medical needs that require ongoing attention,
and parents or caregivers who continue to use drugs
and alcohol may be very challenged to fully address an
infant's needs.
Medical neglect
Not providing for a child's medical or mental health needs may
be considered medical neglect. Of special sensitivity are those
situations where a child has a life-threatening condition that
requires medical intervention and a parent or caregiver does
not or will not attain professional services. The contextual
considerations for these matters include parental or caregiver
resources and the capacity to care for a medically fragile child
and religious beliefs that may be counter to recommended
courses of action. In the latter case, court intervention may
be needed to obtain the right balance between community
and familial standards in consideration of the best interest of
the child.
Examples of medical neglect include
• Inadequate or delayed health care: A parent or caregiver is not
able to care for a child's acute or chronic medical condition,
physical, or mental disability that if left untreated causes
harm or danger to a child.
• Inadequate nutrition: A parent or caregiver is unable to pro-
vide for a child's basic nutritional needs or purposefully
withholds food and water.
• Lack of treatment for mental health: A parent or caregiver does
not provide for necessary mental health concerns causing
potential health and emotional impairment for a child.
Educational neglect
Inattention to a child's educational needs or failing to attend to
a child's special education needs may be viewed as educational
neglect. Considering community standards and resources as
well as the mandatory education requirements of a culture,
educational neglect involves deliberate lack of attention to a
child's educational needs even when a parent or caregiver has
been fully informed of his or her responsibility for that child's
education and services have been made available. In these
instances, a parent or caregiver may lack the physical or mental
capacity to help a child with educational needs so efforts may
be necessary to ameliorate both a child's and parent's or care-
giver's needs.
Examples of educational neglect include
• Unmet special education needs: A parent or caregiver is unable
or unwilling to have a child participate in special learning
or remedial education programs, contributing to a child's
learning disabilities.
• Chronic truancy: A parent or caregiver does not provide for a
child's learning needs by not enrolling a child in school or
not supporting a child's attendance.
Emotional Abuse
Emotional abuse, or psychological abuse, is a pattern of par-
enting or caregiving behavior that degrades a child or impairs
healthy emotional development. This may involve constant
demeaning or belittling comments and criticisms that impede
a child's capacity to function and causes emotional strain for a
Child Abuse 463
child. Withholding of parental or caregiver love, support, and
guidance may also be considered emotional abuse if it causes
undue strain and anxiety for an involved child.
Examples of emotional abuse include
• Isolation : A parent or caregiver does not allow a child to
interact with peers or others causing undue dependency
and potential harm to a child's physical and emotional
development.
• Ridiculing and degrading : A parent or caregiver constantly
belittles, degrades, and ridicules a child causing emotional
harm and shame.
• Witnessing violence: A parent or caregiver allows a child to
repeatedly witness criminal activities between caregivers
and others.
Human Trafficking and Exploitation
ffuman trafficking is a modern-day form of slavery and is
increasingly recognized as an emerging type of child abuse. It
is estimated that 600 000-800 000 people worldwide are traf-
ficked annually. No country in the world has been exempted
from human trafficking. The Palermo Protocol adopted by the
United Nations provides the most commonly recognized defi-
nition of human trafficking described as
the recruitment, transportation, transfer, harboring, or receipt of
persons by means of threat, or use of force or other forms
of coercion, of abduction, of fraud, of deception, of the abuse of
power or of a position of vulnerability, or the giving or receiving
of payments or benefits to achieve the consent of a person having
control over another person, for the purpose of exploitation. Exploi-
tation shall include, at a minimum, the exploitation of the prostitu-
tion of others or other forms of sexual exploitation, forced labor or
services, slavery or practices similar to slavery, servitude, or the
removal of organs.
The primary forms of human trafficking are sex trafficking, sex
trafficking of minors, and labor trafficking. Other forms of
human trafficking include the trafficking of children for adop-
tion and trafficking of organs. Women and children are the
most vulnerable groups of individuals trafficked, with ~50%
of all persons trafficked being children under the age of
18 years.
Causes and Contributing Factors
The underlying causes of human trafficking are issues of
poverty that have been compounded by the rapid growth
of communication technology and globalization. Victims are
trafficked within countries of origin and transported across
international borders. Collapses of economic systems in
some governments and regions have left large segments of
the population desperate for money for survival. Women and
children have been hit particularly hard by poverty. It is esti-
mated that internationally 70% of all persons who are living in
poverty are women. Desperate for work and a way to support
their families, many women have been lured into slavery
through employment agencies and false advertisements that
promise work abroad. Many women who have responded to
these false advertisements have disappeared and are assumed
to have been trafficked. In poor countries, children are left
alone to parent themselves and younger siblings. The issue of
families without an adult parent or caregiver is so relatively
common in some parts of the developing world (e.g., areas of
extreme poverty) that the term 'child-led families' is used to
describe parent-less families. Children, left to fend for them-
selves, have become victims of traffickers and have also dis-
appeared without a parent, family, or community to protect
them or seek assistance when they disappear.
In the United States, many children who have run away
from child welfare services (e.g., foster care) have become
victims of traffickers and prostituted. Adolescents have been
found to have been blackmailed into prostitution by gangs
who first provide them with drugs, and when addicted require
payment to support the addiction. Some traffickers take por-
nographic photos of victims, then threaten to expose them to
their parents unless they cooperate, causing some adolescents
to be secretly forced into prostitution while still living at home.
Psychological and Health Issues
Human trafficking has consequences that interfere with the
developmental milestones expected during childhood and
adolescence. Captivity interferes with the development of
one's identity and the establishment of societal or cultural
morals and values. They experience psychological distress
that includes depression and posttraumatic stress disorder.
The psychological stress is so great that many captive women
and children commit suicide. The overall psychological impact
of the abuses of human trafficking can profoundly change the
development of one's character and personality.
During captivity, victims are exposed to sexually transmit-
ted diseases and other health problems that worsen due to a
lack of appropriate medical care. Women become pregnant
and experience lasting health problems as a result of poorly
performed or botched abortions, the lack of adequate medical
care, and poorly met nutritional needs. As a result, those who
are released from bondage are often seriously incapacitated
or terminally ill.
Long-Term Consequences
Children born into captivity are exposed from birth to incon-
sistent care. From the earliest period of awareness, babies
experience deeply traumatized and despairing mothers creat-
ing insecure attachment systems and poor internal working
models of relationships. They experience their mothers as
powerless and unable to protect them from their abusers.
Bom into captivity, these psychological conditions leave
many children to identify with a detached and emotionally
absent mother.
Many women who give birth during captivity are minor
children and unable to parent. For these women, their children
are a reminder of the rape and abuse they experienced while in
captivity and the men who abused them. These issues interfere
with the emotions of giving and receiving love as well as a
capacity to establish secure attachment systems for self or
others.
There are long-term consequences for children born into
captivity. If a captive mother is unable to psychologically
464 Child Abuse
nurture a child, that child is at risk of becoming detached,
limiting the emotional capacity for giving and receiving love.
Some captive mothers are able to give and receive love from
their children. When this happens, a child's need for safety may
cause a mother to risk the consequences of an attempt to
escape captivity. In other, more desperate situations, the love
returned by a child may keep a mother from committing
suicide. The loss of psychologically healthy women able to
nurture the next generation of children will have long-term
consequences on society. Consequently, emerging research
and information about human trafficking is worthy of addi-
tional consideration in future reviews of child abuse both as
a distinct type of child abuse and as a consequence of other
forms of child maltreatment.
Risk Factors and Protective Factors
While child abuse is found in all cultures and all socioeco-
nomic strata, social science data and research, mostly from
Western countries, reveal some general risk and protective
factors relative to child abuse. Risk factors are the individual
and familial variables that appear associated with children,
individuals, and families for whom child abuse has been a
problem. Protective factors represent the variables that seem
to thwart occurrences of child abuse for individuals and within
families or communities. These factors should be considered
global indicators, as the presence of, or lack of, any factors does
not predict, or counteract, child maltreatment. Rather, risk and
protective factors can guide assessment of child abuse and
planning for programs to ameliorate its effects on children
and families.
Some suggest that risk and protective factors are additive
features that can be simply balanced out as needed. Specifi-
cally, an increase in protective factors and a reduction in risk
factors would theoretically reduce the chances of harm to a
child. This may be true in some cases but it has yet to be
determined across all aspects of child abuse. Instead, risk fac-
tors should be viewed as signals of concern and protective
factors as areas of strength. The following review of risk factors
and protective factors summarizes some of these factors in the
contexts of parents or caregivers, children, and community.
Risk Factors
Parents or caregivers
For parents or caregivers, risk factors generally correspond to a
lack of resources, materially, emotionally, and socially. Many
parents or caregivers who abused their children were them-
selves victims of abuse or had insecure attachments with their
parents or primary caregivers. Relatedly, abusive parents have
less knowledge or understanding about the emotional or
developmental needs of their children, expecting more than
what might be appropriate for their children's age or attribut-
ing negative attitudes of feelings to their children. These mis-
matches of developmental expectations may be exacerbated by
a parent's own poor romantic relationships or high conflict
within the home with other adults or other children.
Mental health issues are clear risk factors for a parent or
caregiver, especially depression and anxiety. Coupled with
social isolation, single parenthood, and poor social support,
compromised mental health can contribute to elevated stress
that can manifest in child abuse. Moreover, a parent or care-
giver with a history of being abused or neglected is more likely
to repeat that approach with his or her own children when
there is an accumulation of other risk factors.
Alcohol and drug addiction is a major risk factor that seems
increasingly correlated with numerous other social problems
in addition to child abuse and neglect. It is difficult to disen-
tangle other risk factors from drug and alcohol addiction, but it
is the one risk factor most associated with child abuse, possibly
because it is both an outcome and a contributor to other child
abuse risk factors. Additionally, it creates a social scenario and
lifestyle for a parent or caregiver that contains risk for the adults
involved, let alone for the dependent children living with them
and relying on their good judgment.
Children
Risk factors for children begin in utero where a forming child
may be exposed to toxins from a mother's use of drugs or
alcohol. The effects of these toxins can have profound impacts
on a child's lifetime capacity to regulate intellectually and
emotionally. Fetal Alcohol Spectrum Disorder, for example,
can seriously alter a child's ability to read social cues, including
verbal and nonverbal, contributing to challenging social inter-
actions with family and others.
Children with difficult temperaments are prone to abuse
and neglect because they tend to be more challenging behav-
iorally than other children. In fact, behavior problems or
externalizing behavior (acting-out behavior) from children
are often cited by nonabusing parents as well as abusing par-
ents as the most challenging part of parenting. Also at greater
risk of abuse are children with developmental, intellectual, and
physical disabilities as they can be less responsive to caregivers
and may cause additional stress given their special needs.
Further, children who have experienced trauma, including
those maltreated by a parent or caregiver, respond to caregivers
in ways that appear to reject affection and care creating new
risks of maltreatment for those children with histories of
being abused.
Community
Stressful environments or neighborhoods can contribute to
disruptions in parent-child relationships and create contexts
for maltreatment of a child. Parental unemployment and other
stressful family life events, including uncertain housing and
homelessness, can be environmental risk factors for child
abuse. Racial, economic, or cultural discrimination also form
a broad context of systemic risk for children and families that
may potentially elevate risks for child abuse.
Protective Factors
Parents and caregivers
In contrast to risk factors, protective factors for parents and
caregivers correlate with material, emotional, and social re-
sources. Strong, early parent-child relationships defined by
secure attachment systems can help inoculate a child against
maltreatment from a parent or caregiver. This protective factor
depends on a parent's or caregiver's own attachment systems
Child Abuse 465
with his or her own parents or significant others. In fact,
marriage or parenting with another supportive caregiver can
be a protective factor. Networks of supportive friends and
family who can serve as substitute caregivers and with whom
a child can develop independent, secure attachment relation-
ship help protect a child from incidents of abuse. Further,
parental, familial, or community knowledge of child develop-
ment and associated parenting skills connected to a child's
varying needs at different periods of life can help protect a
child against episodes of maltreatment.
Children
Healthy development including a stress-free prenatal period
helps protect a child against child abuse as well as a range of
other health, cognitive, and social challenges. Meeting or
exceeding developmental milestones supports a child's capac-
ity for normative emotional skills and his or her ability to form
a secure attachment system with a parent or caregiver. An
agreeable temperament and adequate cognitive abilities can
also help a child with a range of social tasks with peers and
adults, potentially reducing stress and conflict with a parent
or caregiver.
Community
Supportive extended family and neighborhoods can help form
protective environments for children and families. Formal
and informal community networks, especially those that offer
judgment-free support for parents or caregivers, can help alle-
viate stress at home and help potentiate healthy relationships
between children and their parents or caregivers. Additionally,
positive peer relationships for children and adequate educa-
tional and extracurricular resources are essential components
of a healthy community.
Treatment and Support
Safety
The first step in treating issues of child abuse is child safety.
This may take the form of removing a perpetrator of child
abuse from a home or placing a child temporarily in alternative
care with a safe kin or nonkin foster family. In most cultures,
these initial moves are viewed as temporary to allow for
additional planning to assess the core issues underlying the
history of maltreatment. Viewed in the context of relationship-
disruptions and potential insecure attachment systems, child
abuse interventions reviewed here will be those targeting
young children and their parents or alternative caregivers
(e.g., kin or nonkin foster parents, adoptive parents).
In advance of relationship-enhancing interventions, spe-
cialized treatment may be needed for a maltreating family,
especially the perpetrator as the source of continuing danger,
relative to reabuse. For example, child sexual abuse requires
specialized treatment for perpetrators, usually addressing sex-
ual deviancy with a combination of cognitive-behavioral and
medical methods. Similarly, families and perpetrators of phys-
ical abuse need ongoing treatment addressing cycles of rage
and assault. In these instances, it is likely and necessary that
court or criminal jurisdiction will be applied to a child abuse
matter, creating additional external accountability.
Interventions
Abused children need individual support and interventions to
address issues of trauma associated with abuse and the poten-
tial familial disruptions associated with it. While stability of a
safe home and a secure relationship with a caregiver who will
protect him or her are paramount to the emotional needs of a
child, attention must also be paid to the individual traumatic
events of abuse. Mental health professionals skilled in trauma
counseling should be engaged to support a child's needs as
soon as possible following interventions due to maltreatment.
Maltreatment has dire effects on a child's health and well-
being. Children who are abused are at far greater risk for
developmental, health, and educational problems than the
general population of children. Abused children have more
behavioral and mental health problems than nonabused chil-
dren and are more likely to use mental health services. Atten-
tion to the unique mental health needs of these children is a
critical component of their long-term health.
Using an attachment framework, the following interven-
tions have been demonstrated to be effective with maltreated
children and their parents or caregivers in repairing or im-
proving parent-child relationships. These interventions are
described here as representations of potential models for pro-
fessionals treating abused children and their families. The cate-
gories of parental internal working models, parental behavior,
and parent-intervener relationship reflect the general ap-
proach of the interventions.
Parental internal working models
Child-parent psychotherapy (CPP) is a dyadic intervention for
young children and their parents. It is intended to help parents
come to grips with their own compromised childhood attach-
ment systems to help prevent child maltreatment and im-
prove current parent-child relationships. In general, CPP
helps address insecure attachment issues that undermine
parental capacity and contribute to risks of maltreatment.
Attachment and Bio-behavioral Catch-up is a brief home
visiting program for foster parents and young foster children.
Paying close attention to a foster parent's own attachment
history, home visitors help address a foster child's need for
security and help foster parents understand behavior from a
child in their care that may appear to be rejecting the foster
parents. It helps address problematic internal working models
for the foster parent, redirecting attention to the emotional
needs of a foster child.
A group-based educational approach, circle of security,
helps parents or caregivers better understand a child's relation-
ships needs. Using guided reviews of parent-child interactions,
parents or caregivers learn to better read a child's cues for
security and exploration. Initial studies with families who
experienced the trauma of child abuse indicate that inse-
cure attachment patterns were significantly reduced following
interventions.
Parental behavior
Parent-child interaction therapy is designed for families with
children who are behaviorally disruptive or have mental health
concerns. Using in vivo coaching, parents learn new techni-
ques to address a child's externalizing behavior. Treatment
466 Child Abuse
concludes once parents or other caregivers master the techni-
ques for managing a child's behavior.
Triple P - Positive Parenting Program promotes positive
parenting approaches through a variety of sources: multimedia,
professional consultations, and self-directed modules. Triple P
originated in Australia, and its multilevel program is intended
to prevent severe emotional and behavioral problems in chil-
dren by helping parents or other caregivers manage difficult
behaviors. Triple P has been successful in reducing parent-
reported child behavior problems and parenting problems.
In the Netherlands, models of short-term intense home
visiting have helped mothers of irritable infants and young
children with externalizing behavior problems. After learning
sensitive disciplinary practices and reviewing cues about their
children's behaviors, mothers appeared to be more securely
attached to their children than those who did not participate,
and the effects of the interventions persisted over time. These
models may have broader application across cultures as they
are short term and simple for both the involved parent or
caregiver and home visitors.
Parent-intervener relationship
Two similar interventions with at-risk mothers and their
babies, the UCLA family development project and Minding
the Baby, use the relationship of the intervener with a mother
to establish secure attachment with a baby or expected
baby. Through an established secure base with an intervener,
a participating mother was able to build reflective capacity to
address the emotional needs of her child. With both ap-
proaches, early interventions in support of expectant mothers'
emotional and relational needs helped prevent development
of risk factors for abuse by strengthening protective factors of
secure attachments.
Summary
Child abuse is a global social problem defined broadly as
physical abuse, sexual abuse, neglect, and emotional abuse of
children by adults who are usually family or community
members. Human trafficking involves exploitation of children
who are vulnerable due to extreme poverty, child abuse, and
other difficult social pressures. Maltreatment by parents and
caregivers contributes to insecure attachment systems for chil-
dren, potentially leading to unhealthy internal working mod-
els of relationships that may be transmitted to maltreated
children's subsequent relationships, including those with a
new generation of children. Risk factors relative to child
abuse can help intervening professionals identify those chil-
dren who may potentially be harmed, while protective factors
can help families and communities build personal and envi-
ronmental strengths toward preventing child maltreatment.
Evidence-based treatment programs aimed at enhancing rela-
tionship capacity for children and parents or caregivers are
available to help maltreated children and their families address
severe relationship-disruptions characteristic of child abuse.
See also: Family Systems; Parenting; Parent-Offspring Conflict.
Further Reading
Berlin LJ, Zeanah CH, and Lieberman AF (2008) Prevention and intervention programs
for supporting early attachment security. In: Cassidy J and Shaver PR (eds.)
Handbook of Attachment: Theory, Research, and Clinical Applications, 2nd edn.,
pp. 745-761. New York: Guilford Press.
Cassidy J and Shaver PR (eds.) (2008) Handbook of Attachment: Theory, Research, and
Clinical Applications, 2nd edn. New York: Guilford Press.
Fraiberg S (1987) Ghosts in the nursery: A psychoanalytic approach to the problems of
impaired infant-mother relationships. In: Fraiberg L (ed.) Selected Writings of
Selma Fraiberg. Columbus, OH: Ohio State University Press.
Lawler MJ, Shaver PR, and Goodman GS (2011) Toward relationship-based
child welfare services. Children and Youth Services Review 33: 473-480.
McCabe KA and Manian S (2010) Sex Trafficking a Global Perspective. New York:
Rowman & Littlefield.
Pecora PJ, Whittaker JK, Maluccio AN, Barth RP, DePanfilis D, and
Plotnick RD (2009) The Child Welfare Challenge. New Brunswick, NJ:
Transaction Publishers.
United Nations General Assembly (1990) Convention on the Rights of the Child.
New York: United Nations.
World Health Organization and International Society for Prevention of Child
Abuse and Neglect (2006) Preventing Child Maltreatment: A Guide to Taking
Action and Generating Evidence. Geneva: World Health Organization.
Childhood Mental Disorders
A Scarpa and L Wilson, Virginia Tech, Blacksburg, VA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Child psychopathology The study of psychological
and behavioral disorders in children and adolescents,
taking into account developmental processes and
milestones.
Comorbidity The diagnosis of more than one mental
disorder in any given individual.
Diagnostic and Statistical Manual of Mental Disorders
(DSM-IV) A manual published by the American
Psychiatric Association that provides a common
language and criteria for the diagnosis of mental
disorders.
Prevalence The proportion of individuals in a population
diagnosed with a specific disorder.
Prognosis The likely outcome associated with a diagnosis
or disorder.
Somatic complaints Unexplained medical symptoms.
Taxonomy The science of classifying plants, animals, and
microorganisms into increasingly broader categories based
on shared features.
Introduction
The current field of study for childhood mental disorders is
often referred to as child psychopathology, which involves
the study of psychological and behavioral disorders in children
and adolescents. In order to be considered as meeting the
definition of a mental disorder, there must be a pattern of
symptoms that leads to some level of distress, disability or
interference with daily functioning, and an increase in risk for
pain, suffering, or loss of freedom (e.g., institutionalization).
In addition, the symptoms are not consistent with the norms
of the culture, though they may appear to be exaggerations of
normal behavior or can occur in adaptation to some unusual
circumstance (e.g., chronic illness, maltreatment).
The Diagnostic and Statistical Manual of Mental Disorders
(DSM) was originally developed as a taxonomy to categorize
these patterns of behaviors into distinct mental disorders.
Currently, the DSM is in its fourth edition (DSM-IV) and a
fifth edition is expected to be released in 2013. The DSM-IV
lists numerous disorders as 'usually first diagnosed in infancy,
childhood, or adolescence,' and these disorders will be dis-
cussed in the current article. Mood disorders and anxiety
disorders are also common in children and adolescents, though
in some instances they may not be diagnosed until adulthood.
This review will focus on childhood mental disorders that are
commonly seen in clinic populations, including the mood and
anxiety disorders, attention-deficit and disruptive behavior dis-
orders, and pervasive developmental disorders (PDDs).
Recent estimates suggest that 10-20% ofyouth meet criteria
for a specific psychological disorder. Unfortunately, the major-
ity of children with mental health problems do not receive
services. About 20% of children with mental health problems
will have significant difficulty throughout their lives, suggest-
ing that childhood psychological disorders are long-lasting and
continue into adulthood.
Anxiety Disorders
Anxiety in children is often an unrecognized and undiagnosed
problem because some symptoms of anxiety are normal in
development (e.g., stranger anxiety in toddlerhood), many
symptoms are not easily visible (e.g., increased heart rate),
and the effects are not as damaging as those arising from
disruptive behavior disorders (e.g., fighting, property destruc-
tion). While not easily identified, anxiety disorders in young
people do exist and can be chronic.
Anxiety is a mood state that is often viewed as having three
interrelated components - physical, cognitive, and behavioral.
The physical component is mediated by the sympathetic ner-
vous system and the endocrine system, and essentially involves
the chemical and stress reactions that mobilize the body for
action. This response is sometimes referred to as the fight/flight
response, and prepares us to either flee or defend ourselves in
dangerous situations. The cognitive component includes men-
tal processes that allow us to search for potential sources of
threat. This component can lead to worry or nervous apprehen-
sion. Finally, the behavioral component includes acts associated
with anxiety, particularly aggression, avoidance or escape of the
situation, and other nervous behaviors (e.g., finger tapping).
Therefore, anxiety is characterized by a strong negative emotion
accompanied with bodily fight/flight symptoms, apprehension
of future danger, and a desire to avoid the feared situation.
In general, anxiety is thought to have some evolutionary
survival value by preparing individuals physically and psycho-
logically to cope with potentially dangerous situations. Some
anxiety, therefore, is necessary for normal adjustment and cop-
ing. When anxiety becomes excessive and disabling, however,
the child may have an anxiety disorder. Multiple anxiety disor-
ders are specified in the DSM-IV and they will be reviewed here.
In general, the onset of anxiety disorders has been attribu-
ted to biological, behavioral, and social influences that lead to
the dysregulation of the normal anxiety response. This disrup-
tion then results in undifferentiated anxiety, and ultimately
distress and impairment. For example, being bitten by a dog
may result in an animal phobia, being teased by classmates at
school may result in social anxiety, and biological vulnerability
may result in panic disorder. More specifically, the biological
mechanisms associated with the onset of anxiety include an
overactive behavioral inhibition system, overactive amygdala,
increased cortisol levels, increased heart rate, blood pressure
and electrodermal responding, as well as genetic influences
467
468 Childhood Mental Disorders
(i.e., 33% of the variance in anxiety is believed to be genetic).
Family influences include excessive parental control, rejection,
overprotection, or modeling. In terms of temperament, chil-
dren who are born with a low threshold for novelty and
unexpected stimuli are at greater risk for anxiety disorders.
Countless influences have been identified as playing a role in
the onset of anxiety disorders in children and a few of these
mechanisms have been reviewed here.
Separation Anxiety Disorder
Separation anxiety disorder (SAD) is the only anxiety disorder
in the DSM-IV that is specified as usually the first diagnosed in
infancy or early childhood. SAD involves age-inappropriate,
excessive, disabling anxiety about being away from home
or from those individuals to whom the child is attached.
According to the diagnostic criteria in the DSM-IV, the distur-
bance must last at least 4 weeks and begin before the child is
18 years old.
Separation anxiety is typical in toddlerhood, and so this
disorder may be difficult to distinguish from anxiety that is
developmentally normal. SAD occurs in about 4-10% of chil-
dren and is equally common in boys and girls. Children with
SAD may experience excessive distress when separated from
their major attachment figures. They often need to know the
location of their attachment figures and need to stay in fre-
quent contact with them. They may become very 'homesick'
when away from home and may become worried that some-
thing tragic will happen to their attachment figures while they
are separated. Because of these fears, the children may avoid
traveling independently away from home. For example, they
may refuse to attend school, camp, or a sleepover at a friend's
house. Children with SAD may also refuse to engage in activ-
ities where they are alone, such as sleeping in their room. These
children may sometimes have nightmares about their separa-
tion fears (e.g., being kidnapped), and they may also have
physical complaints such as stomachaches and headaches.
Of all the anxiety disorders, SAD has the earliest onset.
It often occurs after the child has experienced a stressful event
(e.g., the hospitalization of a parent). SAD generally has a good
prognosis, although it persists into adulthood for about one-
third of children and adolescents. Friendships and academics
may suffer as the child avoids school or other typical social
events where they may be separated from their parent. In
addition, children with SAD are at risk for developing other
anxiety disorders or a secondary depression in reaction to their
distress.
Generalized Anxiety Disorder
Generalized anxiety disorder (GAD) is excessive, uncontrolla-
ble anxiety and worry about numerous aspects of the child's
day-to-day life. According to the DSM-IV, it must occur
on more days than not for more than 6 months. The onset of
GAD is often around 10-14years of age. Many people with
GAD report having felt this way all their lives. It occurs in 3-6%
of children, with equal rates in boys and girls, and is one of
the most common anxiety disorders in children.
Children with GAD often find it difficult to control their
worry, which is far out of proportion to the actual likelihood or
negative impact of the feared event. While adults with GAD
tend to worry about everyday matters, children with GAD tend
to worry about their own competence (e.g., grades). They may
have excessive concerns about being punctual and about cata-
strophes, such as earthquakes. Children with GAD may also be
overly conforming, perfectionistic, and uncertain of them-
selves, thus often requiring approval and excessive reassurance.
In addition to this apprehension, the disorder is accompanied
by at least one additional symptom, such as restlessness, fatigue,
difficulty concentrating, irritability, muscle tension, and dis-
turbed sleep (three of these symptoms are required in adults).
It is commonly accompanied by physical symptoms such as
headaches, twitching, shakiness, or muscle tension. GAD has a
high comorbidity with other anxiety disorders, mood disor-
ders, and substance related disorders. The course is chronic,
but appears to fluctuate and worsen in times of stress.
Specific Phobia
Specific phobia is the extreme, disabling fear of specific objects
(e.g., a dog) or situations (e.g., flying) that pose little or
no danger. For example, if a child's fear of costume characters
impedes their ability to attend a friend's birthday party, it may
be considered clinically impairing. According to the DSM-IV,
for individuals under 1 8 years old, the marked and persistent
fear must occur for more than 6 months. Exposure to the feared
stimulus is typically accompanied by an immediate anxiety
response, which in children may be expressed as a temper
tantrum or clinging to an attachment figure. In adults,
DSM-IV criteria require that the individual recognizes that the
fear is excessive or unreasonable; however, this characteristic is
often absent in children and is not required for diagnosis. The
onset of specific phobias can occur at any age, but the highest
prevalence of specific phobias in children occurs between the
ages of 10 and 13. Specific phobias occur in 2-4% of children,
however only a small number of children are referred for
treatment. Specific phobias are more common in girls than in
boys, and the symptoms tend to occur at younger ages in girls
than in boys.
Social Phobia
Social phobia is the marked, persistent fear of and desire to
avoid being the focus of attention or scrutiny, or doing some-
thing humiliating in front of others. Subsequently, individuals
with social phobia avoid situations where they may experience
embarrassment or be perceived as stupid, including eating in
public or speaking in front of the class. In children, the DSM-IV
further specifies that there must be evidence of age-appropriate
social interactions with familiar people (e.g., parents), but
the child must become markedly anxious in peer social situa-
tions (e.g., school). Similar to the diagnosis of specific phobias,
the DSM-IV specifies that, unlike adults, children frequently
do not recognize that the fear is unreasonable, and for indivi-
duals under 18 years old, the fear must occur for more than
6 months. Social phobia in children may also take the form
of selective mutism or the failure to speak in specific situa-
tions. The onset of social phobia typically occurs in early to
mid-adolescence because of the greater emphasis on self-focus
and -consciousness. It occurs in 1-3% of children, appears to
Childhood Mental Disorders 469
increase with age, and is slightly more common in girls than in
boys. Younger children experience more somatic complaints,
such as stomachaches, whereas adolescents report more
concern about visible signs of anxiety, such as trembling or
sweating.
Obsessive-Compulsive Disorder
Obsessive-compulsive disorder (OCD) is a disorder that is
characterized by two types of symptoms: obsessions and com-
pulsions. Specifically, individuals with OCD exhibit repeated,
intrusive, irrational, and unwanted thoughts that cause anxiety
(i.e., obsessions), which are often accompanied by ritualized
behaviors or compulsions to relieve the anxiety (i.e., compul-
sions). For instance, experiencing irrational and repeated fears
of germs (i.e., obsession) may lead a child to engage in hand
washing (i.e., compulsion).
To be diagnosed with OCD, an individual must experience
the symptoms for more than 1 h a day and the symptoms must
cause marked distress or impairment in everyday life. In fact,
OCD is typically associated with severe disruptions in health,
social and family relations, and school performance. For
instance, children with OCD frequently evidence declines in
grades because of the secondary impairment in their concen-
tration. Children diagnosed with OCD are extremely resistant
to reason, even in instances where they recognize the irrational
and obsessive nature of their fears and rituals.
The onset of OCD typically occurs between 9 and 12 years
of age and tends to be gradual. The disorder is twice as
common in boys than in girls during childhood, but the gen-
der ratio equalizes by adolescence. Approximately 2-3% of
children are diagnosed with OCD, the prognosis is poor, and
there is a high rate of comorbidity with other anxiety disorders,
depression and disruptive behavior problems. Fortunately,
most children with OCD show some improvement in func-
tioning following treatment.
Panic Disorder
Panic disorder is characterized by the presence of recurrent
and unexpected panic attacks, as well as at least 1 month of
persistent and marked worry about having another attack
or worrying about the implications of having another
attack. Panic attacks are sudden and overwhelming periods of
intense fear or discomfort accompanied by the flight/ fight
response. Thus, panic disorder includes physical (e.g., elevated
heart rate) and cognitive (e.g., fear of going crazy) elements of
anxiety. Furthermore, situational avoidance may lead to agora-
phobia, which is worry about being in situations where it
would be difficult for one to escape or get help, if one were
to have a panic attack. If agoraphobia is present in addition to
panic disorder, then the diagnosis is coded as panic disorder
with agoraphobia. Conversely, if the individual's symptoms do
not meet DSM-IV criteria for panic disorder, then the symp-
toms are labeled as agoraphobia without history of panic
disorder.
Panic attacks are fairly common in the population; how-
ever, panic disorder is relatively rare (i.e., 0.6%). The typical
onset of panic disorder is 15- 19 years old and is more com-
mon in girls. It is interesting that studies show that 95% of
adolescents with the disorder are postpubertal, supporting
the importance of considering developmental milestones.
Research suggests that panic disorder has the worst prognosis
of all of the anxiety disorders, evidenced by the lowest rate of
remission.
Posttraumatic and Acute Stress Disorders
Posttraumatic stress disorder (PTSD) is a disorder character-
ized by the development of symptomatology following direct
exposure to a traumatic stressor during which an individual
perceived a threat of death or serious injury, or after learning
about the unexpected or violent death or serious injury
of a close associate, such as a family member. Examples of
traumatic stressors include natural disasters, car accidents,
witnessing war, being kidnapped and experiencing sexual
abuse. A diagnosis of PTSD involves three core features.
First, individuals must report persistent reexperiencing of
the event (e.g., flashbacks, nightmares). Second, PTSD is asso-
ciated with the avoidance of associated stimuli and numbing of
general responsiveness (e.g., dissociation). Third, individuals
with PTSD experience extreme arousal, including increased
autonomic reactivity. To meet DSM-IV criteria, the individual
must experience symptoms for more than 1 month and the
symptoms must cause significant distress or impairment.
PTSD in children may take a different form than the
responses typically seen in adults. With regard to reexperien-
cing the event, children may engage in repetitive play involving
themes from the trauma or they may report having nightmares
without distinguishable content. Furthermore, PTSD in adults
is associated with a response of fear, helplessness or horror
following the trauma; however, children may exhibit disorga-
nized or agitated behavior. This last criterion is being recom-
mended for removal from future DSM editions, as it is difficult
to operationalize in children.
Acute stress disorder is the development of symptoms
within 1 month of being exposed to a traumatic stressor, dur-
ing which the symptoms last for 2 days but no longer than
1 month. Thus, acute stress disorder is a short-lived stress
response that involves many of the characteristics associated
with PTSD.
PTSD occurs in ~3.7% of boys and 6.3% of girls in the
United States. Furthermore, 75% of individuals diagnosed
with PTSD also display comorbid depression and/or substance
abuse. Prognosis following the diagnosis of PTSD is influenced
by pretrauma child characteristics, exposure to additional trau-
matic stressors, features of the posttrauma environment, and
specific PTSD symptomatology reported. For example, social
support following exposure to a traumatic stressor is associated
with better prognosis, whereas continued sexual abuse is
typically associated with worse adjustment.
In sum, anxiety disorders in children are associated with
cognitive disturbances, physical symptoms, social and emo-
tional deficits, and high rates of comorbid mental disorders.
These characteristics result in significant impairments in multi-
ples areas of their lives, including academic difficulties, family
conflict and peer problems. There are several therapeutic
approaches that are used to address anxiety symptoms in chil-
dren, such as behavior therapy, cognitive-behavioral therapy,
medications, and family intervention.
470 Childhood Mental Disorders
Mood Disorders
Mood disorders, including depressive disorders and bipolar
disorder (BD), are primarily categorized by a disturbance in
mood and typically involve some form of depression. In the
past, it was mistakenly believed that depression did not exist in
children in a form comparable to adults; however, it is now
known that children do in fact experience depression. Much of
the confusion in the past occurred because childhood depres-
sion frequently co-occurs with more visible disorders and dif-
ficulties, such as aggression and attention-deficit/hyperactivity
disorder (ADHD). Thus, the depression was typically masked
under more observable symptoms. There is evidence now
suggesting that depression is one of the most disabling child-
hood disorders. Furthermore, the prevalence of depression in
children is increasing and the age of onset is decreasing.
Although depression in children has many of the same features
as the disorder seen in adulthood, the features and nature of
the disorder change with age. For example, depressed preschoo-
lers typically present as withdrawn and somber. Conversely,
school-age children with depression tend to be argumentative
and combative. Thus, it is important to take age into account
when examining mood symptomatology.
Numerous causes of mood disorders have been identified.
First, neurobiological influences have been implicated, such as a
heightened stress reaction. Also, brain activity appears to be
attenuated in areas associated with attention and sensory pro-
cesses, but is more active in brain regions linked to recognizing
and regulating emotions. Second, heritability estimates for
depression range from 0.35 to 0.75, suggesting moderate genetic
influences. In fact, if a parent has major depressive disorder
(MDD) then their child is two to three times more likely to
have depression. Family characteristics, such as more anger and
conflict, poorer communication, parental over-involvement, less
warmth and support, more disorganization, and higher levels of
stress are also associated with mood disorders. Finally, stressful
life events, including frequent moves, death of a parent, family
illness, divorce, and violence may increase the likelihood that a
child develops a mood disorder. In sum, no one influence has
been linked to the onset of mood disorders, but several mechan-
isms have been identified that may increase the probability that
a child develops a mood disorder, such as depression.
Depressive Disorders
The depressive disorders include MDD and dysthymic disorder
(DD). MDD is a disorder that is characterized by one or more
major depressive episode, which is a period of at least 2 weeks
during which the individual experiences depressed mood or
the loss of interest in almost all activities. To meet DSM-IV
criteria, the individual must also report four additional symp-
toms, such as significant weight change, insomnia or hyper-
somnia, psychomotor agitation or retardation, fatigue, feelings
of worthlessness, diminished ability to concentrate, or recur-
rent thoughts of death. For diagnosis, it is also necessary to
exclude the presence of prior manic episodes, organic causes,
normal bereavement, and underlying thought disorders. In
childhood MDD, it is common to see academic difficulties
unrelated to intellectual deficits, few close friendships, and
poor relations with parents and siblings. Evidence supports
suicidal ideations in most young people with depression,
with between 16% and 30% attempting suicide.
Although the DSM-IV criteria for MDD are the same for all
ages, children with depression often present differently than
adults. For example, although children with depression may
be sad, they typically cannot verbalize their emotions in the
same way adults frequently do. Therefore, they may attempt
to express their feelings through withdrawal or irritability.
Depressed children are more likely to report somatic com-
plaints than adults, such as headaches. Also, anhedonia, psy-
chomotor retardation, psychosis, lethality of suicide attempts
and impairment of functioning increase with age. Relative to
adults, children also are more likely to be experiencing their
first episode of depression, tend to recover faster, and are at
increased risk of developing BD.
The average length of a depressive episode is 8 months, and
almost all children recover from the episode, but recurrence
is highly probable. Furthermore, even after recovery, children
typically continue to exhibit adjustment and health problems,
and appear to display mild symptoms of depression.
MDD occurs in ~2-8% of children between the ages of
4- 18 years old. It is less common in preschool and school-age
children (1-2%), but the prevalence increases to 8% by adoles-
cence, which is the typical onset age of adult depression. The
increase in prevalence of MDD from preschool to school-age has
been attributed to the child's growing sense of self-awareness,
cognitive capacity, verbal ability, and performance and social
pressures. The subsequent drastic increase in prevalence that
occurs during adolescence is likely due to biological maturation.
There are no gender differences until puberty, after which girls are
two to three times more likely to be diagnosed with MDD. The
lifetime prevalence of MDD is between 10-25% in women and 5 -
12% in men. Comorbidity is very common, with as many as 70%
of individuals with MDD reporting at least one other disorder,
such as conduct problems or ADHD. Most co-occurring disor-
ders precede MDD and persist after the child is longer depressed.
DD is similar to MDD, but is less severe and more chronic.
Specifically, individuals with DD exhibit depressed mood more
days than not for at least 1 year, have at least two additional
symptoms of depression, and are never symptom-free for more
than 2 months. This criterion is different than in adults, which
requires a duration of at least 2 years. DD is also associated with
less anhedonia, social withdrawal, impaired concentration,
death thoughts, and physical complaints than MDD. Con-
versely, children with DD tend to display more constant sadness,
self-depreciation, low self-esteem, anxiety, irritability, anger, and
temper tantmms. Individuals that have both MDD and DD are
said to have 'double depression,' which is associated with
longer, more severe depressive episodes, higher rates of other
mental disorders, more suicide attempts, and worse social
impairment. DD is less common than MDD, with prevalence
rates of ~1% in children and 5% in teenagers. The onset of DD
is earlier than MDD (ll-12years old) and the average episode
length is 2-5 years. Most children with DD recover, but they
are then at greater risk of other disorders, particularly MDD.
Bipolar Disorders
The essential feature of BD is periods of depression alternating
with or accompanied by abnormally and persistently elevated,
Childhood Mental Disorders 471
expansive, or irritable mood. During this time, individuals
with BD may display symptoms of overexcitement, restless-
ness, agitation, sleeplessness, pressured speech, flight of ideas,
sexual disinhibition, inflated self-esteem, and reckless behav-
ior. Manic episodes in adolescents are more commonly asso-
ciated with psychotic features, school truancy, school failure
and substance abuse than MDD.
There are several subtypes of BD differentiated based on
whether the child displays a manic (i.e., elevated or irritable
mood lasting at least lweek or requiring hospitalization),
mixed (i.e., both manic and depressed episodes within a
lweek period), or hypomanic episode (i.e., manic for 4 days
but with no deterioration of functioning). Bipolar I is having
one or more depressive episodes alternating with one or more
manic/mixed episodes. Bipolar II is having one or more depres-
sive episodes alternating with one or more hypomanic episodes.
Cyclothymia is at least lyear of numerous hypomanic and
depressive symptoms that do not meet full criteria for manic,
mixed, or depressive episodes, no more than 2 months without
symptoms, and significant distress or impairment.
The lifetime prevalence of BD is 0.4- 1.2%, and is less
common than MDD in children. BD is rare in young children,
but the prevalence rate increases after puberty when it becomes
comparable to adult rates. Research suggests that early signs of
BD include irritability and anger difficulties, as well as fluctua-
tions in energy levels. However, it is important to note that
these symptoms are also associated with other psychiatric dis-
orders. Typically, the first episode occurs during adolescence
(15- 19 years old) and depression normally appears first. In
fact, ~50% of individuals experience depression first, 25%
report mania and 25% report a mixed episode. In young peo-
ple, it is common to exhibit milder Bipolar II and cyclothymia
rather than Bipolar I. BD is equally common in males and
females; however, early onset is more common in boys. Comor-
bidity is common with anxiety disorders, ADHD, conduct dis-
orders (CDs) and substance abuse. BD is chronic and resistant
to treatment, with poor long-term prognosis.
PediaUic BD is quite conttoversial and although it only
appeared in the literature during the past decade, the rate of
diagnosis has increased exponentially. In fact, prior to 1990, BD
was not thought to exist in children. Specifically, in 1996, BD was
the least frequently diagnosed psychiatric disorder among in-
patient children; however, in 2004, it was the most frequently
diagnosed disorder. This increase in the prevalence of pediatiic
BD is thought to be largely due to a shift in the definition of the
disorder. That is, since DSM-IV, a milder form of mania (i.e.,
hypomanic) is recognized, which is referred to as Bipolar II. As
previously noted, Bipolar II is more common in children. Prior to
the aforementioned change, children who are now diagnosed with
BD would likely have been diagnosed with CD or ADHD. This
broadening of the symptomatology of BD has therefore resulted
in a greater prevalence of BD, particularly in children, because
Bipolar II is more common in this age range than Bipolar I.
Attention-Deficit and Disruptive Behavior Disorders
Attention-Deficit/Hyperactivity Disorder
ADHD is a persistent pattern of severe inattention and or
hyperactivity-impulsivity that persists for at least 6 months.
Some of the symptoms need to have been present prior to
7 years of age. Moreover, impairment in functioning must
occur in at least two settings (e.g., in school and at home).
ADHD is estimated to occur in 4-8% of all school-aged chil-
dren. It is three times more likely to occur in boys and is more
prevalent in groups of lower socioeconomic status.
Difficulties with attention seem especially prominent dur-
ing activities that are long, repetitive, and tedious. The child
may fail to pay attention to details, make careless mistakes,
produce messy work, appear as though they are not listening,
frequently forget or misplace things, and have difficulty com-
pleting work. Children with ADHD tend to perform better on
self-paced tasks that they have chosen, in novel settings, in one-
to-one situations, and when the activity is especially engaging
or offers immediate reward.
In the DSM-IV, hyperactivity and impulsivity are considered
to be a single dimension and reflect a deficit in behavioral
inhibition. This factor refers to behavior that is excessively
energetic, intense, and not goal-directed. Hyperactivity or over-
active motor behavior includes fidgeting, difficulty staying
seated, moving, running, climbing about, excessive talking,
and appearing as if driven by a motor. Impulsive behaviors
include difficulty in stopping an ongoing behavior, inability to
resist immediate gratification, responding too quickly or blurt-
ing out answers, interrupting others, and difficulty waiting for
one's turn.
The DSM-IV specifies three subtypes of ADHD, depending
upon the predominant symptoms in the last 6 months. The
first subtype is the combined type, which accounts for most
children with ADHD. The combined type is classified when six
or more symptoms of inattention and six or more symptoms of
hyperactivity-impulsivity have persisted for the last 6 months.
The predominantly inattentive type (ADHD-PI), which occurs
in about 1% of schoolchildren, is diagnosed if at least six
symptoms for inattention have persisted, but the child displays
fewer than six symptoms of hyperactivity-impulsivity. Finally,
the predominantly hyperactive-impulsive types (ADHD-HI) is
used when six or more symptoms of hyperactivity-impulsivity
have persisted, but the child displays fewer than six symptoms
of inattention. ADHD-PI is frequently associated with learning
disorders, slow processing speed, difficulties with information
retrieval, and anxiety/mood symptoms. The combined and
ADHD-HI types are associated with aggressiveness, defiance,
peer rejection, and placement in special education classrooms.
Literature on the various subtypes is ongoing, and it is not yet
clear if these subtypes are clearly distinct, or if they represent
different stages in the developmental course of the disorder.
Although children with ADHD may have normal levels of
intelligence, studies have indicated that IQ scores range about
5-9 points lower than children without ADHD. Lower IQ scores
may be due to deficits in working memory, speed of processing,
sustained attention, and test-taking behaviors. Many children
with ADHD experience difficulties with school, including aca-
demic underachievement, being held back a grade level, place-
ments in special education classrooms and increased school
dropout. About 25% may have a diagnosable learning disorder
and 30-60% may have speech and language impairments.
Many children with ADHD have difficulties with sleep, and
may also experience motor clumsiness or repetitive motor
movements (i.e., tics). Impulsivity may lead to accidents and
472 Childhood Mental Disorders
risk-taking behaviors, including more pedestrian and traffic
accidents, unsafe sex practices and increased use of substances.
Others may perceive the child as noncompliant or annoying,
which can lead to high levels of conflict with family members
and peers. It is also likely that children with ADHD will display
symptoms of oppositional defiant disorder (ODD) (about
50%) or CD (about 30-50%), which further increases their
social difficulties. They may also experience comorbid anxiety
(about 25%) and depression (about 20-50%). Finally, many
children with ADHD report a higher level of self-esteem than
would be expected in light of their behavior. That is, they have
an exaggerated perception of their own competence, and this
has been referred to as a positive illusory bias.
Developmentally, hyperactive-impulsive symptoms tend
to appear first, especially in the preschool years. In retrospect,
parents may report that their child had a 'difficult' tempera-
ment. Although the disorder may be present at this early age,
and possibly from birth, difficulties are usually not identified
until the school-age years (around ages 5-7), when the child is
faced with the demands of school and is required to engage in
activities that are long, tedious, and repetitive. The prognosis
is poorer for children with co-occurring behavioral problems.
As the child matures into adolescence, about 25-50% of
children outgrow the disorder or learn to cope with its symp-
toms. However, most children still exhibit the disorder into
adolescence. In general, hyperactive-impulsive symptoms tend
to decline with age, while inattention remains stable over time.
It is not clear what causes ADHD, though it is currently
viewed as a neurobiologically based disorder. Genetic evidence
suggests that ADHD runs in families, with a concordance
rate of ~35%, and twin studies show heritability estimates as
high as 80%. Pregnancy and birth complications, smoking
and substance abuse during pregnancy, and minor physical
anomalies have also been implicated, suggesting neural mal-
development. In particular, the prefrontal cortex, basal ganglia
and cerebellum appear to be involved, which are associated
with attention, executive functions, delayed responding and
response organization.
ADHD symptoms and associated impairments are often
treated with stimulant medications. These medications are
effective in about 80% of children with ADHD by increasing
sustained attention, impulse control, and persistence of work
effort, and by decreasing task-irrelevant activity and disruptive
behavior. It is clear, however, that symptoms are never
completely brought to normal levels, and they return as soon
as the medication is stopped. Behavior therapy via parent
management training and classroom management also have
been found to be effective in making short-term behavioral
gains and improving relationships with parents and teachers.
Oppositional Defiant and CDs
Oppositional defiant and CDs involve age-inappropriate
actions and attitudes that violate family and societal rules
and rights of others. Behaviors vary widely and may range
from defiance and noncompliance to assault and murder.
While general surveys indicate that antisocial acts are relatively
common among youth (i.e., about 30-50% of children report
some antisocial act), it is less common for children and ado-
lescents to engage in persistent and serious rule breaking.
It is this small percentage of youth, ~3-5% of youngsters,
which is of particular concern. Their antisocial behavior tends
to remain stable over time and lead to a myriad of associated
problems such as physical injury, academic underachievement,
peer rejection, exacerbated family conflict, and psychopathol-
ogy. Moreover, this small number of youth contributes to a
disproportionately large number of antisocial acts.
In the DSM-IV, children with these sorts of persistent rule-
breaking and antisocial behaviors fall under the categories of
ODD and CD, which have prevalence rates ranging from 6%
to 16% and are more likely to occur in males. In order to be
diagnosed with ODD, the child must show four or more
symptoms of age-inappropriate stubborn, irritable, and defiant
behavior that persisted over 6 months. Symptoms include
losing temper, arguing with adults, active defiance or noncom-
pliance, deliberately annoying others, blaming others for mis-
takes or misbehavior, being touchy or easily annoyed, anger
and resentfulness, and spitefulness or vindictiveness.
CD reflects a repetitive and persistent pattern of behavior in
youth that violates the rights of others. Criteria for the disorder
include aggression toward people and animals (i.e., bullying,
threatening, intimidating others; initiating physical fights;
using a weapon to cause harm; using physical cruelty; stealing
while confronting the victim; and forcing someone into sexual
activity), property destruction (i.e., deliberately setting a fire or
destroying another's property with the intent to cause dam-
age), deceitfulness or theft (i.e., breaking into another's house,
building or car; lying or conning to avoid obligations or to get
something; stealing without confronting the victim), and seri-
ous rule violations (i.e., repeatedly staying out past curfew
beginning before age 13; running away from home at least
twice or for a lengthy period; being truant from school begin-
ning before age 13). Three or more symptoms must have been
present for the last 12 months, with at least one in the last
6 months.
CD is further broken down into childhood-onset type
and adolescent-onset type, based upon the age of onset.
Childhood-onset represents those children for whom at least
one symptom began prior to age ten. These children are usually
male, display frequent physical aggression, have disturbed peer
relationships, have had ODD in early childhood, and usually
show full symptoms for CD before puberty. Adolescent-onset
represents those for whom no symptoms were present prior to
age 10. Relative to the childhood-onset type, these youth are
less aggressive, have a lower male to female ratio, have more
normative peer relationships, and are less likely to persist in
their conduct symptoms. This typing is based upon work look-
ing at the development of antisocial behavior, which suggests a
better prognostic picture for the adolescent-onset type.
Children with ODD or CD tend to have normal levels of
intelligence, but scores generally fall about eight points lower
than their peers, with greater deficits in verbal intelligence.
Some studies also indicate difficulties with executive cognitive
functions, which may interfere with the ability to learn from
consequences. Academic underachievement and learning dis-
orders are also common, which can lead to grade retention and
school dropout. These cognitive deficits are sometimes, but not
always, accounted for by the presence of ADHD. Youth with
conduct problems may also show inflated self-esteem (i.e.,
positive illusory bias) that is easily threatened. They may have
Childhood Mental Disorders 473
difficulty with peers due to their aggressive and threatening
behavior. Other associated problems include family conflict,
instability in the home, and harsh discipline or deficient
parenting practices. Finally, these youth have a higher rate of
premature death, risk of personal injury, early onset of sexual
activity, substance abuse, and comorbidity with ADHD,
depression and anxiety.
Studies on the course of conduct problems from childhood
to adolescence suggest that ODD developmentally precede CD
in the majority of cases. These children begin with minor
behavior problems in early childhood that develop into more
serious problems in later childhood and adolescence. These
studies also indicate that typically, without intervention,
conduct problems tend to remain stable or worsen over time.
Of all children with ODD, ~25% desist from further antisocial
behavior, ~50% maintain a diagnosis of ODD, and ~25%
progress to a diagnosis of CD. The course of development
may also differ between childhood-onset and adolescent-
onset CD, with greater persistence and escalation of behavior
occurring in childhood-onset type.
Antisocial behavior in youth is likely to have multiple
causes, any combination of which may be acting in a given
child. Biosocial models suggest that the interplay of biological,
psychological, and social influences contribute to severe and
persistent antisocial behavior problems. Twin and adoption
studies support a genetic contribution, suggesting that traits
such as impulsivity or difficult temperament may be inherited.
Neurobiological studies have implicated frontal lobe deficits
leading to executive cognitive dysfunction and the persevera-
tion of inappropriate behavior. Psychophysiological studies
have also implicated low heart rate in individuals who are
aggressive, suggesting that low arousal may lead to sensation-
seeking and risky behaviors such as aggression. Low arousal
may also indicate a fearless attitude, in which the child does
not become anxious at the thought of punishment or negative
consequences for their behavior. Cognitive influences could
include a biased way of processing social information, so that
the child makes hostile interpretations of other's actions or views
antisocial behavior as a means to a positive outcome. Family
influences include coercive interaction patterns, in which
parent-child dyads leam to use increasingly intense behavior
to coerce the other person into submitting to their demands.
Family instability and stress, harsh or abusive discipline, parental
criminality, and parental psychopathology are also commonly
associated risk factors for ODD and CD. Finally, societal
influences include neighborhoods with criminal subcultures,
community violence, poverty, or high social disorganization.
There are few effective interventions with empirical sup-
port. Parent management training is especially useful for
younger children with ODD. This behavioral intervention tea-
ches parents skills for managing their child's behavior through
the use of ignoring, attention, reinforcement, and punishment.
Cognitive problem-solving skills training also has some sup-
port. It is typically used in school settings and teaches children
a specific strategy for finding nonaggressive solutions to
problems. Finally, multisystemic treatment has support as a
treatment for severely troubled youth. Based on a systems
approach, this treatment is carried out with all individuals in
all settings of the child's life to address shortcomings at multi-
ple levels.
Pervasive Developmental Disorders
PDDs include an array of chronic and lifelong neuro-
developmental disorders that affect children's social interac-
tions, language/communication, and behavior. In the DSM-IV,
these disorders include autism, Asperger's disorder, Rett's
disorder, childhood disintegrative disorder (CDD), and PDD-
not otherwise specified (NOS). They affect whether children
develop speech, form social relationships, and process sensory
information.
Autism Spectrum Disorders
Together, autism, Asperger's disorder, and PDD-NOS are
referred to as autism spectrum disorders (ASDs). They are con-
sidered spectrum disorders because children vary in the symp-
toms they display and the severity of the disorder. High
functioning autism and Asperger's disorder are often similar in
presentation and there is a controversy over whether they are
truly distinct disorders. Autism is thought to be the fastest
growing developmental disability in the country. Recent statis-
tics from the Centers for Disease Control and Prevention indi-
cate that as many as 1 in 1 10 children are affected with an ASD.
As such, more children are being recognized and require
services.
The diagnosis of autism requires impairments in three key
areas - social functioning, communication, and repetitive
behaviors, interests or activities - with onset prior to age
three. The primary difference between the diagnosis of autism
and Asperger's disorder is that Asperger's disorder does not
require delays in communication. Also, in Asperger's disorder,
there is no clinically significant delay in cognitive develop-
ment, self-help skills, adaptive behavior, and curiosity about
the environment. However, research has not clearly shown
distinctions between Asperger's disorder and high functioning
autism in outcome, course/outcome, etiology, neurocognitive
profile, or treatment needs. As such, current recommendations
are that Asperger's disorder be subsumed under the autism
category in future editions of the DSM.
Impairments in social interaction include nonverbal beha-
viors (e.g., eye contact, gestures) to regulate social interaction,
peer relationships, spontaneous sharing of enjoyment, and
social or emotional reciprocity. Impairments in communica-
tion include delay or lack of spoken language, initiating or
sustaining a conversation, stereotyped or repetitive language,
make-believe or social imitative play. Impairments in beha-
viors include preoccupation with an interest that is abnormal
in intensity or focus, inflexible adherence to routines or rituals,
motor mannerisms (e.g., hand-flapping, twisting), and persis-
tent preoccupation with parts of objects.
Children with ASDs often evidence skills that are scattered,
meaning that they may have deficits in some areas and
strengths in others. Up to 80% of children with autism fall in
the mental retardation range of intellectual functioning, with
particular weaknesses in verbal IQ (using traditional tests,
which may not be appropriate). About 25% have splinter skills
(above average skills in one area like math) and 5% have
savant abilities (extraordinary or supernormal talent). Other
cognitive deficits might include difficulty in understanding
social situations, impairments in the ability to understand
474 Childhood Mental Disorders
others' and their own mental states (Theory of Mind), deficits
in executive functions (orienting, shifting attention, organiza-
tion, reasoning), and lack of drive for central coherence (do
not filter out details). On the other hand, these individuals
may be very good at sustained attention and memory for
details. Some individuals experience sensory impairments
including over- or under-sensitivities, and possibly both at
the same time.
Raising a child with autism is stressful. For example, the
child may be hyperactive, self-injurious, wandering, or may
not sleep well, in addition, parents may be socially ostracized
by friends and strangers and may experience frustration and
delays before receiving help. Co-occurring symptoms include
hyperactivity, learning disabilities, anxieties/fears, mood pro-
blems, sleep difficulties, and self-injurious behavior. Epilepsy
is also a common co-occurring medical condition.
ASD occurs from birth or early in childhood, although
deficits are often not noticed until around age two when the
lack of language and other skills become especially evident.
At 12-18 months, children with ASD may show a lack of point-
ing, showing, orienting to their spoken name, and looking at the
face of another person. Often, there are gradual improvements
with age, though they are likely to continue to experience many
difficulties. Hyperactivity, self-injury, and compulsions appear
to worsen in about a one-third of adolescents with ASD.
Typically, this is a chronic and lifelong condition; however,
variability in outcome depends on language abilities and IQ.
There is no known cause for ASD. Sometimes, problems
have been noted during pregnancy or birth, such as premature
birth, bleeding in pregnancy, toxemia, and viral infections.
Family and twin studies suggest that the heritability of an under-
lying liability to autism is above 90%, and studies of family
members suggest that there may be a Broader Autism Phenotype
in which less severe symptoms are inherited. Neurobiological
studies have implicated the frontal lobes, cerebellum, medial
temporal lobe, and limbic system structures, which can reflect a
brain system specialized for social cognition. The amygdala, in
particular, may be implicated in deficits of orienting to social
stimuli, motor imitation, empathy, and joint attention.
There are currently no empirically validated treatments for
ASD, though there is a great deal of empirical support for early
intensive behavioral interventions. These comprehensive treat-
ment programs are recommended to begin before 5 years of
age, occur in the home or preschool, actively involve family
members and peers, and involve one-to-one work with the
child for 15-40 h per week. The interventions use behavioral
principles to address the core deficits and excesses in behavior
that are commonly seen in ASD. Some evidence supports
cognitive-behavioral treatments for children with Asperger's
disorder and higher functioning autism. These treatments
teach children that thoughts and knowledge can affect feelings
and behavior. They often use a skill-building approach to teach
social skills, affective education, and management of comorbid
disorders like depression or anxiety.
Rett’s Disorder and CDD
Other PDDs in the DSM-IV include Rett's disorder and CDD.
Rett's disorder is a severe neurobiological disorder character-
ized by deceleration of head growth between 5 and 48 months
of age, loss of previously acquired hand skills and subsequent
development of stereotyped hand movements (e.g., hand-
wringing), loss of social engagement that may reappear later,
appearance of poorly coordinated gait or trunk movements,
severely impaired language development, and psychomotor
retardation. It is a disorder that occurs primarily in females
and is thought to be caused by a specific X-linked gene
mutation. The prevalence rate is ~l-4 per 10 000 females.
The long-term prognosis is poor. Rett's disorders is associated
with mental retardation, epileptic seizures, motor handicaps,
and difficulties with communication. Some studies have indi-
cated that the brain sizes of children with Rett's syndrome are
about 12-34% smaller than the brains of other children. The
inclusion of Rett's syndrome in the PDD category has been
questioned and this disorder may be removed entirely in
future editions of the DSM. The primary rationale for removal
of the disorder is that these children often have PDD symp-
toms for only a brief period during early childhood, and social
interactions often reemerge later in childhood, so inclusion in
the autism spectrum may not be appropriate for most indivi-
duals. Moreover, since Rett's disorder has a specific etiology, it
can be specified as a known medical or genetic condition
associated with PDD, rather than its own category as a mental
disorder.
CDD is characterized by a significant loss of previously
acquired language, social skills, and adaptive behavior prior to
age 10, following a period of apparently normal development in
at least the first 2 years of life. Recent estimates suggest that CDD
is rare, occurring in about 2 per 100 000 children. This diagnosis
was included in the DSM-IV primarily to encourage research on
the disorder, though little research still exists. It has been recom-
mended that CDD be subsumed under the autism diagnosis in
future editions of the DSM because there is no evidence that this
disorder is different from Autism with a regressive course. Also,
some case studies indicate that lost skills reemerged following
treatment, suggesting the possibility that the symptoms were
instead due to some treatable neurological condition.
See also: Anxiety Disorders; Attention; Attention Deficit Hyperactivity
Disorder; Autism and Pervasive Developmental Disorders.
Further Reading
American Psychiatric Association (1 994) Diagnostic and Statistical Manual of Mental
Disorders , 4th edn. Washington, DC: Author.
Arcelus J and Vostanis P (2005) Psychiatric comorbidity in children and adolescents.
Current Opinion in Psychiatry 18: 429-434.
Beauchaine TP and Hinshaw SP (2008) Child and Adolescent Psychopathology.
Hoboken, NJ: Wiley.
Egger H and Angold A (2006) Common emotional and behavioral disorders in
preschool children: Presentation, nosology, and epidemiology, c/ot/ma/ of Child
Psychology and Psychiatry 47: 31 3-337.
Garber J, Gallerani CM, and Frankel SA (2009) Depression in children. In: Gotlib IH and
Hammen CL (eds.) Handbook of Depression, 2nd edn., pp. 405-443. New York:
Guildford.
Jensen PS, Arnold LE, Swanson JM, et al. (2007) Three-year follow-up of the NIMH
MTA study. Journal of the American Academy of Child and Adolescent Psychiatry
46: 989-1002.
Kazdin AE and Weisz JR (2003) Evidence-Based Psychotherapies for Children and
Adolescents. New York: Guilford.
Childhood Mental Disorders 475
Lovaas 01 (1987) Behavioral treatment and normal educational and intellectual
functioning in young autistic children. Journal of Consulting and Clinical
Psychology 55: 3-9.
Mash EJ and Wolfe DA (2007) Abnormal Child Psychology, 4th edn. Belmont, CA:
Wadsworth.
Ollendick TH and King NJ (2000) Empirically supported treatments for children
and adolescents. In: Kendall PC (ed.) Child and Adolescent Therapy:
Cognitive-Behavioral Procedures, pp. 386-425. New York: Guilford.
Pelham WE Jr., Wheeler T, and Chronis A (1998) Empirically supported psychosocial
treatments for attention deficit hyperactivity disorder. Journal of Clinical Child
Psychology 27: 190-205.
Sofronoff K, Attwood T, and Hinton S (2005) A randomised controlled trial of a CBT
intervention for anxiety in children with Asperger syndrome. Journal of Child
Psychology and Psychiatry 46: 1 1 52-1 1 60.
VanVoorhees E and Scarpa A (2004) The effects of child maltreatment on the
hypothalamic-pituitary-adrenal axis. Trauma, Violence and Abuse 5: 333-352.
Volkmar FR, Lord C, Bailey A, Schultz RT, and Klin A (2004) Autism and pervasive
developmental disorders. Journal of Child Psychology and Psychiatry 45: 135-170.
Wells KC, et al. (2000) Psychosocial treatment strategies in the MTA study: Rationale,
methods, and critical issues in design and implementation. Journal of Abnormal
Child Psychology 28: 483-505.
Relevant Websites
http://www.aacap.org/ - American Academy of Child and Adolescent Psychiatry.
http://www.cdc.gov/ - Centers for Disease Control and Prevention.
http://www.apa.org/ - Home Page for American Psychological Association.
http://www.dsm5.org/Pages/Default.aspx - Home Page for American Psychiatric
Association DSM 5 Development.
http://www.aboutourkids.org/-New York University Child Study Center.
Chunking
A L Gilchrist and N Cowan, University of Missouri, Columbia, MO, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Adopted chunk method A method of measuring the
number of chunks adopted by participants in free recall by
examining whether the order of recalled information
matches its initial presentation order; each series recalled in
order is considered a chunk adopted by the participant.
Chunk A unit of storage in memory containing multiple
items that are grouped or bound together to form a
unified whole.
Chunk access measure An estimate of the number of
chunks in working memory; based on the assumption that if
at least one item from a presented group is recalled, working
memory has accessed a unit corresponding to that group.
Chunk completion measure An estimate of the integrity of
a presented group of items or interassociation between the
items; based on the assumption that if the group was stored
as a well-integrated chunk and the chunk was accessed
(i.e., at least one item recalled from it), then recall of
the group should be nearly complete, including most or
all of the items presented in that group.
Chunking A process or strategy that increases the amount of
information that can be stored in memory by meaningfully
grouping information together.
Immediate memory Portion of memory that is currently in
our conscious awareness and is easily reportable; also
known as short-term memory.
Transitional error probabilities (TEPs) method A method
of measuring number and size of chunks in serial recall by
examining the probability in making an error during recall
of items; the proportion of errors in transition from one
item to the next in recall, or TEP, should be smaller within a
chunk and larger between chunks.
Introduction
Chunking is a strategy or process by which one can efficiently
increase the amount of information that can be stored in short-
term memory by reorganizing information in a meaningful
way, or by finding patterns within a set of items to be remem-
bered. Chunking may also involve grouping stimuli together
on the basis of perceptual principles such as the similarity or
proximity of certain items to be remembered. However, this
form of chunking does not typically involve strategic effort in
forming chunks. For present purposes, the more strategic
method of chunking, in which long-term knowledge must be
accessed or constmcted to form larger groups of items, will
be of particular interest. For a simple demonstration of chunk-
ing, consider the following scenario. You are presented with a
string of letters, MAIBIFBCI. Most likely, you will not remem-
ber all nine letters. If the string is rearranged to FBICIAIBM, you
will probably remember all of the letters. Why? The letters can
be grouped into three familiar acronyms; the number of sepa-
rate items be recalled is now much smaller. As a result of this
grouping, the amount of information that can be stored has
effectively been increased. This is a very simple illustration of
one way in which people can use chunking to remember more
items than they would (or could) otherwise.
Chunking is arguably one of the most powerful ways to
increase the amount of information that can be stored in
memory. In this article, we will discuss what is currently
known about chunking. First, we will review a seminal work
on chunking and capacity limits in memory, George Miller's
'The Magical Number Seven, Plus or Minus Two.' Following
this we discuss the purpose of chunking and its role in the
capacity-limited immediate memory (which is sometimes also
known as working memory or short-term memory), as well as
a theoretical model of how chunking operates. We then raise
the issue of measuring chunks, discussing several different
ways to do so. Finally, we will close the article with questions
about chunking that are worth examining in future research.
What Is Chunking?
Published in 1956, George Miller's article “The magical number
seven, plus or minus two: Some limits on our capacity for processing
information" has fundamentally shaped our understanding of
capacity limits in immediate memory and how these limita-
tions may be circumvented through the use of chunking. In
this review, Miller noted that seemingly unrelated tasks had a
puzzling commonality: an upper limit in the number of items
(approximately between 5 and 9) that could be processed or
stored at any given time.
Miller reviewed capacity limits for tasks that tested immedi-
ate recall of stimuli. In these tasks, participants are presented
with a list of items (e.g., words, letters); following the presen-
tation of the list, participants are to recall as many items from
the list as possible. Across various types of immediate recall
tasks, it was found that memory span was consistently between
five and nine items, regardless of stimulus type. Similar capac-
ity limits can be found for tasks that examine the absolute
judgment of stimuli that vary in more than one dimension,
such as lines that differ in length and angular orientation.
Though these tasks differ somewhat in terms of methodology,
in both cases, stimuli in these tasks involve multiple features
that are bound together to form a single, unified item. Because
of these similarities, the limits on how many multifeature
objects can be held in mind theoretically could be the same,
regardless of task type.
Although the consistency of capacity limits across various
immediate memory tasks was the primary emphasis of the
476
Chunking 477
review, Miller closed his paper with a discussion of recoding
(by which he meant the same thing that was later termed
chunking) and its implications for immediate memory perfor-
mance. Recoding was defined by Miller as a process of group-
ing or organizing input information into meaningful units, or
chunks. Because information in immediate memory is limited
to approximately seven items, grouping items together in a way
that allows them to be more easily remembered can effectively
increase the amount of information that can be stored and,
eventually, recalled.
To illustrate the phenomenon of chunking, Miller reviewed
a prior study where participants were to immediately recall
sequences of binary (e.g., 101000100) and octal digits. In
general, persons can correctly recall approximately nine of
these digits when presented with one of these sequences. It
was possible, however, that the number of binary or octal
digits that were correctly recalled from memory could be
increased by recoding the digit information in a more mean-
ingful way. For example, consider the binary sequence of nine
digits that was presented above. Miller suggested that more
digits from the list could be recalled if the digits within the
sequence were reorganized into a series of groups (e.g., 101,
000, and 100). These items might be even better recalled,
however, if one knows how to convert binary digits into digits
with a different base (e.g., decimal digits, which are based on a
base of 10). If one groups the binary digit list in the manner
above, 101 equals 5, 000 equals 0, and 100 equals 4 - grouping
digits into larger chunks reduces demand on memory, as does
converting the digits into a more meaningful form. Initially,
participants only recalled about nine binary digits and seven
octal digits; however, once they learned the recoding schemes
similar to the one described above, span for these digits
increased.
This illustration of recoding and chunking of items that is
described above has little ecological validity outside of a labo-
ratory setting; persons are rarely required to convert digits from
one base to another in their everyday lives. However, Miller
himself noted that using chunking or some means of recoding
as a strategy to increase the amount of information that could
be stored in memory was quite possible in an everyday setting.
In particular, Miller suggested that the most common uses of
these sorts of strategies were for the translation or grouping of
verbal materials, though chunking is advantageous in memor-
izing nonverbal stimuli as well.
For example, much of what we know about chunking of
nonverbal stimuli originates from studies that have examined
immediate memory for various positions in chess. In a series of
experiments, William Chase and Herbert Simon presented
both chess masters and novice players with different config-
urations of chess pieces on a playing board. Participants were
shown these configurations for 5 s, which were later covered up
so that participants could not see them. In a critical manipula-
tion, these configurations were either possible chess positions
or were entirely random. Once the configuration was hidden
from view, participants were to recreate the configuration
from memory. When configurations were actually permissible
within the context of a chess match, masters were more likely
than novices to remember the configurations accurately. How-
ever, for random configurations, the memory performance of
masters was no greater than performance for novices. When
Chase and Simon examined how various chess pieces were
chunked, they found that masters were more likely to group
chess pieces based on prior knowledge about chess - for exam-
ple, they might classify a subset of pieces as an attacking or
defensive position. This is starkly contrasted to the grouping
strategy of novices, who were more likely to group subsets of
pieces based on more shallow aspects (e.g., noting that a group
of pieces form a candlestick). The findings from this study
clearly demonstrated the role that prior knowledge can have
in terms of what is retrieved from memory and how chunks are
organized.
The Purpose of Chunking
One of the most important conclusions in Miller's seminal
paper was that immediate memory, where we are able to
hold information in mind briefly, is capacity-limited. It may
be unclear to some why Miller closed his paper with a discus-
sion of chunking; however, we forward the idea that chunking
cannot be understood without a discussion of capacity limits
in memory. Chunking is a way to circumvent capacity limits in
memory to an extent. To illustrate this point, it is important to
discuss any potential limitations on the amount of informa-
tion that can be attended at any given point in time. As chunk-
ing can only be understood in the context of memory capacity
limits, we discuss evidence supporting a four-item limit in
immediate memory below; we follow this with a discussion
of the implications of capacity limits upon chunking.
Evidence for a Four-Item Limit in Immediate Memory
Why is there a great deal of evidence for a memory capacity of
four items when Miller initially proposed that immediate
memory was limited to approximately seven items? It is possi-
ble that in these prior studies that Miller reviewed, small
chunks could have been formed during item presentation,
allowing participants to recall a higher number of items. Inter-
estingly, when items are either too numerous or are presented
too quickly for chunking to occur, participants typically recall
no more than 3-4 items. This even occurs for items with which
we are very familiar. For example, in a 1975 paper Donald
Broadbent described a study where participants were asked to
recall familiar lists of seven items, such as the Seven Wonders
of the World or the names of the Seven Dwarves from 'Snow
White.' Participants were typically able to recall all of the items;
however, Broadbent noted that long pauses during retrieval of
items tended to occur between the third and fourth items being
recalled; these pauses suggested that additional retrieval of
remaining items had occurred. Here, seven digits could not
be grouped together to form a single chunk. Instead of memory
being capacity-limited to no more than seven items, our imme-
diate memory may only be able to maintain up to four items
before our memory performance declines and retrieving items
takes increasingly greater time.
Early cognitive research that examined capacity limits and
chunking in immediate memory corroborated Broadbent's
view. Many of these early studies involved the presentation
and later recall of phonological or linguistic materials. In
many of these studies, materials that were presented to
478 Chunking
participants in these studies varied in terms of how meaningful
they were. For an example, consider a study by Glanzer and
Razel that examined immediate recall and chunking, where
participants were presented with various types of verbal sti-
muli, including words, sentences, and familiar proverbs. As
one moves from words, to multiword sentences, to proverbs
that can incorporate multiple familiar sentences, it is clear that
the intraitem associations become increasingly stronger as the
presented information becomes more meaningful. It becomes
considerably easier to group words together when they form
a sentence; likewise, it becomes easier to group sentences
together when they form a very familiar passage. Regardless
of how meaningful stimuli were in the context of these various
prior studies, researchers consistently found capacity limits
of approximately 3-4 chunks for each verbal stimulus type.
Herbert Simon, in a similar discussion regarding the role of
meaning in the formation of chunks, obtained similar results.
Though a greater number of words might be recalled from a
familiar nursery rhyme than a novel prose passage, one typi-
cally does not recall more than four chunks across various
linguistic stimuli. This is not to say, however, that increasing
the meaningfulness of verbal stimuli in the manner described
above has no effect upon chunking. As we will discuss in a later
section, the amount of information that is contained within
a single chunk (i.e., chunk size), though not the number of
chunks maintained in memory, may vary as presented stimuli
become increasingly (or decreasingly) meaningful.
More recent research has examined memory capacity limits
when higher-order strategies are precluded. In a 2009 study,
participants were initially presented with various lists of words
that were learned to a 100% criterion. These lists were com-
posed of either word pairs or singletons. In addition to varying
the size of chunks, the number of pairs or singletons within a
list was varied. By varying both of these factors, the influence of
item limits or length limits upon recall could be examined.
Consider an example of three trials in which a participant
might see a list of four singletons, a list of four learned word
pairs, and a list of eight singletons. Assuming that each learned
word pair is grouped into a single two-word chunk, the list of
four word pairs is comparable to the list of four singletons
in terms of the number of chunks that are present within
a list; however, the list of pairs has a longer overall length
(i.e., a larger number of total words) than a list of four single-
tons. In contrast, when overall length is considered, the list of
four word pairs is equivalent to the list of eight singletons;
however, the list of pairs contains a smaller number of chunks.
If a chunk capacity limit is the sole reason why recall is limited,
the prediction is that individuals should recall the same num-
ber of chunks regardless of the list length, and regardless of
whether the chunks are singletons or learned pairs.
Following this training phase, participants were presented
with lists of word pairs that were to be immediately recalled in
correct serial order. To preclude any additional rehearsal of
items, participants were to engage in articulatory suppression
by repeatedly voicing the word the during list presentation.
Ordinarily in serial recall, the list length matters a lot.
Under these suppression conditions, though, when the order
of responses was ignored, a remarkably precise, core capacity
limit of approximately three chunks was consistently observed
across all list types and lengths. For example, in lists with as few
as 4 singletons or as many as 12 singletons, about 3 singletons
were recalled from the list on average; and in lists with 4 or 6
learned pairs, about 3 of the pairs were recalled on average.
This constant number of chunks recalled across many condi-
tions provides strong support for fixed capacity limits in imme-
diate memory. The results of other conditions also suggest,
though, that using strategies like chunking or additional
rehearsal can supplement memory performance a great deal
beyond this core limit.
Chunking Overcomes Capacity Limits
As we have shown above, the number of items that can be
stored in immediate memory is greatly limited; typically, no
more than four items or chunks can be attended simulta-
neously. Because this number is so small, one might expect
that this places a severe limit on the information that we can
attend at any point. This is probably true for many instances
in our everyday lives, as retaining every piece of information
that entered our awareness would cost us the ability to process
information quickly and efficiently. Luckily, our attention is
able to rapidly shift to adapt to new information that enters
our environment, if doing so is needed. However, there are
certainly times when we need to maintain more information
than our limited memory capacity will permit, such as when
one receives a long, verbal list of groceries to buy at the store. In
these cases, chunking is an easy, powerful way to ensure that as
much information as possible is stored. For example, if one is
able to group grocery list items together by their respective
grocery departments (e.g., tomatoes, cucumbers, and grapes
belong in produce; bread, cookies, and apple pie can be
found in the bakery), more information can be stored beyond
the 3-4 items one would recall if no chunking took place.
Various findings from research support the notion that
chunking helps us to exceed severe capacity limits in immediate
memory. The role of chunking is especially exemplified when
linguistic materials are manipulated in some way. Consider, for
example, when sentences vary in terms of their coherence (i.e.,
whether they make sense or are completely nonsensical) . This is
known as order of approximation, which corresponds to how
many words in a row make sense. A sentence that has a low
order of approximation (e.g., first-order) typically makes little
sense, often sounding like a list of random words: telephone call
over easy cheese and then he. Sentences that have higher-orders of
approximation (e.g., fifth-order) sound increasingly like gram-
matically correct text: when we saw the cat attacked by using claws.
If you read both of these sentences out loud and tried to recall
each of them immediately, you probably will find that you
recall a greater number of words from the sentence with a
higher order of approximation. In coherent language, there
are numerous semantic and syntactic associations that exist
between words; as one moves from nonsensical language to a
more coherent form, the various associations between words
increase and a more coherent whole can be apprehended. The
more that words are associated with each other, the easier
chunking words together becomes. As a result, sentences with
a higher order of approximation, like the example above, are
more likely to be encoded as a single chunk. Thus, we see more
words recalled from these sentences than from nonsensical
sentences where interword associations are considerably fewer.
Chunking 479
The notion that increasing associations will lead to a greater
amount of recalled information is also apparent when stan-
dard, grammatical sentences are considered. In cases where the
order of approximation for two sentences is equivalent, more
information will generally be recalled from the sentence that
makes greater use of interword associations. Thus, a greater
number of words will be recalled from a sentence that is a
compound of two smaller sentences (e.g., Grandmother warmed
the oven and she baked a fudge cake ) than would be recalled from
two unrelated sentences.
Dramatic cases of chunking overcoming capacity limits are
found in research that examines effects of deliberate practice
and expert performance. You might be familiar with the digit
span task, an immediate memory test that was mentioned
above. Here, participants are presented with a list of digits
that they are to recall immediately after presentation of the
list, in the presented order. The length of the digit list increases
until a participant makes an error or fails to recall a digit for
several lists in a row. As Miller noted, the average person can
accurately recall between five and nine digits in this task. This is
typically considered a measure of one's basic memory capacity;
however, as we argued above, a small amount of chunking is
still possible in this task.
What would happen to one's digit span if a participant
practiced this task for a few hours each day over the period of
a year? Given findings from a previous study that did just that,
it is very likely that your digit span could be over 80 digits! In
this study, a single participant repeatedly practiced the digit
span task for several months; by the end of the training period,
he was able to report back up to 82 digits in the span task, well
beyond the 5-9 digits that the average person recalls. The
ability to chunk various digits together played a considerable
role in the participant's remarkable span increase. Specifically,
the participant himself noted that he used prior long-term
knowledge of long-distance running times to group digits
together to form multidigit chunks. For example, the digits
4, 5, and 2 could be grouped together to form 4.52, a very
fast time to finish a mile run. This sort of recoding, where
meaningless stimuli are grouped in a way that connects to
one's long-term knowledge, is reported by many expert mne-
monists to be a typical way that they encode stimuli that exceed
capacity limits. The role of chunking should not be down-
played in this case, as it is unlikely that such a high number
of digits could be recalled without some sort of organizational
strategy developing over time.
We should note that chunking only circumvents the capac-
ity limits in memory in the sense that more information can
be recalled. In the grocery example we mentioned above, chunk-
ing allows us to remember a greater number of items on our
shopping list. Although chunking increases the amount of infor-
mation that is able to be stored, the chunks that are formed are
still subject to the four-item capacity limit. A greater amount of
information is being stored within a single chunk (i.e., the size
of the chunk becomes larger) . The number of chunks that can be
stored in immediate memory, however, never exceeds four. This
result has been found even in cases where the number of items
that can be stored or recalled is exceptionally large, including the
participant who recalled 82 random digits in the digit span task.
As we noted above, the amount of digits that were accurately
recalled is far beyond what would normally be recalled in an
Groups of
Figure 1 A hierarchical structure exemplifying the type of performance
observed by Ericsson, et al. (1980) in a participant who learned to use
athletic records and other memorized groups of digits to form multidigit
chunks. The recalled list of digits was chunked into groups of 3-5 items
(seen on the left), which were in turn spatially organized to form even
larger groups, as shown on the right side of the figure. Although the total
digit recall in the figure would greatly exceed what is considered the limit
of immediate memory capacity if each digit were retained as a separate
chunk, the number of larger groups on the right of the figure fall within
this capacity (i.e., 3-5 chunks).
immediate memory task. However, it was found that he had
chunked these digits into a hierarchical structure of four large
supergroups (which were made up of various, smaller groups of
digits), as is shown in Figure 1.
It might be theorized that attention can 'zoom' up and
down to different levels in the structure shown in the figure.
When the mnemonist was thinking at the level of a single chunk
that was just created, the number of digits in the chunk would
be limited to about four. When the mnemonist zooms up to
think on the superchunk level, however, he is limited to about
four chunks per superchunk and at the list level, there should be
only about four superchunks. In this way, the presumption is
that the basic capacity limit does not change in an expert, but
the way in which it can be used improves dramatically with
expertise and practice.
Clearly, chunking does not entirely overcome the capacity
limits inherent in the immediate memory system. Despite this,
we suggest that chunking allows one to remember a greater
amount of information than would be possible if no strategy
were used at all. It is for this reason that chunking is regarded as
such a powerful mnemonic strategy.
How Chunking Works
Chunking can help us increase the amount of information that
can be stored and later recalled. This ability is important, as
it can help us exceed a severely-limited immediate memory
480 Chunking
capacity. If we understand the definition of chunking, as well as
its necessity for the immediate memory system, a question that
follows is how the process of chunking works. Above, we have
suggested that items or chunks in immediate memory are stored
in a capacity-limited region, where no more than four items may
be consciously attended at any given time. Based on what we have
previously learned about this region, we proceed with a discus-
sion of how chunking may work in the context of this theory.
For chunking to occur, some degree of selective attention is
often necessary. This is because items held in immediate mem-
ory receive attention; likewise, forming chunks from these
attended items will also require attention and presence in
conscious awareness. As an example of how this process
might work, consider an instance where a person is provided
with a seven-digit telephone number, 6847083. Assuming that
there are no potential aides to record this number, one will
have to store the series of digits in immediate memory. How
might this proceed? Suppose that the first four digits of the
phone number, 6847, enter into the capacity-limited region of
immediate memory through selective attention. Because four
items are currently being stored within this region, no other
items receive attention or any form of processing. At this point,
if more digits are to be stored, chunking must occur. Specifi-
cally, items within the capacity-limited region can easily be
Figure 2 A model of how chunking may occur in the capacity-limited
region of immediate memory. Digits in a phone number enter the region
and form associations (indicated by dashed lines) to create larger
chunks. Note, however, that the number of chunks that can be stored
never changes. Encoding the phone number occurs in phases, as in this
example: (a) Digits 6, 8, and 4 are first bound into a single chunk;
(b) then 7 and 0 are bound into a chunk, and the second 8 enters the
capacity-limited region; (c) chunks 684 and 70 are bound into a larger
chunk, and 8 is chunked with 3, which has just entered the region; and
(d) The two chunks that comprise the number are chunked, which will
allow the entire phone number to occupy a single chunk and will free up
several other chunks that can be used to accomplish something else
(such as keying in the number or holding a conversation).
chunked or associated with each other, as all of them are
available and accessible in memory. As illustrated in Figure 2,
one could bind the digits 6, 8, and 4 into a single chunk -
doing so reduces the number of items stored in immediate
memory from 4 to 2. This allows room for two more digits
(e.g., 0 and 8) to receive attention, and thus, enter the limited-
capacity region. Further chunking can occur (e.g., 70 are
formed into a single chunk); this again will reduce the number
of chunks occupying the limited-capacity region, allowing the
final digit to receive attention and storage. Through this pro-
cess, the phone number is eventually memorized. The process
of forming chunks in the capacity-limited region can be as
recursive as needed - chunks can be grouped into even larger
chunks, which is certainly beneficial for stimuli that are more
complex or numerous than the example listed here.
In general, research supports the idea that chunking can
occur through at least three possible ways. First, we can chunk
or group items together by forming associations between items
that are located in the capacity-limited region of immediate
memory, in the manner described above. Second, in a similar
vein to the first method, we can take advantage of perceptual
cues present within a given set of stimuli (e.g., pauses in
auditory lists, items in close spatial proximity) to form groups
once items enter the capacity-limited region. Unlike the first
case, where groups can be formed in any sort of fashion, the
cues provide clear chunk boundaries. For example, a pause in
between the third and fourth digit in a phone number (e.g.,
684 7083) allows us to form two chunks - one containing the
first three digits, and another containing the final four digits.
Finally, we can also form an association between items that are
in the capacity-limited region with prior knowledge that is
stored in long-term memory. In general, long-term knowledge
cannot be in conscious awareness unless it has been activated
and brought into the capacity-limited region. Once long-term
knowledge receives selective attention, forming a chunk or an
association between this knowledge and other items in this
region proceeds similarly to the first chunking process we
described above.
How Chunks Are Measured
Above, we discussed the notion of the boundaries of chunks,
places where individuated chunks end and begin. In some
cases, such as when stimuli contain perceptual cues that
encourage chunk formation, the boundaries are relatively
clear; we can tell how many chunks there are (the number of
chunks), as well as how large each chunk is (chunk size). In a
majority of cases, however, stimuli are considerably more
complex (e.g., prose passages), and it may be unclear where
these boundaries occur. This is particularly so when encoding
of information is considered, as internal cognitive processes
are occurring. How can we measure both the number and size
of chunks in these cases?
One of the earliest ways proposed to examine chunk
boundaries was done by measuring adopted chunks. Initially
theorized by Endel Tulving and Jeanette Patkau, an adopted
chunk is defined as recalled items that match the manner
in which the stimuli were actually presented (i.e., order infor-
mation must be preserved in recall). For example, a participant
Chunking 481
is presented with the sentence "The quick brown fox jumped
over the lazy dog. " If the participant recalls "The quick brown
fox lazy dog," two adopted chunks have been recalled. This
is because the participant has recalled two phrases in correct
order, but the whole sentence was not recalled verbatim (if this
were the case, one adopted chunk would have been formed).
Many early studies that measured adopted chunks in recall of
unstructured materials found that across stimulus types that
varied in terms of subjective meaningfulness; in the case of
Tulving and Patkau, participants were presented with sentences
that varied in terms of order of approximation to text, which
was discussed previously. Findings from these various studies
show that the number of adopted chunks that are recalled is
constant, regardless of how meaningful the stimulus is. How-
ever, as we mentioned above, the size of these adopted chunks
increases as stimuli become more meaningful. These studies
concluded that there is a constant limit in terms of the number
of chunks that can be maintained in memory; however, a
greater amount of information (e.g., words within a sentence)
can be grouped into a single chunk as stimuli become more
meaningful and familiar to us.
Another way to measure chunks capitalizes on an impor-
tant assumed properly of chunks: the association between
items contained within a single chunk should be considerably
stronger than the association between items from two different
chunks. Theoretically, this means that if one item that is part of
a given chunk is recalled, other items within that chunk will
likely be recalled as well (i.e., a positive relationship). In con-
trast, there is practically no relationship between recall of items
that come from separate chunks; recall of one does not lead to
recall of the other. In fact, as items from separate chunks are
not bound to each other, we might expect that errors in recall
will increase as one traverses a boundary between chunks. One
way to measure chunk boundaries using these assumptions is
to calculate the transitional error probability of recalled items.
Using a mathematical formula, the probability of erroneously
recalling a presented stimulus is calculated, conditional on the
recall of the item presented immediately prior. In general,
researchers have proposed that items within the same chunk
will have low transitional error probabilities; when a boundary
between two separate chunks occurs, the error probability
should increase considerably.
A third means has similar assumptions to those of transi-
tional error probabilities. As was stated above, items that are
part of the same multi-item chunk are sttongly bound together;
in contrast, items separated by a chunk boundary (i.e., the
items belong to two separate chunks) are not as strongly
bound. It follows, then, that recalling items that are bound
together within the same chunk should be faster than when
items from two different chunks are recalled. This is because,
in the latter case, two separate chunks which have little to no
association to each other must both be retrieved. For example,
Chase and Simon used this logic to measure chunk boundaries
in their series of chess studies that was mentioned previously.
In a chess board reproduction task, chess masters and novice
players viewed a configuration of pieces, and were asked to
copy the configuration onto a blank board; Chase and Simon
were interested in how long it took participants to place suc-
cessive pieces on the reconstruction board. When response
times were examined, both for the immediate memory task
and the reconstruction task, it was found that a chunk bound-
ary was defined at 2 s. If successive pieces were placed on the
board in less than 2 s, those pieces belonged to the same
chunk; if the placing down of two pieces had an interval of
more than 2 s between them, the pieces belonged to different
chunks. In general, it has been found in this and other studies
that training increases the size of chunks; regardless of training
level, the number of chunks that recalled remains the same.
A final method to measure chunks involves a posteriori
assumption regarding how many chunks (and the size of
those chunks) are stored in the capacity-limited region of
immediate memory. Researchers have used two types of mea-
sures to examine chunk boundaries: chunk access and chunk
completion. Chunk access is typically taken as a measure of the
number of independent groups or units that are successfully
retrieved at the time of recall. This is based on the assumption
that items recalled from the same presented group (e.g., word
pair, sentence) are part of the same chunk in immediate mem-
ory. For an example of how this might work, suppose that a
participant were presented with the following word pairs in an
experiment: pop-cube, apple-light, key-fish, and shoe-pill. The
participant recalls the following items: pop, cube, apple, pill.
Here, three separate chunks have been accessed, assuming that
a pair functions as a single chunk - pop-cube, apple-light, and
shoe-pill. The second measure, chunk completion, is taken as a
measure of chunk size. This can be measured as the proportion
of words recalled from the same chunk, provided that it has
been accessed. Completion measures allow us to examine the
coherence of chunks in immediate memory. Using the exam-
ple above, the participant's average completion for the list of
pairs is 0.67 (0.5 for apple-light, 0.5 for shoe-pill, and 1 for
pop-cube). Here, we see that chunking of the pairs is not perfect
(there are instances where the entire pair is not recalled), but
the size of chunks suggests that chunking of items is occurring
somewhat.
Unresolved Issues
Throughout this article, we have described the phenomenon of
chunking to the best of our ability, given what has been found
in previous studies. That said, there is still much to learn about
chunking. Below, we list some questions that, as of now,
remain unresolved; by working to answer these questions,
cognitive researchers will gain a better understanding of how
chunks are formed and processed in immediate memory.
How Many Chunks Can Be Kept in Mind at Once?
Above, we proposed that chunking is necessary to overcome
limitations in the capacity-limited region residing in immedi-
ate memory. In particular, we noted that this region can hold
no more than four items or multi-item chunks at any given
point in time; these items receive selective attention, and are in
our conscious awareness at that time. There are, however, other
views that report slightly different capacity limits. For example,
recent studies by Klaus Oberauer and colleagues suggest that
the capacity-limited region in immediate memory can only
hold a single item at a time. Furthermore, if multiple items
are present in conscious awareness, this is because they are
482 Chunking
bound into a single chunk. Further research that examines
chunk capacity limits is necessary to determine if any resolu-
tion between these conflicting views is possible.
Are Chunks Simple or Complex in Their Structure?
As chunking is often an internally generated process, it can be
difficult for researchers to determine how chunks are organized
in memory. A critical question worth examining in future
research pertains to the organization of chunks in memory:
when chunks are formed, are they separately stored and undif-
ferentiated from each other, or are chunks connected to each
other through some organizational structure? We argue, based
on prior findings, that chunks in immediate memory may
be organized by a hierarchical retrieval structure. For example,
consider the study where a participant spent over a year
practicing the digit span task to the point of accurately recalling
82 digits in the task. Based on retrospective reports, near the
end of the training period, the participant was not only using
his previously-reported strategy of encoding in terms of run-
ning time, but was also using other ways to give digits mean-
ing. Furthermore, he reported deliberately organizing digit
groups spatially, allowing larger groups (termed supergroups)
to be formed; this nested organizational stmcture allowed
digits to function as retrieval cues for remaining digits in a
group or supergroup.
Similar organizational structures are presumed to operate
in the comprehension of discourse or text, where the amount
of information to be stored in memory typically exceeds imme-
diate memory capacity limits. In a text retrieval structure, a
linguistic representation contains several different organiza-
tional levels, allowing surface, semantic, and syntactic features
to coexist in parallel and be chunked together. In general, only
elements that have recently occurred or are central to passage
comprehension are accessible in immediate memory. These
various representations can serve as retrieval cues for what
has previously been read, as recall of what has occurred earlier
in a text passage is necessary for good comprehension.
From these examples, it is clear that chunks in immediate
memory are most likely part of a larger, more complex organi-
zational structure. People often need to know not only which
elements belong together, but also the order in which they
occur (as in a telephone number) or the way in which the
elements are related to one another (as in the understanding
of the slightly different chess configurations that are about
equally good, and those that are not good at all). Flowever,
many studies that have examined organizational structures of
chunks in memory focus upon exceptional domain-specific
performance; even language comprehension can be considered
a type of expertise. It is uncertain how chunks are organized in
cases where expert performance is not required, or where task
performance does not involve long periods of training.
Is Chunking Based on AII-or-None Associative Strength, or
Can Associations Be Continuous?
In our discussion of chunk access and chunk completion
above, we noted in our example that word pairs were not
perfectly chunked. Specifically, in the word pair example that
we discussed above, completion rates were 0.67 overall.
Although items were presented in word pair form, there were
instances where only one word in the pair was recalled. This
potentially suggests that the associative strength between items
within the pair was not sufficient to form two-item chunk.
However, a thorny issue regarding completion rates as a mea-
sure of chunk size involves how we are to interpret associative
strength between items. If two-thirds of the word pairs that are
presented are successfully completed, as they were in our exam-
ple, is it because one is only able to recall two-thirds of all of
the pairs, with all pairs perfectly chunked together (i.e., asso-
ciative strength equaling 1)? Or, alternatively, is it because one
actually had some information about all of the pairs that were
presented, but some pairs were so weakly associated or
chunked that participants were unable to recall them accu-
rately on every trial? Unfortunately, completion is not able to
measure associative strength between items that make up larger
chunks, and it is unclear whether associative strength is dichot-
omous (i.e., items are either chunked or are not) or is subject to
a continuum of potential values. Future measures of chunking
should be sensitive enough to be able to measure the strength
of associations between items.
How Does Chunking Change with Age?
When immediate memory performance across the lifespan is
examined, both children and older adults have worse perfor-
mance than young adults. Two different views have tried to
account for these differences. First, older adults and children
may have a reduced memory capacity, meaning that they hold
a smaller number of chunks in mind. Second, older adults and
children may form smaller chunks. In two recent studies by
Gilchrist and colleagues, participants ranging in age from the
early elementary school years to young adults and older adults
heard lists of unrelated English sentences. Each list of sentences
varied in terms of how many sentences were present, as well as
the length of each sentence. Chunk boundaries were measured
by adapting chunk access and completion measures, discussed
above, to sentence stimuli. The results suggested that both
young children and older adults hold fewer chunks in mind
than did young adults, but that chunk size did not change
across the lifespan. Specifically, for these materials the comple-
tion rate for simple sentences remained at about 0.80 in all age
groups, but the number of sentences accessed was smaller in
children or older adults than in young adults.
Additionally, there is the question of how well young chil-
dren and older adults are able to take advantage of chunking as
an encoding strategy. Although completion rates for simple
sentences were comparable across age groups, the ability to
combine simple clauses to form compound sentences was
not quite as good in the children and older adults. Previous
research suggests that spontaneous use of strategies develops
in later childhood, and then declines with adult aging. More
research is needed to determine how the quality of chunking
changes with age, and why these changes occur.
Concluding Remarks
Chunking is a process that allows us to meaningfully group
items together to increase the amount of information we can
Chunking 483
store and later recall. This process underlies much of how we
learn information. Furthermore, chunking allows us to circum-
vent the capacity limits inherent in the immediate memory
system; though chunks are still subject to capacity limits, we
are able to store more information than if chunking were not
possible, by fitting more information into each of a severely
limited number of chunks that one can hold in mind concur-
rently. In this article, we have discussed chunking in depth,
noting its purpose and a theorized view of how the process of
chunking occurs. Chunking is one of the most powerful stra-
tegies to increase the information we hold in mind, yet there
are many unresolved debates that should be examined in
future research.
See also: Attention; Memory.
Further Reading
Broadbent DE (1975) The magic number seven after fifteen years. In: Kennedy A and
Wilkes A (eds.) Studies in Long-Term Memory, pp. 3-18. Oxford, England:
Wiley.
Chase WG and Simon HA (1973) Perception in chess. Cognitive Psychology
4: 55-81.
Chen Z and Cowan N (2009) Core verbal working memory capacity: The limit in words
retained without covert articulation. The Quarterly Journal of Experimental
Psychology 62: 1420-1429.
Cowan N (2001) The magical number 4 in short-term memory: A reconsideration of
mental storage capacity. The Behavioral and Brain Sciences 24: 87-1 85.
Cowan N (2005) Working Memory Capacity. Hove, East Sussex, UK: Psychology Press.
Ericsson KA, Chase WG, and Faloon S (1980) Acquisition of a memory skill. Science
208: 1181-1182.
Ericsson KA and Kintsch W (1995) Long-term working memory. Psychological Review
102: 211-245.
Gilchrist AL, Cowan N, and Naveh-Benjamin M (2008) Working memory capacity for
spoken sentences decreases with adult aging: Recall of fewer, but not smaller,
chunks. Memory 16: 773-787.
Gilchrist AL, Cowan N, and Naveh-Benjamin M (2009) Investigating the childhood
development of working memory using sentences: New evidence for the growth of
chunk capacity. Journal of Experimental Child Psychology 104: 252-265.
Glanzer M and Razel M (1974) The size of the unit in short-term storage. Journal of
Verbal Learning & Verbal Behavior 1 3: 1 1 4-1 31 .
Gobet F, Lane PCR, Croker S, et al. (2001) Chunking mechanisms in human learning.
Trends in Cognitive Sciences 5: 236-243.
Johnson NF (1978) The memorial structure of organized sequences. Memory &
Cognition 6: 233-239.
Miller GA (1956) The magical number seven, plus or minus two: Some limits on our
capacity for processing information. Psychological Review 63: 81-97.
Oberauer K and Bialkova S (2009) Accessing information in working memory. Can the
focus of attention grasp two elements at the same time? Journal of Experimental
Psychology: General 138: 64-87.
Tulving E and Patkau JE (1962) Concurrent effects of contextual constraint and word
frequency on immediate recall and learning of verbal material. Canadian Journal of
Psychology 16: 83-95.
Classical Conditioning
B 0 Ploog, College of Staten Island, City University of New York, Staten Island, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Acquisition and extinction With a number of pairings of a
conditioned stimulus (CS) and unconditioned stimulus
(US), conditioning will occur (i.e., acquisition of an
association). Conditioning is evidenced if the CS elicits a
conditioned response (CR) that it did not elicit before the
CS-US pairings. (The US elicits an unconditioned response,
UR, regardless of conditioning.) If, after acquisition has
occurred, the CS is then presented without the US, the CS
loses its power to elicit a CR. This phenomenon is referred to
as extinction. Extinction is not simply undoing or forgetting
an acquired association but it is better conceptualized as
learning something new, that is, learning that the CS is not a
predictor of the US anymore as it was before extinction.
Compensatory conditioned response and learned drug
tolerance The compensatory CR is a special form of a CR,
especially in classical conditioning with pharmacological
USs. Its effect is opposite to the effect of the UR. It is thought
that this is a learned response, dependent on the CS's
predictive strength that is activated in anticipation of a
disturbance of an animal's (including humans') homeostasis.
For example, if a CS predicts a US such as morphine, which
lowers the heart and respiration rate, the compensatory CR
elicited by the CS is an increase in heart and respiration rate.
This compensatory CR is thought to be the basis for learned
drug tolerance because the compensatory CR counteracts the
UR, thus making the drug less effective.
Conditioned stimulus (CS), unconditioned stimulus
(US), conditioned response (CR), and unconditioned
response (UR) An unconditioned stimulus (US) elicits a
response without any previous learning. This response ( in the
absence of learning) is known as an unconditioned response
(UR). A conditioned stimulus (CS) elicits a response only if
learning has occurred after the CS has been paired with a US.
Without previous learning, the stimulus to serve as the CS is
neutral and elicits no response. The response elicited by a CS
is known as a conditioned response (CR). Pavlov discovered
that when, for example, a sound (CS) was followed reliably
by food (US), a dog would salivate in response to the sound
(CR) just as it had salivated in response to the food (UR),
even though the sound did not elicit any reaction prior to
conditioning (i.e., CS-US pairing).
Contingeny and contiguity Contingency refers to an if-then
relationship. For example, if the CS occurs, then it will be
followed by the US with a certain probability. Contiguity
refers to the separation between two events. For example, the
US may follow the CS immediately or after a few seconds (high
contiguity) or after a few hours (low contiguity) . In contrast to
Pavlov's original thinking, contiguity is neither necessary nor
sufficient for conditioning to occur. For example, in taste
aversion learning where the US reliably follows the CS but
perhaps only after hours, there is a strong contingency with
low contiguity, yet conditioning is likely to occur.
Excitatory and inhibitory conditioning A CS that predicts
the occurrence of a US is an excitatory CS (CS+). A CS that
predicts the omission of a US is an inhibitory CS (CS-). Note
that the plus or minus designations do not correspond to
'good' or 'bad' outcomes. A CS+ can predict the occurrence
of an aversive or unpleasant US (which would make the CS+
undesirable). A CS- can predict the omission of an aversive
or unpleasant US (which would make the CS- desirable).
Little Albert and John B. Watson's attempt to develop a
behavioral therapy for phobias John B. Watson was driven
by a laudable desire to develop a behavioral therapy to help
people to get rid of their phobias (intense, irrational fear);
he was successful in laying the groundwork for this. In order
to study the elimination of fear scientifically, he sought to
create a fear response in a person such that he would know
its etiology and be able to develop a therapy based on
learning principles. Thus, he exposed Little Albert to a rat
(CS) and then immediately followed this experience by a
loud, fear-inducing noise (US). After a few of such trials,
Little Albert showed a fear response (CR) when just seeing
the rat, similar to the response (UR) he exhibited when
initially hearing the noise without the rat. Watson has been
maligned often because of his work. According to modern
standards, he clearly violated ethical conduct in research,
however, contrary to popular belief, there is no evidence that
Little Albert was permanently or severely traumatized. Many
young children experience aversive events such as loud
noises in their lives, yet they will not be marred for life.
The Rescorla-Wagner model of classical
conditioning The Rescorla-Wagner model is a
mathematical model that attempts to describe phenomena
underlying classical conditioning. Learning is predicted to
be a function of the discrepancy between the maximum
amount that can be learned and the amount that has been
learned thus far. The rate of closing this discrepancy is
determined by the salience ('noticeability') of the CS that is
involved on a given conditioning trial.
Introduction
Classical conditioning is also known as respondent condition-
ing or Pavlovian conditioning. The latter name was given for
obvious reasons, that is, to honor the man who discovered it.
Ivan Petrovich Pavlov. Pavlov was a physiologist, not a psy-
chologist, who discovered this natural phenomenon of reflex-
ive learning serendipitously while investigating the digestive
system. Thus, he became one of the most influential figures in
psychology. Pavlov studied the digestive system using dogs as
484
Classical Conditioning 485
subjects (which also conveniently served as producers of gastric
juices, sold as a digestive aid for people to supplement Pavlov's
underfunded laboratory). The story goes that one day it struck
him that, just before they were fed, his dogs salivated to the
sound of the caretakers' footsteps. Initially, Pavlov labeled this
form of salivation psychic secretion because it was not obvious
how the dogs were capable of anticipating a stimulus (food)
before it actually made contact with the dog's tongue or palate.
Sometimes Pavlov is credited with having invented classical
conditioning, which is simply an incorrect assertion. Classical
conditioning can be assumed to have existed ever since the
evolution of animals capable of learning about their environ-
ment, long before Pavlov. It is also an incorrect assumption that
classical conditioning is a somewhat primitive form of learning.
Classical conditioning operates in a wide variety of situations
(examples below), including complex human behavior such as
emotions, language, and cognition. It is also critically involved
in many other forms of learning and conditioning.
Pavlov's great contribution to psychology was to discover
the existence of classical conditioning, and to develop (and in
this case truly invent) scientific methods amenable to its sys-
tematic investigation. Thus, he introduced experimental con-
trol (i.e., exclusion of confounding variables and ability to
produce classical conditioning on demand), which allowed
him to manipulate factors such as stimulus duration, intensity,
modality, timing, and degree of relationship between two or
more stimuli in order to establish causality (the overarching
goal of all sciences) as related to behavior (the realm of scien-
tific psychology). By doing so, he contributed tremendously to
the application of the natural sciences to psychology - a disci-
pline that was thought of as not being easily amenable to
scientific investigations because of its assumed mental nature.
Pavlov's contributions to the sciences can still be appreciated
when considering that in 2000, Eric Kandel was awarded the
Nobel Prize in Physiology/Medicine for having identified in
the giant sea slug, aplysia, the neural circuitry underlying clas-
sical conditioning, which plays a critical role in the physiologi-
cal mechanisms of memory. (Ironically, Pavlov received the
Nobel Prize in Physiology/Medicine in 1904 for his investiga-
tions of the digestive system, but not for his discovery of
classical conditioning.)
Generally, classical conditioning can explain learning about
situations where events occur in a typical sequence, regardless
of what the specific events are and what a person (or animal)
does. This way, a prior event in the sequence becomes a
predictor for the subsequent event.
Terminology
Classical conditioning can be conceptualized as follows
(Figure 1): on the first trial, a conditioned stimulus (CS; red),
which at this point has no power to elicit a relevant response, is
followed by an unconditioned stimulus (US; blue), which -
without prior learning - has the power to elicit an uncondi-
tioned response (UR; tics on the UR trace). After several such
pairings or trials, the CS acquires the power to elicit a
conditioned response (CR; tics on the CR trace).
In one of Pavlov's first experiments, the sound of a metro-
nome served as the CS, meat powder placed in the dog's mouth
First trial After several trials
Time ►
Figure 1 Schematic representation of classical conditioning involving
conditioned stimulus (CS), unconditioned stimulus (US), conditioned
response (CR), and unconditioned response (UR). Note that CRs only
occur after several conditioning trials.
served as the US, and the salivation response (reflex) consti-
tuted the CR and UR. As can be seen in Figure 1, the metro-
nome (CS) initially does not elicit any salivation (no tics
depicted at the time when the CS is presented first), but the
meat powder (US) does. After several trials (the sufficient num-
ber depends on several factors such as type of CS and US, their
timing, salience, and relationship to each other), CRs (tics on
the CR trace) can be observed when the CS is presented before
the US. The US continues to elicit salivation (UR).
Apparently, the terms conditioned/unconditioned were an
unfortunate inaccuracy when Pavlov's initial work was trans-
lated into English. Perhaps a better choice would have been
the terms conditional and unconditional, because these
indicate the different conditions under which a response
does or does not occur. An unconditional response, for exam-
ple, is a response that occurs unconditionally, that is, regard-
less of what was learned before, and a conditional response
only occurs if certain conditions were previously met. The
controversy regarding terminology, however, is by no means
settled as there is an ongoing debate in the field of animal
learning whether Pavlov, regardless of the issue of proper
translation, indeed referred to conditionality (-al terminology)
or to conditioning (-ed terminology) - the latter reflecting the
notion that through conditioning a stimulus and response are
changed (i.e., conditioned). For consistency in the literature,
most writers nowadays have settled for the -ed, not -al,
terminology.
Forms of Classical Conditioning
Delayed, Trace, Backward, and Temporal Conditioning
Figure 1 depicts what has become known as delayed condi-
tioning. In delayed conditioning, the CS onset precedes the US
onset, but there is no temporal gap between the CS offset and
US onset. The CS and US may even overlap. The stimulus-
stimulus interval, that is, the time between CS onset and US
onset can be relatively short (i.e., seconds, short-delay condi-
tioning) or long (i.e., hours, long-delay conditioning). This
form of conditioning is considered quite effective because, as
will be elaborated below, this arrangement makes the CS a
clear predictor of the US.
Other forms of conditioning include trace conditioning,
which is similar to delayed conditioning in that the CS onset
486 Classical Conditioning
precedes the US onset. However, the CS offset occurs before the
US onset such that there is a temporal gap between CS and US,
when neither stimulus is present. This form of conditioning
has been considered less effective than delayed conditioning,
probably because of the weakened predictive power of the CS
(depending on the temporal gap between CS and US). How-
ever, more recent research has shown that this does not have to
be so. Whether or not one observes conditioning with the trace
procedure depends critically on which CR is selected for obser-
vation. For example, when a pigeon is presented with a keylight
(CS) followed by response-independent food (US) with a
period of a dark key between the CS and US (i.e., trace condi-
tioning), very few if any pecks at the lit key will be observed,
implying that the procedure is ineffective and that little condi-
tioning has taken place. However, if the pigeon's proximity to
food is recorded, the pigeon's position with respect to the
location of food delivery is an orderly function of how much
time has elapsed during the trace. In other words, positioning
becomes a learned response, and with this behavior in mind,
the trace conditioning procedure is very effective in bringing
about conditioning.
A third form of conditioning is backward conditioning.
In this procedure, the US precedes the CS, and therefore the
CS cannot predict the occurrence of the US. Unsurprisingly,
little if any excitatory conditioning of the CS will occur. How-
ever, careful analysis reveals that the CS becomes a predictor
for the US-free period to follow, because the CS explicitly
indicates that there will be no US immediately after the CS.
The CS becomes a so-called inhibitory stimulus (CS-; see
below). As in trace conditioning, judgment of the effectiveness
of the procedure depends critically on what response is being
measured as the CR.
Finally, there is temporal conditioning - a procedure in
which a US is presented repeatedly after a fixed-time interval
has elapsed, without explicit CS. The sensation of passing time
presumably serves as the CS.
Autoshaping: A Special Form of Classical Conditioning?
If a hungry pigeon is exposed to a situation where a keylight is
followed by food regardless of the pigeon's behavior (defining
the autoshaping procedure), the pigeon is likely to peck at the
key after a few trials. Furthermore, if one keylight is followed
by food and another keylight by water, the pigeon pecks at
each of the two keylights reminiscent of the way it would peck
during actual consumption of food or water: for food, one
observes distinct peck movements toward the key, with beak
open, whereas for water, one observes small, rapid beak move-
ments, while the pigeon keeps its head close to the key.
The keylight functions as the CS and food/water as the US;
the peck at the keylight is the CR and actually eating or drink-
ing (consummatory response) is the UR. One human example
may be when people standing in an elevator stare at the floor
indicator. Just as in classical conditioning, this behavior has
no bearing on the outcome (e.g., whether people are staring
or not, the elevator does not move faster or in a different
direction). Autoshaping is of interest because it serves as a
convenient model of classical conditioning with relevance to
nonhumans and humans. For example, it can be shown that
the ratio of the interval between trials (intertrial interval, ITI)
and duration of the CS is critical for learning rates. In general,
under classical conditioning, the larger the ITI:CS ratio, the
faster acquisition will occur.
Human Examples
The Case of Little Albert
One of the most famous case studies in psychology is the case
of Little Albert. John B. Watson had designed a method to
condition a child to be fearful, so he could develop, in a
controlled manner, an effective treatment to subsequently
eliminate fear. To accomplish this, he presented a pet such as
a rat, rabbit, or other animal (CS) to Albert. Immediately after
Albert saw the animal, a metal bar was struck to produce a loud
noise (US). First, Albert did not cry when he saw the pet
animal, but after repeatedly experiencing CSp et -US Ncliss , pair-
ings, he began to cry (CR) as soon as he would see the pet, just
as he would cry (UR) in reaction to the loud noise. The claims
of Watson's putative cruelty and unethical conduct are proba-
bly vastly exaggerated. First, as sketched out in his book, Behav-
iorism, first published in 1924, Watson was driven by a genuine
and admirable desire to develop an effective treatment of irra-
tional fear (phobias) in people, regardless of how misguided
and insensitive his approach may have been. Second, many
babies or young children hear loud noises such as popping
balloons at a children's party or suffer pain from falling, for
example, and then cry because of it. They may even become
afraid of balloons and stairs, which is actually a healthy and
adaptive response. Yet one generally does not worry about
these children being traumatized for the rest of their lives.
Language and Verbal Behavior
One of the most uniquely human behaviors is language, and
classical conditioning (in linguistics sometimes referred to
as statistical learning) is involved in language learning, as
B.F. Skinner already recognized in the 1950s. Classical condi-
tioning is particularly important for children to become effec-
tive listeners. Consider the following situation: A mother might
say "Here's some milk! " or "Have a cookie! " The probability of
occurrence of the actual items, milk and cookie, is much higher
after the vocalizations 'milk' and 'cookie', respectively, than
after the vocalizations of 'juice' and 'sandwich' for example.
Since the sound of 'milk' (CS) often precedes the actual item
milk (US), the CS becomes a predictor for the US. (In collo-
quial terms, the 'meaning' of the utterances 'milk' and 'cookie'
is thus established.) Another example of verbal behavior is
memorizing text (e.g., the word 'pledge' precedes the word
'allegiance').
‘Cognitive Maps' and Memory
The learning of so-called cognitive maps is based in part on
classical conditioning as well. Imagine that each day, on your
way to work riding the bus, you drive by the convenience store
with its bright-green awning immediately before arriving at
the north-east corner of main street where the bus turns right.
To be able to picture your way to work in your mind when, for
example, giving directions to a friend ("Turn right immediately
Classical Conditioning 487
after the convenience store with the green awning!'), you do
not have to be actively driving to work. It is sufficient to be
passively driven and to observe a series of landmarks in order
to learn about specific localities. One landmark becomes the
predictor for the next.
Classical conditioning is critical for some forms of memory
(e.g., implicit memory, which is learning and remembering
past events without explicit awareness). In order for a CS
to remain a CS for a while after conditioning has occurred,
memory - presumably encoded in altered neural connectivity -
is essential. In the example above, learning about landmarks is
only possible if stimulus-stimulus pairs are being remembered.
Embarrassment
Imagine a man finding himself attracted to a woman who rides
the same bus he takes to work every morning. One day, he
musters up all his courage to approach the woman and ask her
if she might like to join him for a cup of coffee. The woman
reacts by saying loudly (so everyone on the bus can hear it)
"You creep! Get lost!' The man blushes and feels embarrassed.
The next day, as soon as the man sees the woman, he blushes
and feels embarrassed all over again. In the framework of
classical conditioning, the sight of the woman serves as the
CS, the woman's initial verbal and embarrassing reaction as
the US. Blushing and the feeling of embarrassment are the CR
(when triggered by the mere sight or thought of the woman) or
UR (when triggered by the woman's original reproach). This
example illustrates that the association was established in one
trial (initial situation) between two stimuli, that is, the sight of
the woman (CS) and the embarrassing words (US).
Acquired Taste Aversion
Joe went to a restaurant where he shared a novelty pizza (duck
meat and feta cheese topping) with his friends. In the middle
of the night, he got violently ill with stomach cramps and
nausea caused by a stomach virus. The next day, after talking
to his friends who did not get sick, he realized that his illness
was not caused by food poisoning. From this day on, even
though he realized that his experience had nothing to do with
the pizza itself, he immediately felt ill just smelling feta cheese
or duck. In the framework of classical conditioning, the smell
and taste of feta cheese and duck represent the CS, the toxin
produced by the virus represents the US, feeling queasy in
response to the smell of feta cheese and duck represents the
CR, and feeling ill in response to the virus's toxin represents
the UR. This example illustrates several additional aspects of
classical conditioning. First, classical conditioning can occur
with just one CS-US pairing. Pavlov's original demonstration
required multiple CS-US pairings. Second, the temporal
separation (referred to as contiguity) between CS and US can
be quite long. Pavlov's original demonstration required rather
close contiguity between metronome and meat powder. (If the
metronome were presented hours before the meat powder,
the dogs would not salivate to the sound of the metronome.)
Third, classical conditioning occurred without awareness and
contrary to what Joe deduced from his talk to his friends. Joe
realized that the nausea after his restaurant visit had nothing to
do with feta cheese and duck meat, yet these specific foods
continue to produce aversive reactions. Fourth, conditioning is
more likely when novel stimuli with no or little prior history
are involved. Joe had never experienced the combination of
feta cheese and duck before. If it had been a standard pizza
with tomato sauce and mozzarella, which he had eaten many
times before in his life, it would have been unlikely that he
develop an aversion to tomatoes and mozzarella cheese. This
point will be elaborated in a subsequent section in the discus-
sion of the so-called Rescorla-Wagner model (RWM) of classi-
cal conditioning, which in this article will serve as an example
of one of several competing theories of classical conditioning.
Fifth, some CS-US associations occur more easily than others,
depending on what species is being considered. This phenom-
enon is referred to as species-specificity and underscores
biological constraints on learning. Specifically, in humans, an
association between a smell/taste CS and a nausea-inducing US
occurs easily, but not an association between a visual CS and a
nausea-inducing US. (The relationship might be opposite for
another species, e.g., quail.) Note that Joe did not develop an
aversion to the sight of the restaurant or of his bedroom even
though these provided stimuli that were more contiguous than
the smell/taste stimuli provided by the food in the restaurant.
This point will be elaborated in the next example.
Fear-Conditioning
Here is another example of fear conditioning (compare with
the case of Little Albert above) that involves pain without an
experimenter's responsibility. Jane went to a restaurant where
she shared the infamous novelty pizza (duck meat and feta
cheese topping) with Joe and her other friends. After they
finished and left the restaurant, she got hit by a car and experi-
enced excruciating pain. After recovery, she got invited by her
friends to go out for dinner again. She still liked novelty pizzas
but to her own surprise, when she approached the restaurant,
she broke out in cold sweat, her heart started racing, and she
felt very fearful about being in another accident. In this exam-
ple, the sight of the restaurant serves as a CS, the pain inflicted
by the accident serves as a US, feeling fearful, breaking out in
cold sweat, and her heart racing in reaction to the sight of
the location where the accident had occurred represents the
CR, and experiencing pain, fear, and dread because of the
injuries she suffered from the accident when it happened rep-
resent the URs. This example, again, illustrates the species-
specificity and biological constraints in classical conditioning
(Jane developed an emotional response to the place where
she ate, not to the food; the US was pain, not nausea).
The mechanisms just described probably represent at least a
partial explanation of the posttraumatic stress syndrome, where
cues that have become associated with stress, pain, and injury
can elicit extreme fear-responses even many years after the
original situation that first established the association between
the cues and the actual event. A second point is of importance:
The CR and UR are not identical. Even though Pavlov hypothe-
sized what he called stimulus substitution (i.e., the CS takes the
place of the US) and thus tried to explain the similarities
between CR and UR (both salivation), he recognized that the
CR and UR were not identical (the saliva differed in chemical
composition). The following example illustrates a situation
where the CR and UR can even be opposite in their effect.
488 Classical Conditioning
Learned Drug Tolerance
Bob and Serena had been using heroin for years, which they
originally used recreationally under very specific circumstances,
with their own routine and ritual. Because they had become
increasingly more concerned about their health, they decided to
quit and clean up. They stayed clean for 18 months, proudly
overcoming the battle with initial physical withdrawal. They
had not even been thinking of their past habit anymore, when
one day they were invited to a party where old friends were
shooting up heroin. As soon as Bob and Serena saw the familiar
paraphernalia of syringe, spoon, lighter, and tourniquet, both
experienced sensations that they immediately recognized as
withdrawal symptoms. Whereas they always felt warm, happy,
and relaxed when using heroin, they now, after months of
abstinence, felt cold, irritated, and tense. This experience caught
them completely by surprise because they had not used heroin
for months and thought they had completely overcome with-
drawal from heroin. An analysis of this situation within the
framework of classical conditioning, somewhat simplified, is
as follows: The paraphernalia function as CSs (cues that predict
the imminent injection of heroin), the injected heroin func-
tions as US, and the sensations and physical reactions resulting
directly from the injected heroin are the URs. (A proper analysis
is somewhat complicated by the fact that physical reactions
themselves provide so-called proprioceptive stimuli.) What
are the CRs? The analysis is in terms of so-called compensatory
CRs. Considering that biological systems typically strive for
homeostasis, CSs are assumed to be cues to allow the body
to prepare and compensate for the impending physiological
disturbances caused by heroin (US). With initial heroin use,
these CRs are assumed to be nonexistent or weak but with
repeated use, the CSs become increasingly strong as predictors
(i.e., classical conditioning has occurred), and accordingly the
elicited CRs become stronger as well. With drug use, the CSs are
always followed by the US, but the URs are counteracted with
increasing force by the CRs. The net effect of the drug thus
becomes diminished. The user counteracts this in turn by increas-
ing the dose of heroin to increase the net drug effect. The sUength
of the CR increases in turn, and thus the cycle continues. So,
when Bob and Serena encountered the familiar CSs (parapher-
nalia) at the party, compensatory CRs were elicited (and experi-
enced as withdrawal) that were not attenuated by the primary
effects of heroin (because they were not using any). These unex-
pected withdrawal symptoms can contribute to relapse. Similar
analyses can be applied to tolerance and withdrawal involving
nicotine, caffeine, and alcohol, but also to less obvious situations
involving the diminishing effects of thrill-seeking behavior (e.g.,
bungee jumping and skydiving), changing romantic attachment
in short- and long-term relationships, and depressive reactions to
retirement after a long work life.
In the example of learned tolerance, the effect of the CR and
UR are in opposite directions. For example, if the UR is a
decrease in heart rate, respiratory rate, pain sensitivity, and
muscle tension, the CR is an increase in these measures. Note
that this form of classical conditioning describes a phenome-
non sometimes called psychological dependence. It occurs
independently of systemic changes in opioid receptors or
drug residuals, as can be logically verified as follows: After
months of abstinence, putative opioid-receptor sensitivity
and drug residuals are identical moments before and after
encountering the CSs, yet withdrawal occurs immediately
upon encountering the CSs (environmental cues). Based on
this analysis, a very effective therapy for drug dependence has
been developed as will be presented in a subsequent section.
The Acquisition of Associative Strength in Classical
Conditioning
Predictiveness: Contingencies and the RWM
The CS can be conceptualized as a signal that allows a person's
(or animal's) physiology to prepare for events that happen
with a certain probability: In the above examples, the metro-
nome served as a cue for Pavlov's dogs' palates to be wet before
the anticipated arrival of dry food; the mother's utterance
'cookie' was a signal that a cookie was on its way; the sight or
thought of the person who inflicted embarrassment was a cue
for more embarrassment; the smell of feta cheese was indica-
tive of food that was likely to cause sickness (regardless of
whether the food was the actual cause or not); the rat was a
warning signal for Little Albert that a loud noise was about to
happen; and the syringe for the heroin user was a cue for
imminent disruption of physiological homeostasis. The critical
point is that in these examples the CS becomes an effective
predictor (more precisely an excitatory CS or CS+) for the
occurrence of the US, if it signals a greater probability of the
US in its presence than in its absence: p(US|CS)> p(US|
no-CS). It is also possible that the CS becomes a predictor for
the absence of the US. For example, a person might walk in a
badly lit alley and worry about getting mugged. But as soon as
the person sees a police station, the person might feel relief.
The police station, CS, functions as a signal for the decreased
likelihood of getting mugged and injured (US) because p(US|
CS) < p(US|no-CS). In this case, the CS is inhibitory or CS-.
Perhaps it is helpful to think about how different types of
CSs (CS+ and CS-)may elicit a range of human emotions with
respect to different types of USs (desirable and aversive). Here
are four examples: a person playing a slot machine who sees
three cherries lined up (CS+ predicting the occurrence of a
desirable US, money) probably feels happiness and excite-
ment. A smoker who sees the 'No Smoking' sign (CS- predict-
ing the omission of a desirable US, nicotine) probably feels
unhappy and disappointed (perhaps even experiences with-
drawal). A pilot seeing the warning light of engine failure
(CS+ predicting the occurrence of an aversive US, danger)
probably experiences fear. And finally, the Homeland Security
Department's Code Green (Low Risk of Attacks; a CS- predict-
ing the absence of the US, danger) probably elicits a feeling of
relief and safety in people. It is important to realize that the
signs in CS+ and CS- are not notations of the desirability of
the CS or US. Instead, the sign signifies the type of contingency,
a positive or negative contingency, which refers to 'if CS, then
US' (positive contingency) and 'if CS, then no US' (negative
contingency) - simply the degree of dependency between the
CS and US. The desirability of the CS depends on the contin-
gency and the type of US. Thus, a CS+ with respect to the
occurrence of danger elicits the unpleasant emotion fear,
whereas a CS- with respect to the omission of danger elicits
the pleasant feeling of relief.
Classical Conditioning 489
Several models of classical conditioning that see the CS as a
signal for presentation or omission of a US have been pro-
posed. One of them, the RWM, will be introduced as the most
representative of these models, even though it has been shown
that it cannot account accurately for all phenomena in classical
conditioning. The RWM is also an example of an early model
that proposed a mathematical equation that successfully pre-
dicted human behavior (even though it has been tested pri-
marily in nonhumans). Here is a simplified version of this
model: AV = s(A - V sum ) where AV is the change in associative
strength of a given CS (i.e., how much more or less condition-
ing has taken place after an additional CS-US pairing), s is the
salience or noticeability of a CS (its value ranging from 0 to
1.0), A is an asymptote denoting the maximum amount of
conditioning that a given US can support (sometimes conve-
niently expressed as percentage; it can be a desirable or unde-
sirable US), and V sllm is the total of associative strengths of all
stimuli that are involved in all previous conditioning trials.
For example, if for a metronome s = 0.50, for a morsel of
food A = 100 drops of saliva, and for all stimuli (metronome
only, for simplicity) V sum = 0 (before the first trial, before any
conditioning has taken place), then AV = 0.50 (100 - 0) = 50
after completion of the first trial. Thus, by completion of
Trial 1, the metronome is predicted to have gained associative
strength and can elicit 50 drops of saliva. After the second
CS-US pairing, AV = 0.50 (100 - 50) = 25; after the third
and fourth pairing, AV = 12.5 and AV = 6.25, respectively. The
cumulative associative strengths of the CS (V sum ) after Trials 1,
2, 3, and 4 are 50, 75, 87.5, and 93.75, respectively. Figure 2
depicts the acquisition curve of a CS over ten trials. AV in Trial
1 is added to V sllm in Trial 1 (amount learned prior to Trial 1)
to produce a new sum (amount learned after Trial 1). This new
7 sum is then inserted into the formula to determine AV for
Trial 2, and so on. (A - V sum ) represents the amount of
associative strength that still can be acquired after a given
Acquisition of associative strength
Figure 2 Hypothetical example of an acquisition curve (in red) as
conceptualized by the Rescorla-Wagner model. Refer to text for
explanation.
trial (i.e., the difference between the maximum of learning
and how much has been learned already). How much of this
difference will actually be added by the next conditioning trial
is proportional to the salience of the CS and the number of
previous CS-US pairings. It should be clear that the equation
describes an increasing, monotonic function of diminishing
returns with an asymptote at A, which is a classic learning curve
similar to the one that Hermann Ebbinghaus described over
100 years ago in his memory research. (Each opportunity to
study a list of nonsense syllables constitutes one trial, and each
trial results in learning additional words, but the length of the
list sets the asymptote because one cannot recall more words
than are on the list.) In the beginning, much can be learned on
one trial; after many trials, very little can be learned on an
additional trial even though performance is ever improving.
The RWM can also account for the phenomenon of extinc-
tion by assuming that A = 0. Extinction refers to a situation
where after several CS-US pairings, the CS is presented by
itself, that is, not being followed by the US anymore. This
results in a decline of associative strength. The resulting curve
looks like a forgetting curve (a horizontal mirror image of
the learning curve in Figure 2.) However, the notion that
extinction is similar to forgetting has not been substantiated.
Extinction is better conceptualized as learning a new contin-
gency involving the same CS. (Evidence against forgetting
as an explanation comes from so-called re-acquisition tests,
for example, in which after complete extinction, the original
CS-US pairings are reintroduced. Reacquisition is typically
faster than the original acquisition, despite the fact that both
start at levels where the CS appears to have zero associative
strength.)
The RWM is a powerful model that has been tested empiri-
cally many times. For example, one can change the value of A
(e.g., increase or decrease US magnitude or introduce extinc-
tion with A = 0) and change the salience (e.g., make the CS, a
light, brighter, or dimmer). Decreasing US magnitude, for
example, results in lowering the asymptote of the obtained
learning curve. Increasing or decreasing the brightness of the
CS results in a higher or lower rate of acquisition, respectively.
Generally, these predictions have been validated empirically.
Contiguity
Contiguity in classical conditioning refers to the closeness of
two stimuli (mostly temporal closeness of two CSs, or between
a CS and US). This was one of Greek philosopher Aristotle's
(384-322 BC) Principles of Association. For example, if one
happens to observe a shooting star (CS) and then immediately
thereafter hears a loud bang (US), many would be tempted to
see a causal relationship between these two events (even if
there is none). Pavlov also recognized this principle and empir-
ically tested it by varying the time between the metronome
(CS) and food (US). The more closely together the CS and
US occurred (however, not simultaneously), the more quickly
conditioning took place. When the temporal separation was
too large, no conditioning occurred. (Even if there was a con-
tingency between metronome and food, that is, US-food
always followed CS-metronome, if there was only negligible
contiguity - for example, presentation of the metronome was
followed by food 8 h later - no conditioning would occur.)
490 Classical Conditioning
Thus, it came as a surprise when acquired taste aversion was
discovered (see example above). Here, a long gap (e.g., hours)
between CS-taste and US-sickness was possible and still
allowed for fast conditioning (sometimes even in just one
trial). Thus, contiguity is not necessary for conditioning to
occur. In fact, even in situations with contiguity but without
contingency (i.e., the CS is not predictive; the US is equally
likely to occur with or without being preceded by the CS),
no conditioning occurs. (Robert A. Rescorla introduced this
no-contingency test as the truely random control group.) Thus,
contiguity is not sufficient for conditioning to occur.
Stimulus Overshadowing and Stimulus Blocking
The failure of conditioning despite contiguity is further illu-
strated by two phenomena, stimulus overshadowing and stim-
ulus blocking. Stimulus overshadowing might occur when two
CSs - one salient, the other not - are competing for associative
strength. The outcome is that the more salient CS accrues most
of the available associative strength whereas the less salient CS,
despite equal contiguity, fails to accrue much associative
strength. A human example is as follows: most elevators are
equipped with a visual signal (up or down arrow) and an
auditory signal (one or two chimes) for the vision impaired
to be informed about the direction of the elevator. In general,
for people visual stimuli are more salient than auditory stimuli.
Therefore, we learn more about the direction of the arrows
(CSs) as predictors for the actual movement of the elevator
(US, to move in the intended direction) than about the signal
value of one versus two chimes. When asked, most people
cannot tell whether one chime indicates up or down move-
ment of the elevator.
Stimulus blocking refers to a situation where one of two (or
more) CSs has a prior predictive (or associative) history with
respect to the US, and as a result blocks a second predictive CS
that is added later. Thus, despite the fact that the second CS is
contiguous with the US, its acquisition of associative strength is
drastically reduced. (According to RWM, A - V sllm , the poten-
tial of subsequent learning, is small if one CS had already
accrued most of the associative strength. Thus, little associative
strength is left to be accrued by the added CS.) A real-life
human example of blocking involving classical conditioning
alone is difficult to identify. However, it can be argued that
a discriminative stimulus, S D , in operant conditioning ac-
quires its value or cue function through classical conditioning
(i.e., given an appropriate response, the S D is followed by a
reinforcing stimulus, S R ). This S D - S R pairing is one of the
three binary associations that has been experimentally identi-
fied in the so-called three-term contingency of operant condi-
tioning, S D : :R - S R (if S D , and if R, then S R ), and, for the current
analysis, is equivalent to the CS-US association in classical
conditioning. The blocking rationale may then apply to a
variety of human situations.
What follows is an experimental demonstration of blocking in
humans (even though similar paradigms have been used to study
blocking in other species): initially, a light (CS L ight) is presented
and followed by a puff of air (US) aimed at one of the partici-
pant's eyes. Conditioning has occurred when CS Lig h t elicits an eye-
blink response (CR), similar to the eye-blink response (UR)
elicited by the US. Then a tone (CSr on e) is added to CS Lig h t ,
forming a CSyght+CSrone compound. Blocking is demonstrated
when even after many presentations of the CSught+CSrone com-
pound CSrone alone fails to elicit a CR. It is said that CS Lig ht
blocked CS-pone in its acquisition of associative strength.
Novelty
One last point in this section deals with novelty of the stimuli
involved. To follow through with the example of learned taste
aversion, as mentioned above, consider that a person may have
eaten regular tomato and mozzarella cheese pizza many times
before. In effect, there were many CSraste-USweiiness pairings in
the past. If it so happens that one day the person gets sick after
having eaten a tomato-mozzarella pizza, not much learned
taste aversion with respect to the taste of tomato and mozzarella
cheese would occur because the CS Tast e would have accrued a
lot of associative strength already. In other words, there were
many CS-US pairings in the past that established the food as
safe, so that one bad experience will not have a large impact on
the firmly established CS. However, if the person gets sick the
first time a novel CS Tast e is experienced, the one CS-US pairing
will have a proportionally great impact. Recall that according to
the RWM, early training trials with a new CS have a relatively
large effect: much can be learned on the first few trials, whereas
relatively little new information is learned on late trials.
Therapies Based on Classical Conditioning
The final section of this entry addresses the role of classical
conditioning in a number of therapies for clinical disorders.
Four examples will be offered as a representative selection of
treatments based on classical conditioning.
Treatment of Phobias
A phobia is an irrational fear that is intense enough to render a
person dysfunctional and distressed. As discussed above, the
case of Little Albert was John Watson's attempt to develop
a model of fear (with known history) in order to develop a
treatment to take fear away from a person. His student Mary
Cover Jones continued his work and developed one of the
earliest behavioral therapies based on what she called uncon-
ditioning. The approach was to introduce the established fear
stimulus (white rat or rabbit) gradually, while keeping the
child relaxed and distracted by allowing him to eat a favorite
food (and the loud noise was omitted). Years later, in 1958,
Joseph Wolpe used this as the basis for his so-called sys-
tematic desensitization through reciprocal inhibition or coun-
terconditioning. Research over the following years revealed
that the clinically relevant component of unconditioning, sys-
tematic desensitization, or reciprocal inhibition was extinction
(though changing the value of the CS from being a predictor
of something unpleasant to a predictor of something pleasant
may contribute to a successful treatment as well). In this and
other treatments such as flooding and implosive therapy,
based on extinction, the client is exposed to the fear CS without
being exposed to the fear US (and without being allowed to
avoid the CS by withdrawal, for example). As a result, the CS
loses its fear-eliciting properties.
Classical Conditioning 491
Treatment of Drug Tolerance
In the previous example of learned drug tolerance, the para-
phernalia elicited withdrawal because of their association with
heroin use. The experience of withdrawal is thought of being
one of the reasons why former addicts relapse, even after
months or years of abstinence because the former addict is
very likely to reduce the unpleasant withdrawal by giving in
and using the drug again. Thus, one aspect of treatment of drug
dependence should focus on eliminating the experience of this
type of withdrawal. In order to do so, the former addict is
intentionally exposed to the drug-cues (CSs) but is not allowed
to actually administer the actual drug (US). For example, an
addict may be allowed to cook up saline solution (an inert
substance) and inject it but is not allowed to do so with actual
heroin. Initially, the addict will go through severe withdrawal
all over again. Each presentation of the drug-cues without
actually administering the drug, however, serves as an extinc-
tion trial. With extinction, once again, the association of the
drug-cue-CS with the actual drug-US is changed such that the
CSs lose their power to elicit the compensatory CRs, which
decreases the experience of withdrawal.
Treatment of Alcoholism
Even though a similar extinction technique, as described
above, can also be applied to alcoholism or any other form
of dependence, so-called aversion therapy, based on classical
conditioning, has also been used for the treatment of alcohol-
ism. The idea is based on the principle of taste aversion
learning. The alcoholic ingests a drug (e.g., Antabuse), which
has no effect until alcohol is ingested. If alcohol is ingested
while being treated with Antabuse, the person gets violently ill.
The idea is that the taste of the alcoholic beverage serves as the
CS, and that the nausea induced by antabuse+ alcohol serves as
the US. After a few CS-US pairings, the taste and smell of
alcohol becomes aversive. This therapy by itself has not proved
to be very successful for reasons that are beyond the scope of
this article.
Treatment of Nocturnal Enuresis
Nocturnal enuresis refers to bedwetting in children who are
considered to be too old developmentally for bedwetting. One
treatment approach is based on classical conditioning, which
is applied after organic or medical conditions have been
excluded as a possible cause for enuresis. For this, a pad (con-
structed like a sandwich of two sheets of mesh wire, separated
by a cotton sheet) is placed under the sleeping child. When the
child urinates on the pad (and, urine being a good conductor,
thus closing the electrical circuit), a battery-operated alarm
goes off and wakes up the child. The sensation of a full bladder
is thought to be the CS, the alarm the US, waking up to the
alarm the UR, and waking up to the sensation of a full bladder
the CR. (When the CR occurs, enuresis is eliminated.) Note
that the sensation of a full bladder always precedes the sound
of the alarm. Thus, the sensation of a full bladder becomes an
effective CS eliciting a waking up response. Note that no
awareness of learning is required for classical conditioning to
be effective (i.e., the conditioning can take place even though
the child is sleeping). Note also that there is an alternative
explanation in terms of operant conditioning, which is beyond
the scope of this article.
Conclusion
The present article on classical conditioning was an attempt
to give a flavor of the richness and complexity of this form of
associative learning. There is a vast literature on the different
models of classical conditioning; for the sake of brevity, only
the RWM was introduced as one example of the many quite
successful models. Other aspects of classical conditioning were
given only a cursory treatment: the application of classical
conditioning to neural networking and cognitive models, the
complexity of recent findings relevant for extinction, and an
appreciation of the fact that in most conditioning situations,
many stimuli are involved, which compete and interact in their
involvement with associations that are being formed. The
interested reader is referred to the list of suggested readings.
See also: Alcohol: Psychosocial Effects; Drugs, the Brain, and
Behavior; Operant Conditioning.
Further Reading
Beck HP, Levinson S, and Irons G (2009) Finding Little Albert. American Psychologist
64: 605-614.
Boakes R (1 984) From Darwin to Behaviorism: Psychology and the Mind of Animals.
Cambridge: Cambridge University Press.
Domjan M (2003) The Principles of Learning and Behavior, 5th edn. Belmont, CA:
Thomson-Wadsworth.
Fantino E and Logan CA (1979) The Experimental Analysis of Behavior: A Biological
Perspective. San Francisco, CA: W.H. Freeman.
Garcia J, Ervin FR, and Koelling RA (1966) Learning with prolonged delay of
reinforcement. Psychonomic Science 5: 121-122.
Garcia J and Koelling RA (1966) Relation of cue to consequence in avoidance learning.
Psychonomic Science 4: 123-124.
Kandel ER and Schwartz JH (1982) Molecular biology of learning: Modulation of
transmitter release. Science 218: 433-443.
Pavlov IP (1927) Conditioned Reflexes: An Investigation of the Physiological Activity of
the Cerebral Cortex. New York: Dover.
Ploog BO (2001) Net amount of food affects autoshaped response rate, response
latency, and gape amplitude in pigeons. Learning & Motivation 32: 383-400.
Ploog BO and Zeigler HP (1996) Effects of food-pellet size on rate, latency, and
topography of autoshaped key pecks and gapes in pigeons. Journal of the
Experimental Analysis of Behavior 65: 21-35.
Siegel S (2005) Drug tolerance, drug addiction, and drug anticipation. Current
Directions in Psychological Science 14: 296-300.
Siegel S, Baptista MAS, Kim JA, McDonald RV, and Weise-Kelly LA (2000) Pavlovian
psychopharmacology: The associative basis of tolerance. Experimental and Clinical
Therapeutics 8: 276-293.
Watson JB (1924) Behaviorism. New York: W.W. Norton.
The Clinical and Cognitive Psychology of Conflict
A Parker, Gothenburg University, Gothenburg, Sweden
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cognitive factors Higher-order psychological factors
relating to intellect, reasoning, belief, and construct systems
which may have an influence on lower-order events such as
learning, perception, and conflict resolution.
Cognitive unconscious Information presented at a speed
too rapid for conscious recognition nevertheless establishes
some kind of memory trace. This information is
subsequently learned more quickly than nonpresented
information. The designation 'cognitive' is added to
emphasize that the effect is in the area of learning and
perception and is more limited in its motivational or an
emotional influence than alleged to occur with the Freudian
or dynamic unconscious.
Conflict A perceptual state involving the executive function
of the organism where the immediate choices in the
organism's repertoire, together with the outcome of these
choices, are seen to involve incompatible motives and needs.
Constructs A term promoted by the American psychologist
George Kelly to describe the belief systems which individuals
use to divide persons (and the world in general) into specific
categories such as pleasant-unpleasant, attractive-repulsive,
and honest-dishonest. Constructs are usually bipolar and
characteristic of the individual.
Coping skills The skills, including those known as defense
mechanisms, for dealing with the demands (especially those
that are psychologically threatening) of the environment.
Defense mechanism The description of the various ways in
which consciousness is said to defend itself against potential
anxiety by distorting the perception of the situation to
become a less or nonthreatening one. The usual defense
mechanisms are denial, projection, displacement,
sublimation, and rationalization.
Emotional unconscious The evoking of activity in the
deeper structures of the brain (the limbic system) which
border on the cortex of the brain and communicate with
bodily organs. While these brain structures mediate
potential emotional experience, the experiences themselves
may take the form of gut feelings or more generally,
so-called somatic markers. They may unconsciously
influence decision making and may be a part of what is
experienced as intuition.
Gestalt psychology Literally, the study of how the whole is
greater than the sum of its parts. An approach within
psychology, which emphasizes a basic principle in facilitating
learning, perception, and even psychotherapeutic change, is to
create a whole and meaningful image from its components.
Conflictual aspects are resolved by combining their inherent
polarizations or promoting a dialog between them.
Mindfulness The focusing of attention on the flow of
immediate experience in the here and now alongwith
enabling the attitude of openness, acceptance, and curiosity
as to the content of experience. It derives from the practice of
Zen meditation.
Mindfulness-based cognitive therapy This is aimed at
breaking habitual thought and reaction patterns and
expanding awareness to include feelings and bodily
reactions that would not otherwise be attended to: in short,
access to otherwise nonconscious processes.
Positive psychology A movement rather than a school,
which derives from humanistic psychology, which aims to be
a counterbalance to psychiatry by providing psychology with
a positive and scientific base in cognitive psychology. It
places an emphasis on the expression of higher motives and
positive emotions involved in the search for meaningfulness,
wisdom, and happiness in life. It is associated with the works
and concepts of Philip Zimbardo, Mihaly Csikszentmihalyi,
Martin Seligman, and Albert Bandura.
Psychoneuroimmunology The study of the way
psychological factors interact via neural and chemical
messengers (hormones) with the body's immune system,
thereby providing a pathway for psychological factors to
have determining effects on illness.
Social identity theory Social identity is composed of several
different identities corresponding to the social groups of
which we are a member. The more positive the images of
these groups, the more positive is our self-esteem, and we
strive to increase the status of our group membership by
creating a series of ingroups and outgroups. This becomes a
basis for social prejudice and group conflict.
State-specific memory This concerns memory that is
specific to a particular state of consciousness and is not
retrievable or only partially retrievable in other states. For
example, events which a person recalls in a drug-induced
state may not be recalled in the normal waking state and vice
versa.
Stroop effect A technique for studying the effect of perceptual
bias on recognition time and learning by using words for
colors written in colors which are incongruous with their
meaning: the word red, for instance, written in blue.
Conflict is an integral part of all human behavior, and its
attempted resolution can be seen as a major part of the effort
toward mastery of the environment. Unresolved and chronic
conflicts are probably the most central problems that clinical
psychology deals with.
A conflict can be defined as a motivational and emotional
state that challenges the cognitive resources of the organism
and which can result in mental and bodily stress. The motiva-
tional aspect is due to the perceived nature of our goals to
which we attribute polarized negative and positive loadings.
492
The Clinical and Cognitive Psychology of Conflict 493
This often means that there is an increase in emotions such as
anxiety associated with nearing the goals and their associated
loadings or with the threat of loss of the reward associated with
goal. Conflicts, when they are experienced consciously, also
involve a cognitive and executive aspect in terms of applying
experience and eliciting skills which can lead to their resolution.
Knowledge concerning conflicts and their effects can ac-
cordingly be organized under the aforementioned headings
of motivational aspects and emotional aspects relating to
their unconscious and dynamic significance, and the cognitive
and executive aspects concerning how they are consciously
conceptualized and resolved. Since conflicts inevitably pro-
duce bodily stress, this is also an important component.
Axiomatic to the understanding of conflict behavior is the
view first put forward by the philosopher of science John
Dewey: "the brain is primarily the organ of a certain type of
behavior, not an organ of knowing the world." This is to
acknowledge that with respect to evolutionary demands, the
human psyche would seem to be best equipped for handling
straight decision making in the face of practical conflicts and
then carrying out these choices. The brain may not be designed
to give any simple resolution of abstract philosophical and
ontological problems such as the meaning of life. This apparent
truism may explain not only why existential conflicts are the
most intractable but also why behavioral conflicts appear to be
the most readily amenable to psychological intervention.
In this vein, it is true that early psychologists chose to study
simple paradigms of conflict situations, often attempting to
determine the variables governing their resolution or persistence
in the form of symptoms. The approach known as behavioral
psychology emerged from this with its foundations in the study
of how rat behavior is determined by simple laws relating to the
contingency and frequency of noxious and rewarding reinfor-
cers. In contrast, the early psychoanalytic approach focused on
the dynamic forces and defenses governing emotional conflict.
Historically (with the exception of Gestalt psychology and a few
partisan approaches such as Jungian psychology), it was much
later that therapies and theories developed to take into account
the unique nature of the human species: man's ability to concep-
tualize the environment and enter into existential conflicts.
Recent developments are, however, giving this aspect a central
role in illness. Unresolved psychological conflict with its result-
ing long-term stress appears to be an important and sometimes a
crucial determining factor not only in psychological disturbance
but also in the genesis of a wide range of illnesses. However,
rather than continuing to focus on diagnosis and disturbance,
there is a contemporary focus, known as positive psychology, on
the enormous human potential for change and on the impor-
tance of existential questions and conflicts.
The Motivational Aspects of Conflicts
It was because of its simplicity that the behavioral approach
made rapid progress in the study of learning; whenever it was
applied to the study of conflict, it had to adapt to a more holistic
approach, that of Gestalt psychology. One of the most influen-
tial gestalt psychologists, Edward C. Tolman, introduced con-
cepts such as expectancies, signs, and cognitions as so-called
'intervening variables' placed in the middle of the simple
stimulus-response analysis of behavioral psychology. For
Tolman, even rats were cognizant of the salient features of their
environment. Another gestalt psychologist, Kurt Lewin, pro-
posed what is now regarded as the classic analysis of conflict in
terms of conflicting goals. This was of the trilogy of approach-
approach, approach-avoidance, and avoidance-avoidance goal
conflict.
In approach-approach conflict, both goals are accredited
equal positive values and a choice naturally entails loss of
one. Analogies are easy to find in the human situation, for
example, when one is faced with a choice between equally
attractive job offers or equally attractive roles, such as in career
fulfillment and in parenthood.
Approach-avoidance conflict involves by definition an aspect
of punishment or loss in achieving the desired goal and evokes
fear or anxiety. Challenges by their nature often entail fears and
contemporary fears may not be only of physical harm but of
psychological annihilation or humiliation, for instance as
often experienced by the challenges and risks involved in the
public performance of skills. Many life crises, from the first step
the child makes to his or her first date, can be seen as conflicts
of overcoming anxiety. Their attempted resolution creates the
learning experience for future skills and confidence in dealing
with the demands of the environment.
Avoidance-avoidance conflict is often regarded as the most
deleterious of conflict paradigms, since behaviorally, the
organism cannot avoid punishment. Some conflicts can be of
relatively minor significance: hard work or the prospect of
exam failure. Other conflicts of this type can, however, have
serious effects: a monotonous yet stressful work situation
against the demeaning prospect of being out of work. (The
situation is not helped by accumulating evidence of a rising
incidence of myocardial disorders with this type of job.) The
conflict can be one between physical harm and psychological
damage. Consider the anxiety evoked by and the consequences
involved in the choice between risking death and physical
harm in war or alternatively refusing to take arms in war and
undergoing group rejection and imprisonment.
Nonresolution of all types of conflict in terms of an outcome
in making behavioral choice (or at least some form of behav-
ioral action) necessarily implies stress. Stress, independent of
whether or not it is of a chronic nature, can produce a variety
of behavioral and, as will be discussed later, physiological
symptoms.
Whereas the aforementioned conflict analysis can be
regarded as a contribution from Gestalt psychology, the behav-
ioral approach using animals resulted in some exact lawful
relationships. For instance, in observations of animal behav-
ior, whether in the laboratory or in a natural setting, one may
speak of conflict between two incompatible types of behavior
if tendencies for these two behaviors are simultaneously
present.
An approach-approach conflict involves two tendencies in
conflict approaching two different objects some distance apart,
which implies that the incompatibility is a physical one. Here,
the animal may reach a point in between where there exists a
balance between the tendencies to be drawn to each object.
The position of this point will be unstable as the tendency to
approach either goal increases with its proximity. Any departure
from the point of balance toward one goal increases the
494 The Clinical and Cognitive Psychology of Conflict
tendency to approach the other goal with a concomitant
decreased tendency to approach the other.
In avoidance-avoidance conflict, an animal is confronted
by two objects it attempts to avoid. The tendency to avoid
either goal is likely to increase with its proximity. Therefore,
any movement toward either object is then likely to result in a
return to the point of balance.
While behavioral psychologists initially studied these
situations in terms of animal behavior, they eagerly sought
parallels in humans. The Russian psychologist I.P. Pavlov
described the behavior of his dogs when they were unable
to discriminate between competing but similar stimuli as a
basis for choosing one of them. He coined the term experi-
mental neurosis to describe how they yelped, bit aggressively,
or lay down and gave up the task. Many years later, the
American psychologist, Martin Seligman, focused on behavior
in the avoidance-avoidance paradigm and described a condi-
tion he termed learned helplessness. Seligman's dogs, on suc-
cessively receiving electric shocks at whatever end of the cage
they chose to move to, finally gave up in their efforts to avoid
the shocks and became passive and behaved as if they were
resigned to their fate. This passivity continued even when the
experimental paradigm changed and the active behavior of
the animal would have removed it from the noxious stimu-
lus. Theoretically, the learned helplessness condition assumes
altered responding in at least three major categories of behav-
ior: emotional, motivational, and cognitive. Seligman and his
coworkers assume parallels here with the kind of reactive
depression in humans that results from the carminative
effects of negative life events. By this account, one develops
the strategy of giving up in the face of apparently irresolvable
conflict, becomes depressed, and loses the belief that own
actions can change outcomes. This simple observation has
become the cornerstone of a group of therapeutic approaches
known as cognitive therapy. The underlying principle here is
to teach the human patient alternative ways of viewing his
environment, create new choice situations, and thereby
encourage certain activities that may lead to positive rein-
forcement, either in purely behavioral or in conceptual (cog-
nitive) terms.
Another motivational behavior concerns fear and anxiety as
a learned drive or secondary drive in which the first stage is
shock fear and the second concerns the signs of fear. Fear or
avoidance behavior becomes an acquired drive in its own right.
In a classic experiment by Dollard and Miller, rats learned to
work hard at pedal pressing for the 'reward' of being released
from a cage they associated with electric shocks. Dollard and
Miller regarded the dependency effects of drugs such as barbi-
turates as due to their reinforcement of drug using behavior via
fear reduction. The behavioral psychologist, Hans Eysenck,
emphasized that the persistence of neurotic behavior is due
to avoidance of the feared object - by always running away at
the first possible opportunity. For Eysenck, it is the first physi-
ological signs of anxiety - increased pulsation of the heart or
increased rate of breathing - that induce this behavior rather
than fear as a learned drive.
'Conscience' for behavioral psychologists can be, to some
extent, explained in terms of conditioning which occurred
according to the reinforcement and contingency laws of
learning or else through imitation as a basic form of social
learning. In these terms, neurosis is seen then as an internal
conflict between conscience and drives.
Motivation was until recent years studied either as a largely
conscious process in terms of behavioral psychology or as
largely unconscious process in terms of the psychodynamic
psychology of Freud and his followers. However in recent
years, there has been some rapprochement between these pre-
viously opposed views. Nowadays, cognitive behavioral psy-
chology views defense mechanism as forms of copying (or
noncopying) skills. Even the concept of nonintentional moti-
vated behavior is used to describe motives that influence behav-
ior but which cannot easily be consciously expressed or
integrated into the self-concept or role. For instance, a person
may deny what is seen by others as obvious flirtatious behavior
by maintaining it was only innocent friendliness because this is
not part of the person's self-concept.
The Unconscious and Emotional Aspects of Conflicts
Conflicts involve emotion, especially anxiety, which is disturb-
ing to the psychological function and efficacy. There is little
doubt that the major contribution of the psychoanalysis, and
in particular that of Freud, is in having provided a rich catalog
of the means of dealing with anxiety from conflicts between
drives and the demands of reality and social learning (the
superego). These are the classical defense mechanisms originated
by the ego as strategies to defend it against the ensuing anxiety.
Their common feature is that they result in an initial ameliora-
tion of the conflict by distorting the perception of reality to fit
in with the needs of the individual. However, some defense
mechanisms such as intellectualization can also be seen as
adaptive coping strategies.
As such, defense mechanisms may be regarded as the stra-
tegies an individual learns to enable personality to function in
the face of conflict from the environment. Regression, for
instance, often operates in situations where efforts at problem
solving or conflict resolution, based on current learning, are
blocked and frustration results. The individual then regresses
to a behavior repertoire, which may have earlier given success-
ful solutions.
In the case of other defense mechanisms such as the denial
of the conflict or its projection onto others, the price for their
operation is a distortion of reality, and this is often in the long
run rather maladaptive. An accurate perception of reality - and
one which includes the individual's own motives and behav-
ior - is important from the point of view of successful conflict
resolution and ultimately for the harmonious functioning of
personality.
Nevertheless, the primary function of defense mechanisms,
as the term implies, is to protect the executive function of
personality from threat and breakdown. Experimentally, there
are techniques which demonstrate how defenses operate.
Many years ago, the term perceptual defense was coined to
describe the higher recognition threshold required for the
identification of emotionally loaded words that were rapidly
flashed (tachistoscopically presented) on a screen.
To demonstrate this convincingly, one needs techniques
that circumvent the problem of response bias. It might be
that individuals recognize the emotionally taboo words but
The Clinical and Cognitive Psychology of Conflict 495
refuse to say them or even admit that they see them. One of
the techniques that can demonstrate that genuine subliminal
conflicts do occur makes use of the Stroop effect. In the Stroop
effect, subjects are asked to read the names of various colors
that are printed in colors that conflict with the meaning of the
word, for instance, the word RED is printed in green. Nor-
mally, this interferes with the naming of the colors by delaying
the naming response. The Stroop effect can be shown to oper-
ate even subliminally, that is below the threshold for conscious
recognition. Subjects report they are unaware of when a clini-
cally loaded word is being presented, but its interference on
their performance can nevertheless be demonstrated if they
were required to say the meaning of color words. Such experi-
ments have led many authorities to reject the notion that
perception is unitary and accept that degrees of perceptual
processing or at least perceptual registration of events can
occur without full conscious awareness.
While some aspects of this approach may be intuitively
appealing, the notion of the dynamic manipulation of the
contents of consciousness by an all-powerful Unconscious
lurking beneath its surface has never been accepted by main-
stream psychology. Nevertheless, a concept known as the
cognitive unconscious has come into vogue on the basis of exper-
imental findings from the study of memory and perception.
This notion describes how perceptions and learning can occur
without the individual being cognizant of the information
involved but nevertheless showing evidence of its acquisition
in his or her behavior. There appears to be some form of
unconscious monitoring and correctional activity present in
routine decision making. The originator of the term the cogni-
tive unconscious, John Kihlstrom, is a coauthor of a recent
paper where he has expressed his contemporary view that
"the unconscious is not just an oxymoron" but that noncon-
scious vigilance occurs for nonconsciously occurring biases
and automatic compensatory processes can be unconsciously
activated. It is well known that the skepticism toward this area
owes its origin to the ill-founded claims for the subliminal
effects on coca cola selling. Nevertheless, a current review of
the literature by Ran Hassin, James Uleman, and John Bargh
concludes that the outright dismissal of the claims was
unwarranted.
There appears now to be a consensus that information can
be cognitively registered in this way as subliminal perception. It is
often said that there is no evidence for subliminal perception
effecting behavior or decision-making processes. Nevertheless,
the rapid exposure (at nonrecognition threshold) of emotion-
ally loaded words such as those relating to hostility, rudeness,
or politeness and also of concepts relating to behavioral stereo-
types such as 'professor' and 'secretary' or 'soccer hooligan' has
been shown to have an effect on later behavior. When tested in
staged ambiguous situations, it was found that, depending on
the nature of the exposure, participants behave more politely
or rudely, intelligently or aggressively, etc. than control groups.
Another contemporary reconceptualization of unconscious
processes is found in the work of Antoni Damascio. Damascio
downplays the role of conscious processes in decision making
and maintains that emotions often steer what we believe are
our logical decisions. An important concept here concerns
somatic markers, which means that emotions arise through the
influence of bodily processes and the limbic system or deep
structures in the brain and exert subtle effects on us without
our knowledge.
Nowadays, the term dynamic unconscious is restricted to the
older Freudian use, while the terms cognitive and the emotional
unconscious are generally reserved for the contemporary use of
the unconscious. Some authorities prefer the more limited
term learning without awareness. Damascio's concept, often
termed the emotional unconscious, is seen, at least in a limited
sense, to be a neurobiological concession to Freud.
The Bodily Effects of Stressful Conflicts
As has been described earlier, defenses are not just a liability
but can also serve, at least in a limited way, a positive function
in enabling the individual to cope with conflicts that would
otherwise be overwhelming. It is then not surprising that sev-
eral styles of coping have been identified which show differing
relationships to mental and physical health. What are the
'healthy' ways of coping with stress and conflict? There are
findings suggesting that both a goal-oriented and an emotionally
reactive way of dealing with threats may be effective strategies.
Both these strategies of handling threat show a positive rela-
tionship to indices of a healthy immune system. A fourth
means of handling conflict is described as defensive hostility,
which involves projection outward of hostility, related only
negatively to psychosomatic health. Indeed, it appears likely
that this means of defense, involving as it does the projection
of threat and hostility, can hardly be considered to be adaptive
as its effect would be most likely self-reinforcing, thereby lead-
ing to only more stress and conflict.
The earlier-mentioned studies of the use of various defense
mechanisms in handling threat and conflict indicate that stress
is not just a mental and behavioral event but also has, in some
cases, a profound and lasting physiological impact. Indeed,
stress can be defined in terms of this threat (where the stimulus
can be of an internal or external origin) to the psychophysio-
logical integrity or well-being of the organism. Generally,
chronic stress has profound deleterious effects on psychophys-
iology through increased release of corticosteroid hormones
and catecholamines. Under normal conditions, the body is
programmed to provide short-term muscular responses to
stress situations. The short-term effect of the aforementioned
hormones is vasoconstriction, and the increased mobilization
of fat and protein provides energy to the tissues in order to
meet the demands of conflict. However, these same mechan-
isms, through the action of the hormones, lead to hyperten-
sion and vascular disorders in the long term. In addition, a
significant reduction in lymphocyte number and activity
occurs, an effect which may explain the now well-established
finding that chronic stress leads to increased susceptibility to a
wide range of illnesses.
The immune system is almost certainly phylogenetically
older than the central nervous system and therefore is accre-
dited as having the primary role of defending the organism
against, as well as adapting it to, the environment. Like the
nervous system, the immune system is also capable of learning,
so much so that the term psychoneuroimmunology is now used to
describe the study of how various components of the immune
system learn to respond to conditioned stimuli. Stimuli
496 The Clinical and Cognitive Psychology of Conflict
that are consistently and associatively paired with those that
previously induced a stress response come also to have a stres-
s-inducing effect when presented alone. The significance of this
can hardly be overstated since it implies a direct link between
illness, hormones, and psychological factors. Already, two adap-
tive systems have been described. Allergic reactions can be seen
as overreactions analogous to the nervous system's anxiety states,
and autoimmune diseases can be regarded as an analogy with
depressive states in that the response becomes internalized
and self-destructive. To quote the actor-cum-director Woody
Allen: "I don't get depressed, I develop a tumor."
The Cognitive and Executive Aspects of Conflict
A concept from cognitive psychology that has been frequently
applied to the study of conflict is called attribution. The notion
of attribution concerns the type of rationalizations that are
given by individuals as explanatory hypotheses for their behav-
ior. If the explanation for one's behavior is said to be attributed
to environmental causes and therefore beyond the control of
the individual, then it is said to be externally determined. This
contrasts with the attribution of personal responsibility for
behavior whence the individual is said to be internally deter-
mined. Research indicates that the type of attributions indivi-
duals make predicts behavior. The positive loading of the
events that are being attributed are of course crucial. For
instance, patients who externally attribute positive events to
others (external attribution) and negative events to themselves
(internal attribution) are reported to score higher on scales of
depression and hopelessness. By contrast, those who persis-
tently attribute positive events to themselves and negative ones
to others and to external forces, are more likely to score high
on paranoia.
The growth of attribution theory has been a major factor in
leading to what is sometimes called the cognitive revolution.
This is a mark of progress in psychology which to some extent
stands in contrast to the other revolution taking place, that of
reducing behavior to neurobiology. The pivotal question is
of course 'is the individual a victim of his neurology' or can
major changes in our concepts of ourselves and the world lead
to changes in behavior and even in our neurological processes.
Some leading authorities in the area of emotion, such as clini-
cal psychologist Arne Ohman, are impressed by the findings
that argue that the brain appears to respond about a half
second before consciousness reports making the decision
(SEE LIBET EFFECT) and conclude therefore that the role of
consciousness is merely to tidy-up afterward by creating mean-
ing out of what has in fact already been decided by the brain.
Yet, it cannot be denied that the recent history of psy-
chology supports the cognitive revolution by the upsurge of
interest in 'consciousness' as an object for scientific study. Max
Velmans and Susan Schneider, the editors of a recent volume
on Consciousness Studies, note how the field, during its 15 years
of existence, has increased to 6 00 000 books and articles.
Consciousness Studies gives organizing structures such as 'the
self and state of consciousness the individual is in, as having
a determining influence on behavior.
The issue, which naturally arises here concerning so-called
nonconscious roots of behavior, is the supposition that not
all the contents of experience and perception are represented
in the self-concept. Perceptions, which are in conflict with
how the person has been taught to view himself and behave,
are either reinterpreted in accordance with this or denied and
therefore not given verbal identification. They are nevertheless
present in the form of so-called nonverbal feelings.
While the principles involved in this process appear similar
to those postulated by the psychoanalytic and psychodynamic
theories of Freud, the descriptive language is clearly different in
terms of being a more humanistic and existential one. Defenses
are thus seen as styles for handling perceptual data and arise in
order to enable the self, as the organizing principle in con-
sciousness, to cope with the demands of the environment.
Defenses arise when the more biological side of personality
(temperament) is in conflict with the 'conditions of worth,'
that is the conditional demands from the environment govern-
ing the expression of love and care toward the individual.
These demands lead naturally to the learning of self-concepts,
social skills, and roles but, in doing so, lead also to the dissoci-
ation of the incongruous, nonaccepted self-experiences that are
in conflict with the learned role.
Flow can this approach deal with the problem that has beset
psychoanalysis: the dubious ontological status of the uncon-
scious? If one follows the phenomenology-cum-functionalism
of William James and Carl Rogers, the solution may be to
simply accept that consciousness exists in many different
forms. Some of these are dissociated from one and another,
and the various forms may utilize different languages of sym-
bolic representation for memories and ranges of experiences.
Supporting this conceptualization are the now well-established
findings relating to state-specific memories. The study of
altered states of consciousness and dissociated states some-
times reveals that these memories are organized as distinct
and conflictual self-representations.
According to this view, conflict and anxiety can then have a
positive function since there is assumed to be a force for growth
existing as a biological principle. This drive operates during the
opportunity for change - in a crisis or in a therapeutic context -
and strives after creating the reintegration of experiences and
split-off parts of the self. Indeed, dissociated and altered states
of consciousness have often been claimed to be associated with
creative problem solving and therapeutic change.
For many years, a medicopsychiatric view dominated clini-
cal psychology, and the findings from behavioral research had
an explanatory input limited to anxiety and phobic states. Even
psychoanalytic categorizations were based on an essentially
nineteenth century Kraeplian system of medical classification.
During the 1980s, although retaining the spirit of the Krae-
plian system of category thinking, there has been an attempt to
produce a theory-free system known as DSM-IV. While the
scientific validity of DSM-IV has been severely criticized, it is
nevertheless generally agreed that the system has incorporated
a more sophisticated view of how predispositional and sit-
uational factors interact. The consensus view among psy-
chologists appears to be that long-term conflicts in the form
of stress-related events - so-called life events - interact with
biological (predispositional) factors to produce a variety of
psychotic and neurotic disorders.
In dealing with conflicts and ensuing stress, research has
identified that there are some basic coping skills. Being
The Clinical and Cognitive Psychology of Conflict 497
successful at skilled performance, however, means not only
learning from one's own experience and gaining positive feed-
back from successes, but also being inspired and learning from
others and then becoming fully involved in meeting the chal-
lenge. It is this combination of factors which Albert Bandura
calls self-efficacy beliefs and enables people to overcome obsta-
cles and reach goals. Bandura was one of the first to emphasize
that the most effective source of learning in mammals and
humans is not that of the classical conditioning but that of
imitation and observational learning.
Positive Psychology
A refreshingly new approach which incorporates much of the
aforementioned points has enabled clinical and health psy-
chology to build an alternative model to the medical one for
the study of mental disturbance. This is known as positive
psychology. Positive psychology approach combines the behav-
ioral work on cognitive functioning with approaches that were
more phenomenological in using terms like self-concept and
self-esteem. Some of the research findings challenge the tradi-
tional ways of seeing mental illness. For instance, those of the
English psychologist Richard Bentall and his coworkers give us
good reason to question the unity and homogeneity of such
disorders as schizophrenia and neurotic disorders. Their find-
ings provide an alternative way of viewing these disorders to
that of illness, in terms of degrees of cognitive dysfunction
arising from biological predispositions and life events. By hav-
ing a knowledge of biological factors such as temperament and
important life events from early attachment onward, one can
then see how certain patterns of dysfunction in beliefs, thought
processes, and emotional expression - what one sees as 'symp-
toms' - can come to develop.
Self-esteem and self-concepts are clearly part of the execu-
tive functioning of the organism and can be seen as closely
connected to the process of attribution. As mentioned earlier,
by using these concepts, it has been shown that patients with
depressive 'symptoms' have excessive internal attributions in the
form of being fixated on negative events relating to themselves
and will tend to interpret events to conform to their pre-
conceived self-concepts. In contrast, deluded and paranoid
patients have been found to make excessive external attributions
in the form of persecutory ideas, perhaps as a defense against
impending depression and low self-esteem, by blaming others
for negative events. Generally speaking, these types of attribu-
tions become the foundation for an optimistic versus a pessi-
mistic lifestyle. A pessimistic lifestyle is an important predictor
of later illness.
Positive psychology is, however, not so much about illness
per se but about our higher and healthy motives: the need for
meaningfulness, happiness, and wisdom. In looking at what
seems to be a healthy functioning style, a series of such rela-
tionships has recently been found by Trevor Archer and his
colleagues at Gothenburg University. They discovered that an
optimistic lifestyle and so-called intrinsic motivation which is
the enjoyment of activities for their own sake rather than for
external reward, together with the expression of positive emo-
tions, were all closely related and formed a personality pattern
of healthy functioning.
The recent history of clinical psychology can be said to
follow the career of one of its most esteemed leaders, Martin
Seligman, who began his work as a behavioral psychologist
contributing to the concept of learned helplessness. Seligman
moved on to become one of the founders of cognitive psychol-
ogy focusing on attributional styles and now finally is the
major proponent of 'positive psychology' with its emphasis
on existential problems and authentic happiness in life. This
reflects how the old divisions between behavioral, psychoana-
lytic, and humanistic psychology have begun to fade away.
Behavioral psychology used to have the dictum: change behav-
ior and you change the way of seeing things. For psychoanaly-
sis, it was first insight and then change. While the aphorism for
today's cognitive psychology is change the person's constructs
and you can change his behavior, it is probably true to say that
in practice, most psychologists use all the three means where
and when appropriate.
The research behind this change has led to tangible effects
and the current interest in Europe in cognitive behavior therapy
(CBT) is one outcome of this. Following the recent evalua-
tion of antidepressive medication, showing this to be a costly
and largely ineffective treatment, there is now government-
empowered support for providing the public with cost-
effective evidence-based treatment. CBT tends to be the most
favored among the various therapies, although there is no
decisive evidence for its superiority over the more personal
and dynamic forms of therapy. Nevertheless according to lead-
ing psychologist and governmental advisor, David Clark, the
UK government plans during the next 3 years to educate 3600
psychological therapists under the supervision of clinical psy-
chologists with the aim of providing a treatment for the
900 000 citizens who are estimated to need it for the treatment
of anxiety and depression.
The enormous interest in consciousness and its role in
cognitive processes mentioned earlier may be the reason why
there is now an otherwise unlikely allegiance between Zen
Buddhism and behavioral psychology teaching 'mindfulness'
meditation. Mindfulness is the focusing of attention on the
flow of immediate experience in the here and now along with
openness, acceptance, and curiosity to its content. Mindfulness-
Based Cognitive Therapy is aimed at breaking habitual thought
and reaction patterns and expanding awareness to include
feelings and bodily reactions that would not otherwise be
attended to: in short, access to otherwise nonconscious pro-
cesses. Clearly, the current focus of psychology gives us reason
to modify Dewey's axiom, which we began with, as being a
simplification. Humans may not be very effective as achieving
it, but they do seek meaningfulness.
Resolving Social Conflict
While it is a long way from the study of rat behavior and the
psychoanalytic couch to the resolution of internal and group
conflict, already during the 1960s psychologists were becom-
ing interested in the application of theories about group iden-
tity to social behavior. This will be covered more fully in other
articles, but some landmarks are worth noting.
Muzafer Sherif and his coworkers carried out what is now
regarded as a classic experiment in the resolutions of group
498 The Clinical and Cognitive Psychology of Conflict
conflict. Working with boys at a summer camp, the researchers
followed the process of how group identity and rivalry arose.
Different codes of behavior evolved and with it conflict and
opposition. Open hostility developed between the groups.
At this stage, various techniques for intervention and conflict
resolution proved successful. One of these was the formation
of superordinate goals that required the cooperation of both
groups in order to resolve an externally imposed threat.
Another was the introduction of methods to promote interac-
tion and contact between the groups.
There are a few examples of psychological intervention in
real group conflict. One well-known example, albeit almost on
an anecdotal scale, was the successful involvement of Carl
Rogers in facilitating openness and communication between
members of a small group of Catholics and Protestants in
Northern Ireland. Rogers relied on creating a therapeutic atmo-
sphere of empathy, acceptance, and warmth.
For many years, practical solutions to racial and group
conflict were influenced by Gordon Allport and his suggestions
for reducing conflict by increasing contact between groups
and focusing on similarities between individuals. However,
with the recent development of the approach known as Social
Identity Theory, the emphasis has now shifted away from the
individual toward dealing with intergroup differences and pro-
moting an understanding of the need to maintain these in the
face of threat and discrimination.
Another major contribution came from social psychologist
Irving Janis who described how group decisions often make
systematic errors of judgment which he called groupthink and
mindguarding. Groupthink refers to the pressures in the group
to maintain an illusion of unanimity, and mindguarding refers
to the efforts to maintain solidarity and protect the leader from
stress and contradicting viewpoints. Irving recommended the
use of a 'devil's advocate' as a counter to groupthink.
Other psychologists such as Max Bazerman have extended
their expertise to the area of negotiations. Bazerman describes
the common mistakes that negotiators make such as being
locked in and committed to one perspective, seeing the others'
gains as their losses, and being overconfident of success.
Although a strong case can be made for the involvement of
a psychologist in international conflict, this seems to rarely
occur. However, one psychologist who has made an important
contribution in the global context is Albert Bandura who was
mentioned earlier in the context of self-efficacy and observa-
tional learning. Bandura now works with the film media in the
entertainment-education area, producing films in order to
encourage social change. These are films intended primarily
for third -world countries addressing major issues such as over-
population, genital mutilation, the education of women, and
literacy. Role models, in the form of highly admired actresses
and actors, and plot-lines are used to debunk false beliefs and
encourage people's beliefs about their own efficacy.
There is a classical quotation from the days of behaviorism
by the well-known psychobiologist, Frank Beach: "Psychology
is the study of the white rat and the college sophomore."
Clearly, the aforementioned research shows that today's psy-
chology has come a long way and now is a dynamic and reality-
relevant subject.
See also: Attribution; Cognitive Behavior Therapy; Gestalt Psychology
and the Development of Perceptual Organization; Positive Psychology;
Subliminal Perception.
Further Reading
Bandura A (2009) Social cognitive theory goes global. The Psychologist
22(6): 504-506.
Barsky AE (2000) Conflict Resolution for the Helping Professions. Stamford, CT:
Thomson Learning.
Baumgardner SR and Crothers MK (2010) Positive Psychology. Upper Saddle River,
NJ: Pearson.
Bentall R (2009) Doctoring the Mind. London: Penguin Books.
Dijksterius A and Van knippenberg A (1998) The relation between perception and
behavior, or how to win a game of trivial pursuit. Journal of Personality and Social
Psychology 74: 865.
Hassin R, Uleman J, and Barh J (eds.) (2005) The New Unconscious. Oxford: Oxford
University Press.
Janis IL (1972) Victims of Groupthink. Boston, MA: Houghton Mifflin.
Kelly EF, Kelly W, Crabtree A, Gauld A, Grosso M, and Greyson B (2007) Irreducible
Mind. Toward a Psychology for the 21st Century. Lanham, MD: Rowan & Littlefield.
Kihlstrom J (1990) The psychological unconscious. In: Pervin L (ed.) Handbook of
Personality: Theory and Research. New York: Guilford.
Kihlstrom JF, Mulvaney S, Tobias BA, and Tobis IP (2000) The emotional unconscious.
In: Eich E, Kihlstrom JF, Bower GH, Forgas JP, and Niedenthal PM (eds.) Cognition
and Emotion, pp. 30-86. New York: Oxford University Press.
Oilman A (2000) Fear and anxiety: Evolutionary, cognitive, and clinical perspectives.
In: Lewis M and Haviland-Jones J (eds ) Handbook of Emotions. New York: Guilford
Publications.
Synder CR and Lopez SJ (2002) Handbook of Positive Psychology. Oxford: Oxford
University Press.
Velmans M and Schneider S (eds.) (2007) The Blackwell Companion to Consciousness.
Oxford: Blackwell.
Relevant Websites
http://crinfo.org/ - The Conflict Resolution Information Source.
http://www.ppc.sas.upenn.edu/ - Positive Psychology Center.
http://www1.eur.nl/fsw/happiness/ - World Database of Happiness.
http://mbct.co.uk/about-mbct/ - Mindfulness-Based Cognitive Therapy.
http://www.ted.com/talks/martin_seligman_on_the_state_of_psychology.html -
Martin Seligman on positive psychology.
Clinical Assessment
E Gambrill University of California, Berkeley, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Antecedents Events that immediately precede behavior and
influence its frequency.
Behavior Any measurable or observable act or response.
Behavior is defined broadly in some perspectives to
include cognitions, feelings, and physiological reactions
which, although they are not directly observable, are
defined so that they can be measured.
Behavioral Approaches to understanding and altering
behavior in which learning histories and current
environmental contingencies of reinforcement and behavior
repertoires are considered. These approaches differ in the
degree of attention devoted to thoughts (e.g., cognitive-
behavioral compared to applied-behavior analysis).
Clinical inferences Assumptions about causes and
remedies regarding problems.
Cognitions Beliefs, schemas, self-statements, attributions.
Cognitive approach Attention is devoted to thoughts.
Cognitive-behavioral approach Attention is devoted to
thoughts as well as behavior.
Consequences Events that follow behavior and influence its
frequency.
Contextual (ecological) assessment Attention is devoted
to individual and environmental factors. This is an
integrated appeal.
Contingencies Relationships between behaviors and the
events that follow (consequences) and precede
(antecedents) them.
Evidence-based practice A process and philosophy
designed to forward effective integration of clinical,
evidentiary and ethical concerns.
Interpersonal approach There is a focus on interpersonal
relationships.
Psychodynamic Approaches to understanding
behavior in which unconscious mental and emotional
processes (e.g., motives, defenses, conflicts) related
to past and current experiences are emphasized for
example in discussing moral conflicts.
Reliability The extent to which a measure is stable
over time (stability). There are other kinds of reliability
such as interitem reliability.
Transdiagnostic approaches Approaches emphasizing
similarities across 'disorders.'
Validity The extent to which a measure measures
what it was designed to assess. There are different
kinds of validity (e.g., predictive, content, concurrent,
construct).
Clinical assessment involves the clarification of presenting
concerns and related factors including identification of out-
comes that will be focused on. It should offer guidelines
for selection of intervention methods. Assessment lays the
groundwork for selecting plans and should indicate how likely
it is that hoped-for outcomes can be attained. The likelihood
may be slim when needed resources are not available or
when problem-related circumstances cannot be rearranged.
Assessment should indicate what factors influence clients'
options. It should suggest 'leverage points' for pursuing valued
outcomes. Interrelated goals (sometimes referred to as case
formulation or case conceptualization) include: identifying
and clarifying presenting concerns, detecting characteristics of
clients and their environments that influence those concerns,
interpreting and integrating data collected, and, in collabora-
tion with the client, selecting outcomes to focus on. Assess-
ment should offer clients more helpful views of problems and
a useful vocabulary for describing concerns and options. Deci-
sions must be made about what data to collect, how to gather
this, and how to organize it. This phase is an ongoing process
in which assumptions are altered as needed in response to new
information, such as degree of progress.
The assessment methods used differ because of differences
in theoretical perspectives which influence the kind of data
collected, how it is gathered and the uses and functions of
these data. Carrying out an assessment is like locating the
pieces of the puzzle. Certain pieces of the puzzle are sought
rather than others depending on the clinician's theoretical
orientation and knowledge, and puzzle completion may be
declared at diverse points. Related decisions and inferences
differ in how closely they attend to related empirical findings.
Issues of practicality may limit evidence available. Data are
sought that are useful, reliable, and valid. Specialized knowl-
edge may be required and critical thinking skills are needed to
weigh the value of evidence and examine the soundness of
assumptions. Although decisions must typically be made on
the basis of incomplete data, without a sound assessment
framework, opportunities to gather useful data may be lost
and ineffective or harmful plans suggested. Collecting irrele-
vant data wastes time and money and increases the likelihood
of incorrect decisions. Assessment methods can be harmful as
well as helpful. Limitations of and related potential adverse
consequences of assessment/diagnostic methods, such as anxi-
ety and worry because of a high rate of false positives, are often
not candidly acknowledged. Even when the decision to use a
test is well argued, there may be risks which should be clearly
described to clients. Hiding potential harm and inaccuracy of
assessment methods are common in the helping professions.
499
500 Clinical Assessment
There is general agreement that an individualized assess-
ment which considers cultural differences should be con-
ducted. This does not mean that this is indeed done and
practice perspectives differ in what is focused on. Individualized
assessment avoids the client uniformity myth in which clients
(or families, or groups) are mistakenly assumed to be similar.
Behavior consists of different response systems, which may or
may not be related depending on each individual's unique
history (1) overt behavior (e.g., avoidance of crowds) and
verbal reports (verbal descriptions of anxiety); (2) cognitions
(thoughts about crowds); (3) physiological reactions (e.g.,
increased heart rate when in crowds). Each person may have a
different pattern of responses in a situation. Only through an
individualized assessment can these unique patterns and
related situations be discovered. Suicidal potential should be
assessed as relevant. Recognizing the signs of pathology is
important anytime. This would be helpful in understanding
what can be accomplished and how it can be accomplished.
A clear mutual agreement between clinicians and clients about
the focus of helping increases the likelihood of client
engagement.
Forming a conceptualization of concerns, one that is
shared by both the clinician and the client that will be helpful
in attaining hoped-for outcomes, is an integral aspect of
assessment. The kind of conceptualization suggested will
depend on the theoretical orientation of the clinician. It is
important to arrive at a common view of hoped-for-outcomes
and how to attain them. This common view is a motivating
factor in that, if clients accept it and if it makes sense to them,
there will be a greater willingness to try out interventions that
flow from this account. Mutually agreed-on views are fostered
in a variety of ways, including questions asked, assessment
methods used, and rationales offered. Focused summaries
help to put material together within a new conceptualization.
Identifying similar themes among seemingly disparate events
can be used to suggest new views. High levels of facilitating
relationship skills are associated with success and thus vital to
develop and use.
The Guiding Role of Practice Theories
How presenting concerns are structured is a key part of clinical
decision-making. Assessment frameworks differ in what is
focused on, the kinds of assessment methods used, and how
closely assessment is tied to selection of intervention methods.
Preferred practice theories influence what clinicians look for
and what they notice as well as how they process and organize
the data collected. Theories favored influence beliefs about
what can be and is known about behavior and how related
knowledge can be developed. Theories differ along the follow-
ing dimensions:
• Unit of concern (individual, family, community, society).
• Goals pursued (e.g., explanation and interpretation alone
or understanding based on prediction and influence).
• Clarity of goals pursued.
• Criteria used to evaluate the accuracy of explanations
(e.g., consensus, authority, scientific).
• Range of problems addressed with success.
• Causal importance attributed to feelings, thoughts, and/or
environmental factors.
• Range of environmental characteristics considered (family,
community, society).
• Causal importance attributed to biochemical causes.
• Attention devoted to past experiences.
• Degree of optimism about how much change is possible.
• Degree to which a perspective lends itself to and encourages
empirical inquiry (finding out whether it is accurate).
• Degree of empirical support (evidence for and against a
theory).
• Attention given to documenting degree of progress.
• Ease with which practice guidelines can be developed.
• Degree of parsimony.
Frameworks differ in the attention given to observation of
interactions in real-life settings, in whether significant others
(those who interact with and influence clients) are involved,
and how directive clinicians are. They differ in the degree of
attention paid to cognitions (thoughts), feelings, environmen-
tal characteristics (such as reactions of significant others), and
biomedical factors. Different frameworks are based on differ-
ent beliefs about the causes of behavior. Beliefs about behav-
ior, thoughts, and feelings, and how they are maintained and
can be changed influence what data are gathered and how
data are weighted and organized. Presenting concerns can be
viewed from a perspective of deficiencies (e.g., psychiatric
labels) or from a contextual view in which both personal
and environmental factors are attended to. For example, a
key point of feminist counseling is helping clients to under-
stand the effects of the political on the personal, both past
and present. Frameworks that focus on psychological charac-
teristics of individuals are based on the view that behaviors is
influenced mainly by such characteristics. There is a focus on
interiors - on understanding private phenomena believed to
cause overt behavior such as mental schemas and disposi-
tions. In interactional perspectives, attention is given not
only to the individual but to people with whom he or she
interacts. Interactional views differ in how reciprocal the rela-
tionship between the individual and the environment is
believed to be and in the range of environmental events
considered.
In contextual, ecological perspectives, individual, family,
community, and societal characteristics (political, social, and
economic) are considered as they may relate to desired out-
comes. A contextual framework decreases the likelihood of
focusing on individual pathology (blaming the victim), and
neglecting environmental causes and resources. Practice per-
spectives that focus on individual causes of personal problems
may result in 'psychologizing' rather than helping clients. The
Diagnostic and Statistical Manual of Mental Disorders (2000)
(DSM) published by the American Psychiatric Association
encourages a focus on individual pathology, listing hundreds
of behaviors as potential indicators of mental disorders (ill-
nesses). The fifth edition will include even more. Transdiag-
nostic approaches focus on what is common across a variety of
disorders. Assessment frameworks differ in the extent to which
they take advantage of empirical information about behavior
including factors related to client concerns, and the accuracy of
different sources of data.
Clinical Assessment 501
Sign and Sample Approaches: Diagnosis
and Assessment
In a sign approach to assessment, observed behaviors are
viewed as indicators of more important underlying (and unob-
served) personality dispositions (typically of a pathological
nature) or traits. Traits can be defined as a general tendency
to react in consistent and stable ways. Examples are 'aggression'
and 'extraversion.' Inherent in sign approaches, is the assump-
tion that observable behaviors are the outward signs of some
underlying process which must be altered to bring about any
lasting change. For example, a clinician may conclude that a
child who has difficulty concentrating on his school work and
sitting in his seat is hyperactive. A focus of change efforts on the
behavior itself, according to this model, would not succeed,
because no change has supposedly been brought about in
underlying causes. The underlying hypothetical constructs are
viewed as of major importance in understanding and predict-
ing behavior. Such dispositional attributions shift attention
away from observing what people do in specific situations to
speculating about what they have or are. In the DSM, behaviors
are assumed to equate with mental disorders. This classifica-
tion system has come under increased criticism for example for
ignoring contextual factors related to client concerns and
pathologizing every day problems in living such as common
(mis)behaviors on the part of children.
The interactions between wishes, anticipated threats if
wishes are expressed, and the processes used to cope with or
defend against conflictual situations, are of interest in psycho-
dynamic frameworks. Important elements in such processes
are believed to be beyond conscious recognition of the indi-
vidual experiencing them even when they may be recognized
or inferred by others. The concepts of developmental stages
and 'mechanisms' (psychological processes such as defense
mechanisms) are central concepts. There is a focus on interiors.
Defensive aims, processes, and outcomes are of interest.
Defense mechanisms include suppression, undoing, repres-
sion, role reversal, projection, and regression. The defenses
are believed to be heightened under conditions of high emo-
tion, stress, and conflict. Motives include the wish to avoid
unpleasant, overwhelming, or out-of-control states. Some
unconscious processes anticipate such outcomes. Other aspects
of psychoanalytic approaches include an emphasis on verbal
reports concerning early histories and efforts to alter inner
processes by verbal means. Compared to behavioral assess-
ment, less attention is devoted to environmental variables
that may influence behavior because of the assumed core
relevance and stability of underlying dispositions. There are
different kinds of psychodynamic assessment frameworks. For
example, there are variants of object relations theory, each of
which may have a somewhat different approach to assessment.
In object relations theory, the concepts of mirroring and self
objects are key ones. Attention is given to internal mental
representations of the self and significant others. The term
'object relations' refers to the interplay between the images of
self and others. Ego psychology emphasizes identification and
support of strengths and working within the 'defenses' rather
than breaking them down. Proponents consider resistance to
change natural and work with and support adaptive strengths.
Defense mechanisms, such as rationalization of actions and
projection of feelings onto others are identified but not neces-
sarily discussed.
Behavioral assessment involves a sample approach. In a sam-
ple approach, direct observation of behavior in real-life settings
or, if this is not possible, in situations that resemble these, is
valued. Labels are used as summarizing categories rather than
as terms indicating some underlying mental disorder. Unlike
in sign approaches where the cause of behavior is assumed to
be underlying dispositions, the cause of behavior is assumed
to lie largely in environmental contingencies. Behavioral fra-
meworks differ in the attention devoted to thoughts and envi-
ronmental contingencies. Differences in focus are so marked
that they have resulted in the formation of different journals,
societies, and names (e.g., 'cognitive,' 'cognitive-behavioral,'
'applied behavior analysis'). Differences in emphasis are
related to the role attributed to thoughts in influencing behav-
ior. This role varies from a causal to a mediating role. In the
former, reflected in cognitive and cognitive-behavioral frame-
works, thoughts are presumed to play a key role in creating
feelings and behaviors. In the latter, reflected in applied behav-
ior analysis, thoughts are assumed to influence feelings and
behavior in a mediating (not causal) manner. It is assumed
that one must look to past and present environmental contin-
gencies to account for both thoughts and feelings. It is assumed
that feelings and thoughts are (covert) behaviors which them-
selves require explanation.
In cognitive-behavioral methods, attention is devoted to
thoughts as well as behaviors. Thoughts of interest include
attributions for behavior, feelings, and outcomes, negative
and positive self-statements, expectations, and cognitive distor-
tions. Attention is devoted to identifying the particular kinds of
thoughts that occur in problem-related situations including
related 'schema.' Cognitive-behavioral approaches differ in
their assumptions about the kinds of thoughts that underlie
behavior. However, all share certain assumptions such as the
belief that individuals respond to cognitive representations of
environmental events rather than to the events per se. It is
assumed that learning is cognitively mediated and that cogni-
tion mediates emotional and behavioral dysfunction. In
applied behavior analysis, environmental contingencies are
focused on. A contingency analysis requires identification of
the environmental events and occasion that maintain behav-
ior. There is an interest in describing the relationships between
behavior and what happens right before and after as well as
'metacontingencies' - the relationships between cultural prac-
tices and the outcomes of these practices. There is an emphasis
on current contingencies. Attention is directed toward the
change of 'deviant' environments rather than the change of
'deviant' client behaviors. There is an interest in identifying
functional relationships. A behavioral analysis includes a
description of behaviors of concern as well as evidence that
specific antecedents and consequences influence these beha-
viors; it requires a functional as well as a descriptive analysis.
Although there are differences, all behavioral approaches
share many characteristics that distinguish them from sign
approaches. Assessment is an ongoing process in behavioral
assessment. This contrasts with some traditional approaches
in which assessment is used to 'diagnose' a client in order to
502 Clinical Assessment
decide on treatment methods. What a person does is of interest
in behavioral approaches rather than what she has. Behavior is
of great interest, especially the behaviors of individuals in real-
life contexts and those with whom they interact. Identifying
variables that influence the frequency of behaviors of interest
including positive alternatives to behaviors to be decreased is a
key assessment goal. Behavior is assumed to vary in different
contexts because of different learning histories and different
current contingencies as well, for example, different levels of
deprivation and fatigue.
The focus on behavior has a number of implications for
assessment. One is the importance of observing people in real-
life contexts whenever feasible, ethical, and necessary to acquire
helpful data. A range of assessment methods is used including
observation in real-life settings as well as role plays. Multiple
assessment methods are also called for because of the lack of
synchrony in overt behavior, physiological reactions, cogni-
tions (thoughts), and feelings. Assessment and treatment are
closely related in a behavioral model. It is assumed that assess-
ment should have treatment utility. There is an emphasis on the
use of valid assessment methods. The principles of behavior are
relied on to guide assessment and intervention. There is a pref-
erence for limited inference and a focus on constructing reper-
toires (on helping clients to acquire additional knowledge and
skills that will increase opportunities for reinforcement). Cli-
ents are viewed in terms of their assets rather than their deficien-
cies. The preference for enhancement of knowledge and skills
requires a focus on behaviors that are effective in real-life con-
texts. In a task analysis, the specific behaviors that are required
to achieve an outcome are identified. For each step, perfor-
mance is clearly described as well as the conditions in which
it is expected to occur.
Some Important Distinctions
The form of a behavior (its topography) does not necessarily
indicate its function (why the behavior occurs). Identical forms
of behavior maybe maintained by very different contingencies.
Just as the same behavior may have different functions, different
behaviors may have identical functions. The distinction be-
tween motivational and behavioral deficits is also important.
If a desired behavior does not occur, this may indicate either that
the behavior exists but is not reinforced on an effective schedule
or is punished (a motivational deficit) or that the behavior is not
present in the client's repertoire (a behavior deficit). Motiva-
tional deficits are often mistaken for behavioral deficits. Moti-
vational and behavioral deficits can be distinguished by
arranging conditions for performance of a behavior. For exam-
ple, clients could be requested to role play behaviors and asked
whether similar or identical behaviors occur in other situations.
Behavior surfeits are often related to behavior deficits. For exam-
ple, aggression on the part of a child may be related to a lack of
friendship skills. It is also important to distinguish response
inhibitions from behavior deficits. Emotional reactions such
as anxiety may interfere with desired behavior.
Past History
Although the past is viewed as important in influencing current
behavior in just about all perspectives, assessment frameworks
differ in how much attention is devoted to the past and what is
focused on. Past experiences are a major focus in psychody-
namic frameworks. Knowledge about past circumstances may
be of value when it is difficult to identify current maintaining
factors. Information about a client's past may provide valuable
information about unusual histories related to concerns. An
understanding of how problems began can be useful in clarify-
ing the origins of what seem to be puzzling reactions. New ways
of viewing past events may be helpful to clients. Information
about a client's past may help clients to understand the source
of current reactions. Demographic indicators about a client's
past behavior in certain contexts may be better predictors of
future behavior than personality tests or clinical judgments.
Information about the past offers a view of current events in
a more comprehensive context. Major areas include medical
history, educational and work history, significant relationships,
family history, and developmental history. Helpful coping
skills may be discovered by finding out what clients have tried
in the past to attain valued outcomes. Research concerning
autobiographical memory suggests that memories change
over time, making it difficult to know whether reports are
accurate. From a psychodynamic perspective, accuracy would
not be an issue. Rather, the client's memories of events, whether
accurate or not, are the substance of import. It is assumed in
fact that memories may be distorted by unconscious motives/
conflicts and so on. Excessive attention to past troubles may
create pessimism about the future and encourage rationaliza-
tions and excuses that interfere with change, especially if this is
not fruitful in selecting effective intervention plans.
What About Labels?
The term diagnosis refers to accurately identifying the cause(s)
of symptoms of an individual. The term diagnosis was bor-
rowed from medicine, in which a physician makes a diagnosis
of a patient's condition and then recommends a treatment
based on this diagnosis. Diagnoses are used as a guide to
selection of interventions. Evidence-based diagnosis can be
defined as the use of certain symptoms and signs (feeling
hot and having a fever as determined by a thermometer) to
identify an underlying condition (a bacterial infection). Such a
diagnosis is evidence-based in two senses (1) the signs and
symptoms accurately indicate the underlying condition and
(2) accurate identification of the underlying condition points
to an effective remedy (e.g., antibiotics). Thus in medicine,
we often have both signs (an X-ray) and symptoms (a cough).
Is this true of psychological concerns such as depression
and anxiety? The answer from psychiatry is yes. It is assumed
that symptoms (e.g., anxiety in social situations) reflect an
underlying disorder (or dysfunction). But what signs do they
use? Here, signs are observed by the examiner. Diagnosis and
assessment are not either/or endeavors. Providing a diagnosis
for clients is a requirement for third-party payment. This does
not mean that clinicians cannot or should not carry out
an evidence-informed contextual assessment describing, for
example, personal and environmental factors that create risks
and contribute to resilience.
Labels are used in assessment in two main ways. One is
as a shorthand term to refer to specific behaviors. The term
unmanageable may refer to the fact that a student often gets
Clinical Assessment 503
out of his seat and talks out of turn in class. A teacher may use
the label 'hyperactive' as a summary term to refer to these
behaviors. With psychiatric labels, the label connotes more
than a cluster of (mis)behaviors. It involves additional
assumptions about the person so labeled. The Diagnostic and
Statistical Manual of Mental Disorders (DSM-IV) includes
hundreds of labels. Methodological and conceptual problems
connected with the use of psychiatric labels include lack of
agreement about what label to assign clients and lack of asso-
ciation between a label and indications of etiology or what
intervention will be effective. Indicators on which psychiatric
labels are based are vague, say little about positive attributes,
potential for change, and change that does occur. Both trait
and diagnostic labels offer little detail about what people do in
specific situations and what specific circumstances influence
behavior. Some behaviors may appear 'trait-like' in that they
are similar over time and situations because of similar contin-
gencies of reinforcement. Degree of consistency should be
empirically explored for particular clients and behavior rather
than assumed.
Labels are classifications and are thus subject to all the ways
in which classifications are wrong (e.g., under or over inclu-
sion). The classification system may be faulty in not accurately
distinguishing between different 'syndromes,' that is, there
maybe 'boundary' problems. Clients maybe given an incorrect
label (it does not really describe them) or a label that does
accurately describe them may not be used and valuable services
may be withheld. Labels can obscure important individual
variations among people and thus encourage the 'patient uni-
formity myth.' Use of labels may overlook cultural differences
and encourage 'the psychopathologist's fallacy - the belief that
because a child has been brought in as a patient, there must be
something wrong with him or her.' We may start to think of a
person as the label he or she is given. We may think our work is
over when we give a person a diagnostic label rather than
realizing it is just beginning in relation to carrying out a con-
textual assessment. Having a label may result in seeking out
characteristics that complement the label.
Acceptance of a psychiatric label may prematurely close off
consideration of promising options. The tendency to use a
binary classification system (people are labeled as either hav-
ing or not having something, e.g., as being an alcoholic or not),
obscures the varied individual patterns referred to by a term.
Psychiatric labels have been applied to an ever-increasing vari-
ety of behaviors viewed as 'mental illnesses.' Critics of the DSM
highlight the consensual nature of what is included (reliance
on agreement rather than empirical criteria) and the role of
political and economic influences in its creation. Some writers
argue that psychiatric classification systems trivialize human
problems in living and encourage blaming victims for their
plights rather than examining and altering the social circum-
stances that are often responsible. For example, they argue that
they direct attention away from political, social, and economic
conditions related to expected gender roles and focus on indi-
vidual deficiencies. Many scholars argue that labels, such as
'mental illness,' are used for social control purposes and often
result in harming rather than helping people. The American
Psychiatric Association claims that its classification system is
based on scientific evidence. However critiques raise doubts
about these claims and discuss the role of political and
economic considerations in the creation and 'selling' of the
DSM and document reliability and validity problems with this
system. These include the consensual nature of what is
included (agreement among individuals is relied on rather
than empirical criteria), lack of agreement about what label
to assign clients (poor reliability) and lack of association
between a diagnosis and indications of what plans will be
effective.
Although they may sound sophisticated, too often labels
offer few if any guidelines about what to do to resolve pro-
blems. In the introduction to the DSM, it states that diagnostic
labels say nothing about etiology nor offer guidelines for selec-
tion of service. Neither trait nor diagnostic labels offer enough
detail about what people do in specific situations and what
personal and environmental events influence their behaviors
to be of value in clarifying problems and related factors. The
individual variability of behavior is overlooked, resulting in
lost opportunities to discover unique relationships between a
client's behavior and his or her environments. Dispositional
attributions shift attention away from observing what people
do in specific situations to speculating about what they might
do or have done.
Labels that are instrumental (they point to effective inter-
ventions) are helpful. Failure to use labels that are informative
may prevent clients from receiving appropriate intervention.
Labels can normalize client concerns. Parents who have been
struggling to understand why their child is developmentally
slow may view themselves as failures. Recognition that their
child has a specific kind of developmental disability that
accounts for this can be a relief.
A Contextual Evidence-Informed Assessment
A contextual assessment includes a clear description of hoped-
for-outcomes and related factors and description of what a
person can do and cannot do, what he or she can learn to do,
and what is expected of him or her, as well as environmental
factors that influence problem-related behaviors and positive
alternatives. It is assumed that only if complaints are clearly
described can they be translated into specific changes that
would result in their removal. The emphasis on behavior and
the influence of environmental contingencies call for the trans-
lation of problems into observable behaviors and the discovery
of ways in which the environment can be rearranged. Clients
are encouraged to recognize and alter the role they play in
maintaining problems. For example, teachers and parents
often reinforce behaviors they complain about. Assessment is
individualized; each person, group, family, organization or
community is viewed as unique. Data about group differences
do not offer precise information about what an individual does
in specific situations and what cues and consequences influ-
ence their behavior.
Practice-related research regarding concerns and possible
causes and ways to discover them are drawn on. This kind of
assessment often reveals that environmental deficiencies con-
tribute to problems such as a low frequency of positive social
interactions, lack of recreational opportunities or day care,
and low wages. A contextual assessment encourages the
description of processes rather than the study of conditions.
504 Clinical Assessment
The characteristics of contextual, evidence-based assessment
include the following:
1. Individually tailored as needed.
2. Use of assessment frameworks and methods that have
been critically tested and found to be of value.
3. A preference for testing inferences.
4. A focus on clients' assets rather than their deficiencies.
5. A focus on the present.
6. Attention to current contingencies in interrelated system
levels.
7. A clear description of concerns and related hoped-for
outcomes.
8. A preference for observation of important interactions (as
ethical, practical) are needed.
9. A clear description of assessment methods.
10. A close relationship between assessment and intervention.
1 1 . Offering high level relationships skills.
A contextual framework decreases the likelihood of errors, such
as emphasizing people's deficiencies.
A contingency analysis involves describing the context
in which problems occur (i.e., the relationships between
behavior in real-life settings and what happens right before
and after), including alternative behaviors that, if increased,
would compete successfully with undesired behavior. Consid-
erable advances have been made in some areas in identify-
ing problem-related contingencies. For example, assessment
protocols have been developed to identify the cues and
consequences related to self-injurious behavior of children.
A descriptive analysis involves identification of problem-related
behaviors and associated setting, events, antecedents, and con-
sequences. A functional analysis includes altering events
identified in a descriptive analysis to determine their actual
influence. Implications for assessment of a focus on contin-
gencies include observing people in real-life contexts when
feasible, ethical and necessary to clarify problems and their
causes. A second is an emphasis on collecting information
about individuals. A focus on vague problems may prevent
the discovery of related factors. If hoped-for outcomes are not
clearly described, it will be difficult to determine the frequency
of behaviors of interest or to assess progress. The specific
referents to which clients refer when using vague terms (such
as uncooperative, immature, or aggressive) may differ.
Sources of Influence
Influences on behavior include other people's actions, the phys-
ical environment, tasks and materials, physiological changes,
thoughts, genetic differences, and developmental factors. Mate-
rial and community resources and related political, economic,
and social conditions influence options. It is important to
obtain an overview of the client's current life as this may relate
to problems, including relationships with significant others,
employment, physical health, recreational activities, and com-
munity and material resources available. Antecedents of behav-
ior, like consequences, have a variety of sources. In addition to
proximal antecedents (those that occur right before a behavior),
distal antecedents may influence current behavior. Past or
future events may be made current by thinking about these.
These thoughts may then influence what we do, feel, and think.
Setting events are antecedents that are closely associated with a
behavior but are not in the situation in which behaviors of
concern occur. For example, an unpleasant exchange with a
teacher may influence how a child responds to his parents at
home. The earlier event alters the likelihood of given reactions
in subsequent situations.
Preferred practice theories influence the attention given to
various sources. Problems vary in the complexity of related
factors. Problems may be complex because clients lack needed
skills, have interfering beliefs, or are threatened by proposed
changes. Distinguishing between problems and efforts to
resolve these will avoid confusion between the results of
attempted solutions and effects of the original concern.
Expected role behaviors in a certain culture may limit options.
Ongoing discrimination against a group may limit opportu-
nities. Clients may lack needed information or skills. A behav-
ior deficit may exist (the client may not know how to perform
a given behavior).
Other People/The Nature of the Client’s Social Relationships
The influence of significant others in the maintenance of a
troubled or troubling behavior should be explored. Behavior
occurs in a context. How significant others respond makes up
an important part of our environment. Significant others are
those who interact with clients and influence their behavior.
Examples include family members and staff in residential set-
tings. Significant others are often involved in assessment. For
example, in family therapy, family members participate in
assessment and understanding relationships among family
members is a key part of assessment. Interactions between
couples is closely examined in relationship counseling. It is
important to assess the nature and quality of the client's social
network and social support system. Clients may lack social
support such as opportunities for intimacy and companion-
ship or the opportunity to provide support to others. Social
interactions may be a source of stress rather than of pleasure
and joy.
The Physical Environment
The influence of the physical environment should also
be examined. Physical arrangements in residential and day
care settings influence behavior. Unwanted behaviors may be
encouraged by available materials. For example, toys that
are visible to children may distract them from educational
tasks. Temperature changes affect behavior as do degree of
crowding and noise level. Characteristics of the community
in which clients live that may influence complaints and possi-
ble intervention options should be assessed. Neighborhood
quality influences well-being. For example, children who live
in lower quality environments (e.g., there is little play space,
housing is in industrial neighborhoods, upkeep of streets is
poor) are less satisfied with their lives, experience more nega-
tive emotions, and have more restricted and less positive
friendship patterns. There is a relationship between the num-
ber of nonaccidental injuries to children and the physical con-
ditions of the home which is related to socioeconomic status.
Clinical Assessment 505
Tasks and Activities
The kind of task confronting an individual may influence the
rate of problem behavior. Particular tasks or activities may be
high-risk situations for unwanted behavior. Many studies have
found a relationship between the kind of task and behavior
such as self-injury. Problems may occur because a task is too
tedious or difficult or because an individual is uncomfort-
able or bored, or is told to do something in an unpleasant
manner. In these instances, altering antecedents may correct
the problem.
Biophysiological Factors
Presenting problems may be related to neurological or bio-
chemical factors. Such factors may place boundaries on how
much change is possible. Malnutrition may contribute to beha-
viors complained of. Discovering biochemical abnormalities
only establishes that they are present, not that they cause
certain behaviors. Premature acceptance of biophysical expla-
nations will interfere with discovering alternative explana-
tions that yield knowledge about interventions. Biochemical
changes may be a result of stress related to social conditions
such as limited opportunities due to discrimination. Certain
kinds of illness are associated with particular kinds of psycho-
logical changes. Drugs, whether prescribed or not, alcohol,
environmental pollutants, and nutritional deficiencies may
influence health and behavior. Behavior changes may be due
to brain tumors. Hormonal changes/imbalances may result in
mood changes which may be misattributed to psychological
causes. On the other hand psychological changes may be mis-
attributed to hormonal changes. People differ in return of
diffuse physiological arousal (DPA) to baseline levels. When-
ever physiological factors may be related to a problem as, for
example, with seizures, depression, fatigue, or headaches, a
physical examination should be required. Overlooking physi-
cal causes including nutritional deficiencies and coffee, alco-
hol, or drug intake may result in incorrect inferences.
Reviewing use of prescribed medication is ever more important
in our overmedicated society.
Cognitive-Intellectual Characteristics
People differ in their intellectual abilities which may influence
problems and potential for attaining hoped-for outcomes.
Genetic differences have been found in intelligence as well as
in temperament and conditioning susceptibility. The impor-
tance of assessing what people say to themselves in relevant
situations is emphasized in many assessment frameworks. For
example, in cognitive-behavioral approaches, clients' internal
dialogs (what they say to themselves) and the way this relates to
complaints and desired outcomes is explored and altered as
necessary. Certain thoughts may occur too much, too seldom,
or at the wrong time. A depressed client may have a high
frequency of negative self-statements and a low frequency of
positive self-statements. In a radical behavioral perspective,
thoughts are viewed as covert behaviors to be explained, not
as explanations for other behaviors, although it is assumed they
can serve a mediating function and influence both feelings and
behaviors. The thoughts and feelings in a situation are assumed
to be a function of the contingencies experienced in a particular
situation or in situations that are similar or associated in some
way. A causal role may be misattributed to thoughts because the
histories related to the development of thoughts is overlooked.
The role of thoughts can be examined by varying certain
thoughts and determining the effects on behavior.
Feelings
When feelings are presented as a problem or are related to
a problem, associated personal and environmental factors
should be identified. Assessment frameworks differ in the
role attributed to feelings and in factors sought to account for
them. Some emphasize the role of thoughts in creating feel-
ings. Others emphasize the role of unconscious conflicts and
motives related to early childhood experiences. Other frame-
works focus on the role of environmental contingencies in
influencing emotional reactions. For example, in a radical
behavioral approach, feelings are viewed as by-products of
the relationships between behavior and environmental events.
Feelings can be used as clues to contingencies (relationships
between behavior and environmental events). Changing feel-
ings will not make up for a lack of required skills, or rearrange
contingencies required to attain desired outcomes.
Cultural Differences
Cultural differences may affect both the concern that clients
experience as well as the communication styles and assessment
and intervention methods that will be successful. An indivi-
dualized assessment requires attention to cultural differences
that may be related to problems and potential resolutions.
Culturally sensitive practice requires knowledge of the values
of different groups and their historical experience and how
these differences may influence behavior, motivation and
view of the helping process. Different groups may prefer differ-
ent problem-solving styles and have different beliefs about the
causes of problems. The norms for behavior vary in different
groups. It is important to be knowledgeable about cultural
differences that may be mistakenly viewed as pathology. The
degree of acculturation (the process of adaptation to a new or
different culture) is important to assess. This may influence
drop-out rate, level of stress, attitude toward clinicians, and the
process and goals that are appropriate. Knowledge of chal-
lenges faced and preferred communication styles of people in
different generations will be useful.
Developmental Considerations
Information about required behaviors at different ages and life
transitions can be helpful in assessment. Knowledge of typical
behavior at different times (developmental norms) can be
useful in 'normalizing' behavior - helping clients to realize
that reactions they view as unusual or 'abnormal' are in fact
common. Knowledge about typical changes in different phases
of life (e.g., adolescence, parenthood, retirement) allows pre-
ventative planning. The following kinds of information will be
helpful (1) norms for behavior in specific contexts; (2) tasks
associated with certain life situations such as parenthood and
retirement; (3) the hierarchical nature of some developmental
506 Clinical Assessment
tasks (some behaviors must be learned before others can be
acquired). Different kinds of norms may be used in the selec-
tion of outcomes. Criterion referenced norms rely on what has
been found to be required to attain a certain outcome through
empirical analysis. Another kind of norm is what is usual in a
situation. However, what is usual may not be what is desirable.
For example, although it may be typical for teachers to offer
low rates of positive feedback to students in their classroom, it
is not optimal. The similarities of contingencies for many
people at a given age in a society may lead one to assume
incorrectly that biological development is responsible. The
role of similar contingencies may be overlooked. Acceptance
of a stage theory of development may get in the way of identi-
fying environmental factors that can be rearranged.
Resources and Obstacles
Assessment involves identification of personal assets and
environmental resources that can be used to help clients
attain desired outcomes, as well as personal and environmen-
tal obstacles. Personal resources and/or obstacles include
cognitive-intellectual abilities and deficiencies, physical abil-
ities and handicaps, social skills and social-skill deficits, voca-
tional and recreational skills, financial assets, and social
support systems. Clients differ in their 'reinforcer profile' and
in degree of motivation to alter problematic circumstances.
Environments differ in opportunities for certain kinds of
experiences (see section 'Physical Environment'). Resources
such as money, housing, vocational education programs, med-
ical care, or recreational facilities may be unavailable. Limited
community resources (such as day care programs, high-quality
educational programs, parent training programs) and limited
influence over environmental circumstances may pose an
obstacle. Child maltreatment is related to poverty. Unemploy-
ment is related to substance abuse and spouse violence. Agency
policies and practices influence options. Lack of coordination
of services may limit access to resources; clients may receive
fragmentary, overlapping, or incompatible services.
Sources of Information
Sources of data include interviews, reports on standardized
measures, data gathered by clients and significant others (self-
monitoring), observation in the interview as well as in role
play or in real-life settings, and physiological indicators. A
variety of electromechanical aids are available for collecting
data such as wrist counters, timers, biofeedback devices, and
audio- and videotape recorders. Preferred theories influence
selection of assessment methods. In approaches that focus on
interiors (e.g., thoughts), self-report within the interview may
be the main source of data used. The broader the scope of
change efforts, for example school (or district) -wide positive
behavior support systems, the wider the range of sources and
kinds of assessment data needed.
In behavioral approaches, self-report is supplemented
whenever possible by other sources of data such as observation
in real-life settings, role play, and/or self-monitoring. (Clients
keep track of some behaviors, thoughts, or feelings and sur-
rounding circumstances and consequences in real life.) Some
sources, such as self-report in the interview, are easy to use and
are flexible in the range of content provided. However, accu-
racy varies considerably. The question is: What methods will
offer an accurate description of reactions or conditions of
concern and related events? Individual differences will influ-
ence a client's willingness to participate in a given manner.
Accuracy of decisions can be improved by using multiple
methods, drawing especially on those most likely to offer
accurate relevant data.
Self-report is the most widely used source of information.
There are many different types of self-report including verbal
reports during interviews and answers on written inventories.
Interviews also provide an opportunity to observe clients.
Advantages of self-report include ease of collecting material
and flexibility in the range of material that may be gathered.
Structured interviews have been developed for both children
and adults in a number of areas. The accuracy of self-report
depends on a number of factors including the situation in
which data are collected and the kinds and sequence of ques-
tions asked. Helpful questions in assessing the accuracy of self-
reports include the following (1) Does the situation encourage
an honest answer? (2) Does the client have access to the
information? (3) Can the client comprehend the question?
(4) Does the client have the verbal skills required to answer
questions? Special knowledge and skills may be required when
interviewing children. Play materials and storytelling may be
used to gather data about children's feelings and experiences.
Measures that have uniform procedures for administra-
tion and scoring and that are accompanied by certain kinds
of information, such as norms for different groups, are referred
to as standardized measures. Thousands of standardized
questionnaires have been developed related to hundreds
of different personal and/or environmental characteristics.
Standardized measures are used for a variety of purposes
including (1) describing populations or clients, (2) screening
clients (e.g., making a decision about the need for further
assessment or finding out if a client is eligible for or likely to
require a service), (3) assessing clients (a more detailed review
resulting in decisions about diagnosis or selection of interven-
tion methods), (4) monitoring (evaluating progress), and
(5) making predictions about the likely futures of clients
(e.g., in relation to use of a particular intervention method).
As always, a key concern is validity. Does a measure assess
what it is presumed to assess? Reliability should also be con-
sidered. How stable are responses on a measure given a lack
of real change? Unstable measures are not likely to be valid.
How sensitive will a measure be to change?
Personality tests may be used to collect assessment data.
Objective tests include specific questions, statements, or con-
cepts. Clients respond with direct answers, choices, or ratings.
Projective tests such as the Thematic Apperception Test, incom-
plete sentences test, and the Rorschach Inkblot Test are pur-
posefully vague and ambiguous. It is assumed that each person
will impose on this unstructured stimulus presentation unique
meanings that reflect his or her perceptions of the world and
responses to it. Psychoanalytic concepts underlie the use of
most projective tests. These tests focus on assessing general
personality characteristics and uncovering unconscious pro-
cesses. Tests are used not as samples of the content domain
(as in behavioral approaches), but as signs of important
Clinical Assessment 507
underlying constructs. Whereas content validity is of great
concern in a behavioral perspective, this is not so within a
traditional approach. In fact, items may be made deliberately
obscure and vague.
Valuable information can be obtained from data clients
collect (self-monitoring). As with any other source of data,
not all clients will be able or willing to participate. Observation
of relevant interactions in real-life settings offers a valuable
source of information. This is routinely used in applied behav-
ior analysis. If observation in real-life settings is not possible,
observation in role plays may provide a useful alternative.
Physiological measures have been used with a broad array of
presenting problems. Measures include heart rate, blood pres-
sure, respiration rate, skin conductance, muscle tension, and
urine analysis. Physiological measures are useful when verbal
reports may be inaccurate. Certain kinds of desynchronies
between verbal reports of fear and physiological measures
may provide useful assessment data. Whenever presenting pro-
blems may be related to physical causes, a physical examina-
tion should be obtained. Failure to do so may result in
overlooking physical causes.
Assessing the Value of Data
Assessment methods differ in their accuracy. For example,
self-report of clients or significant others may not accurately
reflect what occurs in real life. Observers may be biased and
offer inaccurate data. General predictions about a person based
on tiny samples of behavior in one context are not likely to
be accurate, especially when behaviors of interest occur in
quite different situations. Measurement inevitably involves
error. One cause of systematic error is social desirability; peo-
ple present themselves in a good light. Important criteria
to consider in judging the value of data include (1) reliability,
(2) validity, (3) sensitivity, (4) utility, (5) feasibility, and
(6) relevance. Reliability refers to the consistency of results
(in the absence of real change) provided by the same person
at different times (time-based reliability), by two different
raters of the same events as in interrater reliability, or by
parallel forms or split-halves of a measure (item-bound reli-
ability). Reliability places an upward boundary on validity. For
example, if responses on a questionnaire vary from time to
time (in the absence of real change), it will not be possible to
use results of a measure to predict what a person will do in the
future.
Validity concerns the question: Does the measure reflect the
characteristic it is supposed to measure? For example, does
behavior in a role play correspond to what a client does in
similar real-life situations? Assessment is more likely to be
informative if valid methods are used (methods that have
been found to offer accurate information). Direct (e.g., observ-
ing teacher student interaction) in contrast to indirect measures
(e.g., asking a student to complete a questionnaire assumed
to offer information about classroom behavior) are typically
more valid. Validity (accuracy) is a concern in all assessment
frameworks; however, the nature of the concern is different in
sign and sample approaches. In a sign approach, behavior is
used as a sign of some entity (such as a personality trait) that
is at a different level. The concern is with vertical validity. Is the
sign an accurate indicator of the underlying trait? Horizontal
validity is of concern in a sample approach. Different levels
(e.g., behavior and personality dispositions) are not involved.
Examples include (1) Does self-report provide an accurate
account of behavior and related circumstances? (2) Does
behavior in role play reflect what occurs in real life? Different
responses (overt, cognitive, and physiological) may or may not
be related to an event. For example, clients may report anxiety
but show no physiological signs of anxiety. This does not mean
that their reports are not accurate. For those individuals, the
experience of anxiety may be cognitive rather than physical.
The sensitivity of measures is important to consider; will a
measure reflect changes that occur? The utility of a measure is
determined by its cost (time, effort, expense) balanced against
information provided. Feasibility is related to utility. Some
measures are not feasible to gather. Utility may be compro-
mised by the absence of empirically derived norms for a mea-
sure. Norms offer information about the typical (or average)
performance of a group of individuals and allow comparison
of data obtained from a client with similar clients. The more
representative the sample is to the client, the greater the utility
of a measure in relation to a client. Relevance should also be
considered. Is a measure relevant to presenting problems and
related outcomes? Do clients and significant others perceive it
as relevant?
The Social Context of Assessment
Clinical assessment takes place in the context of a helper-client
relationship. The nature of this relationship (the alliance) is
considered important in all practice frameworks, including
providing new conceptualizations and support that decreases
demoralization and avoiding negative reactions that damage
this relationship, such as excessive negativity (e.g., criticism).
Influence of the clinician on the client has been found even in
very nondirective approaches. The interpersonal skills of the
clinician are viewed as essential for facilitating a collaborative
working relationship, validating and supporting the client, and
encouraging clients to participate in intervention programs.
Great attention is given to the diagnostic value of transference
and counter transference effects in psychodynamic therapies
and the relationship itself is viewed as the primary vehicle of
change. Traditionally, transference has been viewed as a reen-
actment between the client and the counselor of the client's
relationship with significant others in the past, especially par-
ents. Counter transference effects refer to feelings on the part of
helpers toward their clients. Transferences are distinguished
from therapeutic or working alliances within psychodynamic
perspectives.
There is a continuing need throughout assessment to ex-
plain the roles and requirements of the client and the coun-
selor, the process that will occur, and the rationale for this.
Introductory explanations include an overview of mutual
responsibilities and of the framework that will be employed.
Because different client behaviors may be required during dif-
ferent phases of assessment and intervention, this 'socializa-
tion' is an ongoing task. Behavioral clinicians tend to be more
directive than psychoanalytically oriented clinicians. They
more frequently give instructions, provide information, and
508 Clinical Assessment
direct conversation. Clinicians may err by being too directive
or too nondirective. Overly directive clinicians may not recog-
nize the need to help clients to explore and to understand their
behavior. In contrast, nondirective counselors may err by
assuming that self-understanding is sufficient to achieve desired
outcomes (when it is not).
Common Assessment Errors and Their Sources
Errors in clinical reasoning account for many errors in clinical
practice. Errors may occur in any phase of assessment: (1)
detection of characteristics of the client and his or her life
situation that are related to problems and desired outcomes;
(2) integration and interpretation of data gathered; and (3)
selection of outcomes to pursue in order to remove com-
plaints. Errors made in the first two steps will result in errors
in the third step. Examples of common errors are noted below.
They result in incomplete or misleading assessment. Unexam-
ined speculation may result in misleading accounts. Indoctri-
nation into a certain culture of helping (e.g., biomedical) may
contribute to failure to search for well-argued alternative
accounts and over reliance on one form of data collection.
Not keeping up to date with research regarding options and
controversies maintains questionable beliefs. Some errors
involve or result in inappropriate speculation - assuming that
what is - can be discovered simply by thinking about the topic.
Failure to keep up to date with research describing the eviden-
tiary status of different kinds and sources of data as well as
factors related to different presenting concerns is a key source
of error in clinical assessment (see prior discussion of evidence-
informed assessment). Here are examples of common errors:
• Selecting vague outcomes.
• Premature closure - Hasty assumptions about causes
(failure to search for alternative accounts).
• Speculating when data collection is called for.
• Searching only for data that confirms a preferred view
(confirmation biases).
• Confusing the form and function of behavior.
• Illusory correlation (confusing correlation and causation).
• Using misleading and/or uninformative labels (confusing
naming and explaining).
• Confusing motivational and behavior deficits.
• Ignoring baserate data.
• Focusing on pathology and overlooking assets.
• Misuse of resemblance criteria.
• Collecting irrelevant material.
• Relying on inaccurate sources (e.g., anecdotal experience).
• Being misled by superficial resemblances of a client to other
clients in the past or to a stereotype.
• The fundamental attribution error (attributing behavior to
personality characteristics of individuals and overlooking
environmental causes).
Errors in detection of relevant events include an overly
narrow focus (e.g., confining attention to the client's
thoughts), overly narrow use of data collection methods (e.g.,
reliance on the interview alone), and errors during data collec-
tion (e.g., observer bias). Inaccurate or incomplete accounts
of client concerns and related factors often occur because
attention is narrowly focused on one source of informant
(e.g., thoughts or feelings). There may be no contextual assess-
ment in which environmental factors related to desired out-
comes are identified. Sources of error in integrating and
interpreting data include focusing on consistency rather than
informativeness of data, hasty generalization based on limited
samples, and inadequate conceptualization of problems due
to theoretical biases (e.g., focus only on environmental factors)
or inaccurate of superficial knowledge of a practice framework.
Another source of error at this stage is the use of vague language
and psychological jargon that is not informative. Errors in
selection of outcomes focused on may occur because of errors
in the first two phases. Client ethnicity, race, and characteristics
(such as obesity) influence helper's views of concerns.
Studies of clinical decision-making indicate that decisions
are made on the basis of quite limited data. Even though a great
deal of data are gathered, only a small subset is used. Clinicians
tend to gather more data than are needed and, as the amount
of data gathered increases, so does confidence in its usefulness,
even though accuracy may not increase. Clinicians have a
tendency to confuse consistency of data with informative
value. Irrelevant as well as relevant data may be influential.
Clinicians, like other individuals, are affected by limited
information-processing capacities and motivational factors.
As a consequence, they do not see all there is to see. Because
of preconceptions and biases, things that are not actually pres-
ent may be reported and events that do occur may be over-
looked. Confirmation biases are common. That is, data are
sought that are consistent with preferred theories and precon-
ceptions, and contradictory data tend to be disregarded.
Many factors that are not correlated with the true frequency
of an event influence estimates of its frequency and how
important it seems (such as how visible it is, and how vivid it
is). It is easy to recall bizarre behavior and pay excessive atten-
tion to this, ignoring less vivid appropriate behavior. The fre-
quency of data that are available is overestimated. Clinicians in
a given settings are exposed to particular kinds of clients, which
may predispose them to make certain assumptions. For exam-
ple, a psychologist who sees many severely depressed indivi-
duals may be primed to attend to signs of depression. Base rate
data that are abstract tend to be ignored, which increases the
probability of inaccurate inferences. A lack of concern for
sample size and sample bias can lead to incorrect judgments.
General predictions about a person that are based on tiny
samples of behavior in one context are not likely to be accurate,
especially when behaviors of interest occur in quite different
situations. Not distinguishing between description and infer-
ence may result in incorrect assumptions. Use of multiple
sources of data within a contextual practice framework pro-
vides the greatest opportunity for sound clinical assessment.
Ethical Issues and Future Directions
Evidence-informed assessment encourages use of frameworks
and methods that have survived critical appraisal in relation to
their accuracy. Taking advantage of the steps involved in
evidence-based practice, for example, consulting practice-
related research, should contribute to making well-informed
decisions regarding questions that arise. Lack of up-to-date
Clinical Assessment 509
knowledge and related skills regarding valid assessment meth-
ods may result in the selection of ineffective and/or harmful
interventions. It is thus incumbent on clinicians to draw on
empirically sound methods that are useful in discovering the
potential to attain valued outcomes and in selecting effective
plans. Helpers and clients have preferred views about behavior
and how it develops, is maintained, and can be altered. These
views differ in the extent to which they have been critically
appraised regarding their accuracy and usefulness in providing
guidelines about how to attain valued outcomes or if they can
be attained.
There are great stakes in how problems are framed and
considerable resources are devoted to influencing how people
think about personal and social problems. Many problems
once viewed as sins were then seen as crimes and are now
considered to be 'mental illnesses.' Beliefs about the causes of
behavior influence how people are viewed. In past years,
pathology was often attributed to housewives who wanted to
work. Knowledge about social, political, and economic factors
that influence the very definition of personal and social pro-
blems (their social construction) is vital so that clinicians con-
sider concerns in their social context. This also decreases the
likelihood of pathologizing clients (medicalizing problems in
living). Honoring ethical obligations to clients requires selec-
tion of assessment frameworks and related methods that are
compatible with empirical data regarding behavior. Accurate
descriptions of different assessment perspectives increases the
likelihood that points of convergence and differences among
varied views are correctly identified. Continuing research efforts
are needed to identify valid methods and indicate assessment
frameworks that are most likely to help clients achieve hoped-
for outcomes. Increased interest in clinical reasoning and
related errors and how to avoid them bodes well for enhanced
quality of clinical decisions regarding assessment.
See also: Cognitive Behavior Therapy; Defense Mechanisms; Social
Support.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental
Disorders, 4th edn.. Text revision. Washington, DC: American Psychiatric
Association.
Gambrill E (2005) Critical Thinking in Clinical Practice: Improving the Quality of
Judgments and Decisions, 2nd edn. Hoboken, NJ: Wiley.
Hersen M (ed.) (2006) Clinician’s Handbook of Behavioral Assessment. New York:
Elsevier.
Hunsley J and Mash EJ (2008) A Guide to Assessments that Work. New York: Oxford.
Lilienfeld SO, Lynn SJ, and Lohr JM (2003) Science and Pseudoscience in Clinical
Psychology. New York: Guilford.
Luiselli JK (ed.) (2006) Antecedent Assessment and Intervention: Supporting Children
and Adults with Developmental Disabilities in Community Settings. Baltimore, MD:
Paul H. Brooks.
Clinical Psychology: An Information Processing Approach
P Fisher, University of Liverpool, Liverpool, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Attentional bias Differential allocation of attention to
self-relevant threatening stimuli relative to neutral
stimuli.
Automatic processing Although no absolute consensus
exists on the meaning of automatic processing, it generally
refers to a cognitive process that operates unintentionally,
without conscious awareness, is effortless, and requires
no resources.
Cognitive architecture The organization of mental
subsystems involved in the performance of information
processing tasks.
Metacognition Cognition about cognition; any knowledge
or cognitive process that is involved in guiding, monitoring,
controlling, or appraising cognition.
Strategic processing In contrast to automatic processing,
strategic or controlled processing is a conscious, effortful,
volitional process which requires cognitive resources.
Introduction
Over a century ago, Lightner Witmer established the first psy-
chological clinic and formally proposed and outlined a struc-
ture of clinical psychology. Subsequently, clinical psychology
has blossomed as a discipline and undergone many paradig-
matic transitions; but all have deigned to improve our under-
standing of human behavior and to improve the efficacy of
psychological interventions. The information processing (IP)
approach has become one of preeminent frameworks in clini-
cal psychology, particularly in the United Kingdom and the
United States, in the twenty-first century. The IP approach has
been extensively utilized in an attempt to account for the core
features of a broad range of clinical disorders in all subdivi-
sions of clinical psychology. More importantly, IP models have
moved beyond descriptive phenomenology and are illustrating
how particular types of attentional, interpretative, and memory
biases are involved in the etiology and maintenance of psycho-
logical disorders, ranging from anxiety disorders in children
and adults to IP models of autism. This development has raised
the usefulness of the IP approach to a new level, namely that
theoretical expositions are translated into psychological inter-
ventions, either in an additive fashion to existing treatments or
as innovative and novel approaches.
This article begins with a brief overview of the nature and
development of the IP paradigm within cognitive psychology.
It should be noted that within the rubric of cognitive psychol-
ogy, IP can be considered from many perspectives, which
accounts for the continued enthusiastic exploration of the IP
approach from both theoretical and practical positions. IP can
be studied from multiple perspectives including neurobiologi-
cal and neuropsychological perspective. For example, under-
standing how lesions in specific brain regions map onto
difficulties in learning or how specific neuropsychological def-
icits are associated with theory of mind difficulties in autism
highlights the wide applicability of the approach. In this arti-
cle, we will focus on how IP approach applies to emotional
disorders. Following the overview of the IP paradigm, the value
of the IP paradigm to clinical psychology is then presented,
coupled with a discussion of the factors leading to the paradig-
matic shift from behaviorism to cognitivism. The next section
discusses some of the experimental methods, research findings,
and clinical developments within the IP paradigm using atten-
tional biases in anxiety as the main exemplar. The final section
of this article focuses on two innovative clinical developments
for anxiety disorders: cognitive bias modification procedures
and metacognitive therapy. Both are grounded within the IP
framework, but theoretically are very distinct.
Overview of the Information Processing Paradigm
Empirical investigation within cognitive psychology includes,
but is not limited to, the study of memory, attention, language,
perception, higher level thought, reasoning consciousness, and
cognitive development. The IP approach was developed as a
framework within cognitive psychology to explain human
behavior by reference to a cognitive system that mediates
between incoming stimuli and behavior. The most familiar
metaphor to explain the IP approach is the individual as a
computer, and indeed the development of the IP approach is
partially attributable to developments in computer technology
during and following World War II. Specific guiding principles
of the IP approach in cognitive psychology were outlined in the
late 1970s. The first assumption was that people are autono-
mous individuals who intentionally engage with the other
people and the environment. Next was the idea that the cogni-
tive processor is a general-purpose, symbol-processing system
involving both process, such as the allocation of attention, and
content, such as an image in memory. Of central importance to
the IP paradigm is the assumption that the cognitive system
has a limited capacity. This proposition is crucial as it means
there will be competition for resources leading to selective
attention. For example, every day we are bombarded with
stimuli and only a finite proportion can be selected and pro-
cessed; how does this occur? Selective attention has engaged
cognitive psychologists for over three decades and debate has
centered on the extent to which selective attention is an auto-
matic or strategic process. Although, a general agreement now
exists that attentional bias is a combination of automatic and
controlled processing, considerable theoretical disagreement
continues as to the relative importance of each component in
510
Clinical Psychology: An Information Processing Approach 51 1
psychopathology. More importantly, the adopted theoretical
positions on automaticity have had a profound influence on
research directions and clinical practice. This issue will be
discussed in more detail later.
A further fundamental assumption of the IP paradigm is
that IP takes time and that the time taken for a particular IP task
can be specified. This allows predictions and testable hypoth-
eses to be derived. For, example, attendonal bias is inferred
from differential response times to competing stimuli in a
range of experimental tasks. The IP paradigm also specifies
that cognitive processes are dependent upon, and constrained
by, neuroanatomy and neurochemistry. Exploration of the
interface between neuroscience and IP is an emerging field,
which is providing valuable insights for the practice of clinical
psychology. For example, neuroimaging assessment before and
after psychological interventions is generating important data
on the neurobiological mediators of change.
Evolution of the IP Paradigm Within Clinical Psychology
Clinical psychologists working in adult mental health embraced
the IP paradigm during the 1970s for two main reasons. First,
there was growing dissatisfaction with the prevailing clinical
paradigm of behaviorism. Although behavioral therapy was,
and continues to be, an effective treatment for many clinical
problems, the underpinning theory had insufficient explana-
tory power. Classical and operant conditioning processes were
unable to explain fundamental aspects of specific disorders or
the psychological mechanisms effecting clinical change. Interest-
ingly, contemporary theories of the mechanisms of change in
behavior therapy explicitly focus on cognitive constructs and
processes, rather than on the two-factor model that had domi-
nated theorizing from the 1950s to the 1970s.
The second key influence was the development of schema
theory proposed by Aaron Beck. This clinical theory placed
schemas at the heart of emotional disorders. The construct of
schemas was appropriated from work conducted on early
research on memory processes. Schemas were defined as rela-
tively stable and enduring cognitive structures that influence
how an individual perceives and interprets information.
The content of the schemata is thought to be of paramount
importance and differentiates between anxiety and depressive
disorders. Specifically, depression or anxiety will occur when
a matching life event activates congruent knowledge stored
in the long-term memory. Typically, depressogenic schemas
incorporate negative beliefs about the world, the future, and
themselves as being inadequate or worthless. In anxiety disor-
ders, the content of schemas concerns personal vulnerability,
uncertainty, and danger. It is hypothesized that schema activa-
tion results in biased IP, most readily seen by an increased
presence of negative automatic thoughts and interpretative
biases, more commonly referred to as cognitive distortions.
Examples of cognitive distortions include selectively attending
to one component of an event and ignoring other relevant
aspects, or thinking in all or nothing terms.
The original exposition of schema theory was not explicitly
developed within the IP paradigm; the theory clearly impli-
cates many cognitive processes including memory, attention,
and perception. Furthermore, the theory states that schemas
guide processing of incoming stimuli and influence percep-
tion and understanding, which in turn is translated in how
people behave. However, neither the original exposition nor
subsequent IP accounts of schema theory have provided a
cogent model of how schemas guide IP and result in persis-
tent emotional distress. It remains to be explained why a declar-
ative belief such as 'the world is a threatening place' should
result in heightened levels of anxiety and persistent worry.
Schema theory remains a hugely influential and useful clinical
theory, but has a number of conceptual limitations when con-
sidered from an explicit IP perspective. Notwithstanding this
issue, the undoubted success of cognitive therapy continues
to stimulate considerable research interest in the cognitive
processes involved in the etiology and maintenance of clini-
cal disorders. Cognitive behavior therapy (CBT) has become
the predominant psychological treatment approach of the
twenty-first century, having been extensively evaluated in over
400 randomized controlled trials and the breadth of applica-
tion continues to grow. It is widely recognized that improve-
ment in the efficacy of CBT approaches for specific disorders,
such as social phobia, is a direct result of integrating schema
and IP theories. Therefore, the IP paradigm will continue to
flourish as it is assumed that further unraveling of cognitive
processes, such as attendonal biases in anxiety disorders, will
enable increasingly effective interventions to be developed.
Significance of the IP Paradigm to Clinical Psychology
The value of any paradigm resides in its ability to generate
theories and well-specified models; the IP paradigm has
excelled in this regard. There are a multitude of theories and
associated models derived from the core assumptions of
the IP paradigm across all subspecialities of clinical psychol-
ogy. Within adult mental health, a multitude of generic and
disorder-specific models exist, some drawing very specifically
on the IP paradigm and others integrating components drawn
from the IP paradigm with other schools of thought. Once
models have been developed and empirically evaluated, the
next step is to translate the models into effective psychological
interventions and again work conducted under the IP para-
digm has been extremely successful and has produced many
highly differentiated treatments. Both of these core strengths
of the IP paradigm can be illustrated with reference to post-
traumatic stress disorder (PTSD). The central premise of many
models of PTSD is that memory processes operate differently to
normal memories as a consequence of the traumatic conditions
in which the memories were encoded. For example, the dual
representation theory proposes that different components of
trauma memories are stored in different memory systems. Spe-
cifically, features of the traumatic experiences that are attended
to consciously are stored in the verbally accessible memory
(VAM) system and can be recalled easily. The nature of the
beliefs in this system is hypothesized to be dysfunctional and
represents common beliefs expressed by trauma patients,
such as 'I am to blame, it was my fault.' In contrast, there are
other aspects of the traumatic experience which were not
consciously attended to, but were encoded in a situationally
accessible memory (SAM). These memories are only recalled
when triggered by a trauma cue and result in increased
512 Clinical Psychology: An Information Processing Approach
physiological reactivity, and reexperiencing symptoms. The
model proposes that psychological treatment must correct
information stored in both memory systems; therefore, cogni-
tive restructuring methods are used to correct maladaptive
beliefs, whereas exposure strategies are required to modify the
physiological components of PTSD.
In contrast to the models of PTSD, which specify a critical
role for memory disorganization, the metacognitive model of
PTSD proposes that metacognitive beliefs and processes are
more important in maintaining PTSD than specific aspects of
memory such as encoding and retrieval. Indeed, the metacog-
nitive model argues that maintaining factors in PTSD relates
to a particular processing style called the cognitive atten-
tional syndrome (CAS). The CAS consists of high levels of
perseverative thinking, threat monitoring, and coping beha-
viors that are guided by metacognitive beliefs and procedur-
alized knowledge that become active following the trauma
experience and subsequent symptoms. Patients engage in
high levels of worry because they hold positive metacognitive
beliefs about the usefulness of worry as a coping strategy and
also because they have negative metacognitive beliefs about
the uncontrollable nature of worry. In relation to memory, an
example of positive metacognitive belief would be 'I must be
remember all aspects of the trauma in order to move on,'
whereas negative metacognitive beliefs focus on the dangers
and consequences of not being able to recall every aspect of
the event. These two categories of metacognitive belief com-
bine to keep the person focused on the threat and engage in
continued processing of the trauma experience.
Treatment would focus on the modification of positive and
negative metacognitive beliefs about the CAS and would not
involve cognitive restructuring of beliefs in the cognitive
domain or use exposure exercises. These two examples illus-
trate the development and diversity of disorder-specific models
that have been developed under the broad framework of the
IP paradigm. Moreover, the two models highlight that the resul-
tant treatments are distinct in how individuals with PTSD are
assessed, conceptualized, and treated.
The development of models within an IP framework has
directly influenced diagnostic classificatory systems such as
the International Classification of Disease (ICD) and Diagnos-
tic and Statistical Manual (DSM). Since generalized anxiety
disorder (GAD) entered the diagnostic taxonomy in 1980, it
has undergone substantive changes. Originally, GAD was con-
ceptualized primarily as a disorder of anxiety but with very
limited reference to cognitive processes. In the latest incarna-
tion of the Diagnostic and Statistical Manual (DSM-IV), the
cardinal diagnostic feature of GAD is a cognitive process,
namely excessive worry, which a person finds difficult to con-
trol. Such diagnostic changes are very significant as they pro-
vide both an impetus and direction for research. In the case
of GAD, research is ever more focused on understanding the
psychological processes which underpin worry. Panic disorder
represents another clear illustration of how diagnoses for anxi-
ety disorders have been shaped by research conducted under
the IP rubric. Panic disorder is defined according to the psy-
chological mechanisms responsible for the maintenance of the
disorder. Patients with panic disorder catastrophically misin-
terpret symptoms of anxiety which is both the primary target
for intervention and the cardinal feature of the disorder.
Evidently, clinical psychology has benefited from research
conducted within the IP paradigm in many ways. In addition
to the taxonomic, treatment, and model development, the IP
framework continues to broaden its horizons as it is being
applied to an increasing number of clinical problems. Further-
more, the IP approach is contributing to our understanding
of vulnerability to relapse following treatment or indeed the
genesis of clinical disorders as can be seen in the child psy-
chopathology literature where research is providing illumi-
nating data on potential pathways to emotional difficulties.
Finally, evidence is steadily being accumulated that models
and treatments located within an IP paradigm are producing
increasingly more effective and time-efficient psychological
treatments.
Emotional Disorders and Information
Processing Biases
A wealth of studies using a range of experimental tasks have
demonstrated that people with elevated levels of anxiety dem-
onstrate an attentional bias toward threatening stimuli com-
pared to neutral stimuli. Such results are in accord with clinical
accounts in which patients describe focusing their attention
on signs of threats, for example internal sources such as
unwanted thoughts and feelings or external signs of possible
threat, typified by hypervigilence for signs of danger in PTSD
patients. However, there may be both qualitative and quantita-
tive difference in attentional allocation in the natural environ-
ment compared to laboratory experiments; this conundrum
remains to be resolved. Nevertheless, research into biased IP in
emotional disorders has made substantial advances over the past
two decades. Early work in this area took a rather dichotomized
approach to explain the IP biases observed in emotional disor-
ders. For example, the majority of research found that anxiety
disorders were characterized by selective attention toward threat-
ening stimuli whereas affective disorders did not show this bias.
Alternatively, depression and not anxiety was characterized by
explicit memory biases toward self-relevant negative informa-
tion. However, as the experimental paradigms and theorizing
increased in sophistication, it has become evident that both
anxiety and depression are characterized by a range of IP biases.
Recent studies on depressed adolescents and on adults experien-
cing suicidal ideation have shown preferential attentional bias
for self-relevant stimuli. A general finding is that the degree to
which threatening stimuli are perceived as self-relevant is corre-
lated with the magnitude of attentional bias. In PTSD, the inter-
ference effect observed on the emotional Stroop task is greater
when the stimuli are most self-relevant; so, in combat veterans
with PTSD, the interference effect will be greater for military
stimuli than general trauma stimuli. A similar picture is seen in
obsessive compulsive disorder, whereby biases are only observed
when the stimuli are relevant to the individual specific obses-
sional concerns.
Components of Attentional Bias
Attentional bias in anxiety comprise at least three components,
which change according to the stage of IP. The first stage is
facilitated attention, in which threatening stimuli are detected
Clinical Psychology: An Information Processing Approach 513
more rapidly than neutral or less threatening stimuli. The next
stage is difficultly in disengaging from threat; in other words,
anxious people find it difficult to shift their attention away
from threatening stimuli; and the final stage is attentional
avoidance, where people allocate their attention away from
threatening stimuli. A broad range of experimental tasks
(described below) have repeatedly demonstrated attentional
bias at each stage of processing; however, the mediating mech-
anism(s) underpinning each component remains a source of
controversy. Understanding the mechanisms responsible for
attentional bias would facilitate the development of theoreti-
cally cogent and defined interventions.
Experimental Paradigms
The emotional Stroop was the first task used extensively to
measure attentional bias in anxiety. In this task, participants
are presented with a series of threatening and neutral words
printed in different colors and the objective is to name the
color that each word is printed in as quickly as possible whilst
ignoring the semantic content of the word. Response times
are measured and differences in latencies of color-naming
between threat words and neutral words are indicative of
attentional bias. Across a range of anxiety disorders, studies
using the emotional Stroop have found increased latencies of
color-naming of threat words. The most frequent conclusion is
that slower color naming is attributable to an automatic pro-
cessing bias in anxiety which directs attention toward threat-
ening material. However, the emotional Stroop has been the
center of considerable debate for many years as numerous
authors have argued that the mediating mechanism underpin-
ning is not rapid orientation or engagement with threatening
stimuli but instead primarily reflects a later stage of pro-
cessing, namely difficulty engaging with threatening material.
Researchers have also suggested that attempts by participants
to suspend further processing of the threat cues slow color-
naming. Other explanations for the interference effect on the
emotional Stroop relate to the other cognitive and emotional
processes involved in undertaking the task as it is not a pure
measure of attention. It may be that participants process neu-
tral and emotional stimuli to the same extent; however, the
negative stimuli produce an increase in emotion which inhibits
reaction time performance. This brief account of the emotional
Stroop highlights that interpretation of the observed interfer-
ence effect is highly problematic. Overall, evidence suggests
that the interference effect is not attributable to just one com-
ponent of attentional bias, and that the component which
explains most of the effect appears to be difficultly disengaging
from threat, rather than facilitated attention.
An alternative task that has also been widely employed in
assessing attentional bias toward threat and overcomes some
of the limitations of the emotional Stoop is the dot probe task.
This task is also thought to have greater ecological validity as it
does not rely on words as stimuli and tends to use pictures that
have personal meaning for the particular problem being inves-
tigated. For example, if a spider phobic was completing this
task, then the neutral stimuli might be a range of everyday
objects or animals that do not provoke distress, whereas the
threatening picture would be relevant to the phobia. In this
task, two pictures are presented simultaneously on a computer
screen. The pictures then disappear and a probe replaces one
of the pictures. Participants are asked to indicate as rapidly as
possible (by a button press) which stimulus has been replaced
by the probe. Attentional bias is inferred through faster
responses to the probes that replace threatening stimuli relative
to the neutral stimuli. A major advantage of the dot probe
task is that the interval between stimulus presentation and
the probe can be varied, enabling assessment of the time
course of the attentional bias. Studies which have examined
the effects of facilitated attention and difficulties in disengage-
ment overwhelmingly find that the attentional bias effect is
accounted for by delayed disengagement. This does not mean
that a fast automatic orienting bias does not occur, but it does
imply that the rapid orienting response is not contributing
as significantly to the maintenance of anxiety disorders as
early research indicated.
Two other experimental tasks are frequently used to
measure attentional biases. The visual search task allows assess-
ment of spatial allocation of attention and speed of respond-
ing. Participants are asked to search for the target word
amongst distractor words (as in a word search puzzle) . A single
threat word might be embedded in a matrix of neutral words
and the reverse condition would ask participants to search
for the neutral word amongst a set of self-relevant emotional
words. Attentional bias is inferred from faster detection of
threat words embedded in neutral words relative to detecting
a neutral word embedded amongst other neutral words. The
alternative condition is where participants are slower to detect
neutral words in a matrix of threat words compared to
detection of a neutral target word in a matrix of neutral stimuli.
The final task that has been used frequently in attentional bias
research is the spatial cueing task. In this task, the participant
is asked to rapidly identify a target stimuli but they are cued
(either by a neutral or threatening cue) to a spatial location
on the computer screen before the target appears. The cue
may be valid, in that the target appears in the same spatial
location as the cue; alternatively, it may be invalid, in which
case the target stimulus appears in a different spatial location
to the cue. Attentional bias is inferred by faster response times
for threatening cues in the correct spatial location or by slower
response times on invalid trials. Both the visual search task
and the spatial cueing task have produced robust results
demonstrating attentional bias for threatening stimuli in a
range of anxiety disorders, accounted for by difficulties in
disengaging from threat.
As experimental tasks have increased in sophistication
and ability to measure the different components of attentional
bias, the most robust finding is that people with elevated levels
of anxiety have difficulty in disengaging from threatening
stimuli. In other words, anxiety disorders are characterized
by a lack of flexibility of allocation of attention. In the next
section, the issue of whether attention is primarily an auto-
matic or controlled process is discussed in more detail.
Automatic and Controlled Processing
One of the most enduring theoretical disputes within IP
accounts of emotional disorders and particularly in relation
to attentional biases in anxiety is whether the observed bias is
attributable to automatic or controlled processing. Automatic
514 Clinical Psychology: An Information Processing Approach
processing has been defined as uncontrollable, unintentional,
unconscious, capacity free, and effortless, whereas controlled
or strategic processing is an intentional, volitional process,
which requires conscious awareness, effort and is constrained
by capacity limitations. As IP accounts of anxiety were pro-
posed during the course of the 1980s and 1990s, the distinc-
tion between controlled and automatic processing became
highly significant because it determined how psychological
treatments for anxiety were developed. If attentional and inter-
pretative biases are automatic processes, then this implies that
psychological therapies should focus on enabling patients to
cope and manage with anxiety or possibly through extended
retraining, attentional biases could be alleviated. The alterna-
tive view that attentional bias is primarily a product of con-
trolled processing will result in a very different psychological
intervention, as the focus would be on modifying processing
configurations, guided by beliefs/knowledge which is stored in
long-term memory structures.
Examination of the automaticity of attentional biases in
anxiety has moved away from a simple all-or-nothing dichot-
omy to exploring whether the different stages of processing
are related primarily to automatic or controlled processing.
The predominant view was, and continues to be, that facili-
tated attention represents an automatic process whereas the
later stages of attentional bias (delayed disengagement and
avoidance of threat) come increasingly under strategic control.
However, research thus far has not unequivocally demonstrated
that any stage of processing is either completely strategic or
completely automatic. This is despite many studies concluding
that attentional bias is an automatic process but this conclu-
sion is unwarranted as these studies have generalized from an
experimental task that has only manipulated one component
of automaticity. For example, studies using the emotional
Stroop have tried to determine whether the interference effect
is attributable to an automatic process by using masked pre-
sentations. This task involves very brief presentation of the
stimuli (<20ms) and then the stimulus is masked for up to
480 ms before the next stimulus is presented. The idea behind
this is that such rapid presentation prevents conscious aware-
ness and therefore the observed interference effect reflects an
automatic process. However, only degree of awareness has
been manipulated and there is a question mark over the extent
to which subliminal presentations with backward masking
actually preclude conscious awareness. Therefore at best, stud-
ies have demonstrated that attentional biases are partially
automatic, as no studies have successfully manipulated all
facets of automaticity. This raises an important question,
which is whether research focusing on differentiating auto-
matic from controlled processing is meaningful or clinically
useful as the majority of cognitive processes appear to be a
combination of automatic and controlled processing. The
more clinically useful question is determining whether strate-
gic processing can exert an influence over the more automatic
components of attentional bias.
Theories differ in the importance placed on each stage of
processing. Some theories emphasize that anxious people
exhibit an enhanced automatic bias for threat during the
early stages of processing and avoidance of threat in the later
stages of processing, often labeled as the vigilance-avoidance
model. Others have argued that the early stages of processing
and the automatic or rapid orienting response have little
impact on anxiety, instead arguing that anxiety is maintained
by a difficulty in disengaging from threat. As reviewed earlier in
the article, the evidence from various analyses of experimental
tasks would indicate that early components of attention are
less important than later stage of processing in the mainte-
nance of anxiety disorders.
Clinical Applicability of Information Processing
Models and Treatments
The ultimate test of the value of the IP approach is whether
it has resulted in improved treatment efficacy as a result of
translating theoretical advances into clinical treatments. The
answer to this question depends on the treatment and disorder
being investigated. Cognitive therapy for depression remains
virtually unchanged in terms of treatment delivery and treat-
ment efficacy over the past 30 years. Findings from empirical
work on memory biases and attentional processes in depres-
sion have not been incorporated into cognitive therapy. The
opposite is true for psychological treatments of anxiety disor-
ders, where there are clear developments in the focus and
nature of interventions for almost all anxiety disorders.
Cognitive Bias Modification
Following on from research demonstrating that individuals
with anxiety disorders show a preferential attention for threat-
ening stimuli, it was argued that attentional bias was an
integral component of anxiety disorders. Prospective studies
found that facilitated attention bias was predictive of greater
emotional reactivity to a stressful event. Furthermore, that
when nonanxious people were trained to either direct their
attention toward or away from threat, those individuals who
were trained to orient toward threat showed significantly
higher emotional responses to a stressor task. As a result of
such studies, it was hypothesized that people suffering from an
anxiety disorder could be trained to reduce their attentional
bias toward threat. Therefore, cognitive bias modification
(CBM) procedures, designed to decrease attentional bias in
anxiety disorders, have been developed. It has been argued
that CBM procedures may have an important role in the treat-
ment of anxiety disorders, especially as the current level of
efficacy of psychological treatments for many anxiety disor-
ders is somewhat limited. CBM procedures typically involve
a modification of the dot probe task and train people to
orient rapidly away from threatening stimuli. Recent applica-
tions of CBM procedures in clinical samples have shown
promising results including significant reductions in anxiety
and depression in a GAD sample, after a 4-week training
program. However, a number of issues require resolution.
First, these procedures are predicated on the assumption that
the early stage of attentional bias, facilitated attention toward
threat, is being modified. However, it appears that the CBM
procedures are operating at the later stages of attentive proces-
sing of threat, namely avoidance of threat. In other words,
CBM techniques are modifying strategic rather than automatic
processing. A related issue is that if people are being trained
to look away from threat, then this has important clinical
Clinical Psychology: An Information Processing Approach 515
implications. Patients use avoidance to modulate anxiety, but
with an important consequence, they do not learn more
effective methods of responding to threat. A further issue
with the CBM results is that although GAD patients showed
reductions in overall levels of anxiety and depression, the
cardinal feature of GAD, namely worry, remained compara-
tively unchanged. Studies such as these highlight that modify-
ing attentional processes may result in reductions of anxiety,
but anxiety disorders are not only characterized by attentional
biases and other important aspects of the disorder remain
unchanged. Further research is required to delineate the
mechanism of action, especially as it is possible that these
training procedures are actually counterproductive in the long
run. Despite the need to clarify these issues, the recent studies
on CBM raise a number of fascinating possibilities including
whether the strategies can be successfully incorporated in
existing CBT protocols with a resultant increase in treatment
efficacy.
The Self-Regulatory Model of Executive Dysfunction
in Emotional Disorders
One model that has been developed within an IP framework
to explain the persistence of emotional disorders is the
Self-Regulatory Executive Function model (S-REF). This
model was proposed by Adrian Wells and Gerald Matthews
as a transdiagnostic understanding of the factors involved
in the maintenance of psychological disorder. The model
attempted to overcome some of the limitations of schema
theory and develop a model of psychopathology that was
explicitly developed within the IP framework. The model spe-
cifies that psychological disorders should be conceptualized as
predominantly top-down or conscious processes, with a set of
self-regulatory strategies. According to the model, a particular
style of thinking and coping with negative thoughts and feel-
ings results in the exacerbation and maintenance of emotional
distress. This particular style of thinking has been termed the
CAS and consists of perseverative thinking, threat monitoring,
and counterproductive coping strategies. In developing the
model, the authors returned to earlier work on attention,
which specified that there were different levels of control in
the form of automatic and controlled processing. The S-REF
model proposes that emotional distress is linked primarily
to biases in the selection and execution of controlled processes
for appraising and coping with intrusive cognitive and emo-
tional events. More specifically, an individual's strategy for
thinking and self-regulation in response to unwanted thoughts
and feelings can either intensify and maintain negative emo-
tions or result in brief emotional reactions. Emotional disor-
ders such as anxiety and depression occur when the style of
thinking and coping inadvertently leads to persistence and
strengthening of emotional responses. The CAS is guided by
metacognitive beliefs concerning (a) the benefits of engaging
in each aspect of the CAS, for example 'worry helps me cope'
and (b) the uncontrollable nature of the CAS, for example
'my worry is out of my control, I cannot stop myself from
worrying.' In MCT, the CAS and the supporting metacogni-
tive beliefs are the psychological processes that are targeted
in therapy.
In contrast to cognitive therapy and CBM approaches,
metacognitive theory does not link psychological problems to
automatic processing biases or the content of schemas. Instead,
metacognitive theory specifies that psychological problems are
attributable to conscious and volitional strategies. In relation
to the experimental paradigms used to assess attentional bias
earlier in the article, metacognitive theory does not link the
observed effects to activation of schemas or automatic proces-
sing but attributes them to strategy selection. In anxiety dis-
orders, patients have a strategy of maintaining attention on
sources of threat and engaging in worry-based processing as a
means of coping.
The S-REF model is based on three basic levels of cognition:
a level of reflexive and automatic processes that run with
minimal or no conscious involvement. These processes may
generate intrusions into consciousness that capture attention.
The next level is an online form of processing which is con-
scious and capacity-limited, responsible for regulating and
implementing appraisal and action. The final level is stored
knowledge in long-term memory. The immediate activities of
the online processing require access to stored knowledge
in order to run. Online processing is guided by knowledge or
beliefs that are metacognitive in nature. In line with one of the
core values of the IP approach, the S-REF theory has generated
many disorder-specific models, including models of depres-
sion, GAD, PTSD, social phobia, and obsessive compulsive
disorder. Current research is evaluating the applicability of
the S-REF model in an even broader range of issues such as
alcoholism, psychosis, and persistent distress in survivors of
cancer and myocardial infarctions. Evaluation of treatment
efficacy of metacognitive therapy is in the early stages, but it
is producing very high recovery rates across a range of clinical
disorders.
Conclusions
IP models proliferate in clinical psychology, but the disparate
nature of the underpinning theories leads to very different
treatment approaches. This is a strength of the IP approach as
it provides clinical psychology with a rich theoretical basis,
from which further advancements can be achieved. The gener-
ation of novel interventions in the form of CBM strategies and
metacognitive approaches are exciting innovations that hold
considerable promise for the continually evolving practice of
clinical psychology. In summary, the IP paradigm has resulted
in a wealth of research and clinical developments that ulti-
mately result in significant therapeutic benefits.
See also: Anxiety and Fear; Anxiety Disorders; Attention; Cognitive
Behavior Therapy; Cognitive Bias.
Further Reading
Amir N, Burns M, Beard C, and Bomyea J (2009) Attention modification program in
individuals with generalized anxiety disorder. Journal of Abnormal Psychology
118(1): 28-33.
516 Clinical Psychology: An Information Processing Approach
Bar Haim Y, Lamy D, Pergamin L, Bakermans-Kranenberg M, and van Ijzendoorn M
(2007) Threat-related attentional bias in anxious and nonanxious individuals:
A meta-analytic study. Psychological Bulletin 133: 1-24.
Beck AT (1 976) Cognitive Therapy and the Emotional Disorders. New York: International
Universities Press.
Beck AT and Clark DA (1997) An information processing model of anxiety:
Automatic and strategic processes. Behavior Research and Therapy 35: 49-58.
Brewin CR, Dalgleish T, and Joseph S (1996) A dual representation theory of
posttraumatic stress disorder. Psychological Review 103(4): 670-686.
Cisler J and Koster E (2010) Mechanisms of attentional bias toward threat in anxiety
disorders: An integrative review. Clinical Psychology Review 30: 203-216.
Flavell JH (1979) Metacognition and metacognitive monitoring: A new area of
cognitive-developmental inquiry. American Psychologist 34: 906-91 1 .
Hankin BL, Abela J, Gibb BE, and Flory K (2010) Selective attention to affective stimuli
and clinical depression among youths: Role of anxiety and specificity of emotion.
Journal of Abnormal Psychology 1 1 9: 491-501 .
Ingram RE (1990) Self-focused attention in clinical disorders: Review and conceptual
model. Psychological Bulletin 107: 156-176.
Lachman R, Lachman JL, and Butterfield EC (1979) Cognitive Psychology and
Information Processing: An Introduction. Hillsdale, NJ: Erlbaum.
Mathews A and MacLeod C (2005) Cognitive vulnerability to emotional disorders.
Annual Review of Clinical Psychology 1 : 1 67-1 95.
Phaf RH and Kan K (2007) The automaticity of emotional Stroop: A meta-analysis.
Journal of Behavior Therapy and Experimental Psychiatry
38: 184-199.
Wells A (2000) Emotional Disorders and Metacognition: Innovative Cognitive Therapy.
Chichester, UK: Wiley.
Wells A and Matthews G (1996) Modeling cognition in emotional disorder:
The S-REF Model. Behavior Research and Therapy 34(11): 881-888.
Williams JMG, Watts FN, MacLeod C, and Mathews A (1988) Cognitive Psychology
and Emotional Disorders. Chichester, UK: Wiley.
Cognition and Personality
G Matthews, University of Cincinnati, Cincinnati, OH, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Attentional resources A general-purpose reservoir of
'energy' or 'fuel' for processing that may be flexibly allocated
to support effective attention.
Cognitive patterning A detailed description of the tasks or
cognitive functions that are facilitated or impaired by a
personality trait or stressor.
Executive functioning A set of control operations used to
regulate attention and memory.
Extraversion A personality trait associated with sociability,
assertiveness, and positive affect.
Idiographic approach The assumption that all aspects of
personality are fundamentally unique and idiosyncratic to
each individual.
Neuroticism A personality trait associated with a
predisposition to experience negative affect.
Nomothetic approach The assumption that personality
may be understood using universal concepts and
principles that apply to all people.
Self-regulation The motivational and cognitive processes
controlling goal-directed personal adaptation to the
external environment.
Self-schema A structured set of beliefs about the
self (which may not be fully accessible to
consciousness).
Trait A personal disposition or characteristic showing
long-term stability and influencing behavior in a
range of situations.
Working memory A memory system used for information
processing including both a central executive and
short-term storage components.
Introduction and Historical Context
The essence of the cognitive psychological perspective on per-
sonality is that individuals differ in their understanding of the
outside world and their own place in it. Thus, theories of
individual differences in cognitive processes and structures
should be central to the psychology of personality. Histori-
cally, the route to a cognitive understanding of personality
has been surprisingly circuitous. In psychoanalysis, for exam-
ple, the rationality of the ego often appears subservient to the
basic drives of the id. Later theorists including George Kelly
and Carl Rogers placed cognitive processes at the center of
personality research, through emphasizing the person's active
efforts at making sense of the world and of the self. However,
these clinically oriented approaches lacked the benefits of the
formal models of information processing provided by cogni-
tive psychology.
At the dawn of the 'cognitive revolution' in the 1960s, we
could discern three trends that continue to shape models of
personality. The first comes from H.J. Eysenck's biological
personality theory. Eysenck followed the behaviorist tradition
of seeing cognition as secondary to more basic neural pro-
cesses, but he pioneered the use of performance tasks as a
means for testing personality theory. Tasks requiring memory
and attention were used as a means of testing predictions from
the arousal theory of personality traits. The failure of these
attempts encouraged attempts to use cognitive psychological
models as a basis for predicting personality effects.
The second trend was the development of social learning
theories of personality by Walter Mischel and Albert Bandura.
Such theories emphasized that social learning builds interna-
lized cognitive structures, such as representations of self-efficacy,
that provide both a coherent style of behavioral interaction
with others, and a coherent sense of self. By contrast with trait
theory, social learning theories focus on the cognitive repre-
sentations of the individual.
A third trend was for clinical psychologists to link emo-
tional disturbance to broad individual differences in cognitive
functioning, an approach more compatible with the notions of
general personality traits. A.T. Beck's landmark schema theory
elaborated the role of cognitive knowledge structures as a basis
for personality, in relation to dysfunctional negative self-beliefs
and styles of reasoning that are associated with clinical
depression.
Contemporary personality research remains infused with a
variety of cognitive constructs. However, personality researchers
have adopted cognitive psychological methods and concepts
in order to pursue rather different, although complementary,
research aims. Next, three research directions are distinguished,
relating to experimental studies of performance, the contents
of self- representations in the 'cognitive self/ and long-term
personality development.
Individual Differences in Performance
Personality traits relate to individual differences in perfor-
mance of a variety of tasks. The conceptual model is that the
trait is seen as a fixed quality of the person, which is associated
with a range of biases that feed forward into observed behav-
ior. Studies of human performance seek to relate personality to
information-processing models that specify individual differ-
ences in components of memory, attention, decision making,
and so forth. An important area of this research concerns
cognitive bias in processing emotional stimuli, which has
been related to anxiety and other traits linked to emotionality.
Theory is then concerned with specifying the cognitive pro-
cesses that mediate the behavioral expressions of the trait.
For example, recent anxiety research has explored how anxiety
517
518 Cognition and Personality
effects may be mediated by individual differences in executive
processes. Because information processing can typically be
related to neural systems, work of this kind also lends itself
to cognitive neuroscience theory. Research of this kind is clos-
est to 'standard' cognitive psychology in using experimental
methods to investigate information-processing mechanisms
in detail.
Experimental studies have also been used to investigate
effects of personality on mood and emotion. It is common-
place (although perhaps simplistic) to identify neuroticism
and extraversion traits with negative and positive emotionality,
respectively. Although the dominant theoretical approach has
been to relate these traits to brain systems for emotion and
motivation, researchers are increasingly studying cognitive
processes that may influence emotion, including biases in
appraisal and styles of emotion regulation.
The Cognitive Self
In the classic experimental paradigm, the personality trait is a
'given' that influences cognitive processes and behavior over
a short time interval. However, cognitive theory can also be
applied to understanding the underlying nature of traits, given
that the stability of personality implies that personal character-
istics are represented in long-term memory (LTM). Variation
in traits may correspond to variation in stable cognitive
structures. Building on A.T. Beck's (1967) schema theory,
researchers have explored how traits correspond to stable repre-
sentations of self-knowledge, which in turn shape individual
differences in social behavior. Traits may also correspond to
'implicit' procedural knowledge, such as well-learned scripts
for interpersonal interaction, as well as to 'explicit' declarative
knowledge.
Research on the cognitive self divides sharply into two
separate camps. Following Mischel and Bandura, one line of
research is concerned with 'idiographic' personality coher-
ence at the individual level, and the role of the individual's
cognitive-affective structures in producing consistency of
behavior within specific situations. The alternative perspective
assumes that features of the self such as dysfunctional nega-
tive beliefs may be assessed nomothetically, that is, as an
attribute of all individuals, contributing to general traits. Typi-
cal methods include experimental studies, such as investigat-
ing speed of processing self-referent information, but may
also employ other techniques, including subjective reports
of self-beliefs. Recent research has increasingly focused on
'implicit' aspects of the self, for example, using techniques
based on A.G. Greenwald's Implicit Association Test. For
example, someone whose implicit self-esteem was low might
respond more slowly if required to respond on the basis of an
association between the self and positive qualities.
Personality Development
Temperamental qualities such as emotionality and self-control
are expressed in early childhood, and form the precursors
to adult personality. While temperament is often seen as es-
sentially biologically based, there is increasing interest in
how brain maturation and self-development are interrelated.
For example, attachment theory is concerned with how the
innate neural programs for managing social interaction are
shaped by experience to regulate the child's attachment to the
caregiver. The development of (cognitive) working models of
self and others plays a critical role in intimate relationships
into adulthood. Research on personality and individual differ-
ences in cognitive development uses a variety of methods,
including experimental studies and self-report. Studies of
younger children obtain reports of behavior from parents or
trained observers. Longitudinal studies that may run from
infancy to adulthood are also important.
The developmental perspective helps us to understand how
the cognitive self is acquired, as the child internalizes an
increasingly sophisticated set of self-beliefs and strategies for
self-regulation. Again, traditional social learning theory has
often emphasized self-development as a process best under-
stood at the level of the individual. However, normative traits
may relate to consistent developmental trajectories and styles
of person-environment interaction. For example, aggressive-
ness may relate to beliefs in the hostility of others, the acquisi-
tion of confrontive interpersonal strategies for self-defense, and
willingness to engage with environments in which confronta-
tion is likely.
Structure of this Article
The introductory overview of research on personality and cog-
nition indicates the scope and diversity of this work. The
remainder of this article is organized as three major sections.
The first is concerned with work that is closest to the central
goal of cognitive psychology, that is, to construct and test
detailed models of the information processing that supports
major psychological functions such as attention and memory.
In personality research, the aim is to specify how personality
traits correspond to individual differences in the parameters of
information-processing models. The next section reviews inte-
grative theoretical accounts of empirical findings, focusing on
attentional resources, control processes, and the adaptive func-
tions of cognition. The third section reviews applications of
cognitive psychological research to other research topics which
are concerned not only with individual differences in cogni-
tion, but also affective, motivational, and social functioning.
These topics are self-regulation and emotion, the cognitive self
and social functioning, and abnormal personality and clinical
disorder.
Personality is itself a diffuse field of study, and some com-
ment on its treatment in this article is needed. The research
covered is primarily concerned with nomothetic trait models,
which assume that the major features of personality can be
described in relation to a psychometrically defined standard set
of traits that are meaningful for all individuals. Coverage of
idiographic personality models that emphasize the uniqueness
of the individual will be brief.
The five factor model (FFM) is the dominant trait model
at the present time. It includes extraversion-introversion, neu-
roticism (or emotional instability), conscientiousness, agree-
ableness, and openness as major traits. Traditionally, most
cognitive research has been directed toward extraversion and
neuroticism, and related traits such as anxiety. Work on the
personality 'superfactors' is supplemented by studies of more
narrowly defined traits, sometimes described as primary or
Cognition and Personality 519
midlevel traits. Studies of clinical disorder have referred to
the roles of both traits considered normal, such as neuroti-
cism, and traits labeled as abnormal, such as schizoid and
schizotypal personalities which may be associated with clinical
schizophrenia. Recent work suggests that both types of trait
may be treated as continuous dimensions, rather than discrete
categories. That is, rather than an individual being rigidly
categorized as either schizophrenic or nonschizophrenic, peo-
ple vary in a continuous, graded fashion on the traits that
capture core qualities of schizophrenia, such as experiencing
delusions. In fact, there is quite good convergence between
dimensional models for normal and abnormal domains. This
article will focus primarily on constructs within the normal
domain, but make some reference to abnormal traits.
Personality Traits, Information Processing,
and Performance
The historical impetus for performance studies came from
H.J. Eysenck's (1967) attempts to validate the arousal theory
of personality from studies of various performance tasks.
Interactions between extraversion and various stressful agents
might be explained on the basis that extraverts were vulnera-
ble to under-arousal whereas introverts were prone to over-
arousal. Similarly, neurotic and anxious individuals were seen
as liable to become over-aroused in emotive or stressful
circumstances.
This research was valuable in showing that the impact of
traits on performance is frequently moderated by environmen-
tal factors. However, it has failed to provide an adequate
account of the performance data, for several reasons. First,
evidence suggests that extraversion and neuroticism effects on
personality may be found even with arousal controlled. Per-
sonality effects are not always (or even frequently) mediated by
arousal. Second, the anxiety research of Charles Spielberger
and Irwin Sarason highlighted the primacy of cognitive
mechanisms over emotional physiological arousal in deter-
mining anxiety effects. Anxiety is harmful to performance
because it is accompanied by worries that interfere with atten-
tional processes. This perspective has been especially influen-
tial in studies of test anxiety. Third, personality effects vary
substantially with the information-processing demands of
the task. Performance is not a unitary construct, and traits are
differentially related to different types of task. Fourth, the
emerging cognitive neuroscience of personality, further dis-
cussed below, points toward brain mechanisms more specific
than general cortical arousal as critical to personality.
Cognitive psychologists faced similar issues in seeking to
understand effects of stressors such as noise and heat on
performance. An influential critique set out by G.R.J. Hockey
offered an alternative to arousal theory. He argued that the
diversity of stressor effects across different tasks suggested a
cognitive patterning of effects. The typical stressor appears to
impair only certain components of information processing,
while leaving others intact, or even improving processing.
Hockey also pointed out that stressors may affect strategy as
well as basic processing efficiency. Individuals under stress
may appraise tasks and the priorities for applying effort
differently.
The cognitive patterning approach provides a means for
better understanding personality effects on performance.
The assumption is that traits typically correspond to multiple,
often small, biases in information processing. (Typically, cog-
nitive ability is a stronger influence on processing and perfor-
mance). There remains a need for a systematic investigation of
the processing correlates of traits, but published studies allow
at least an outline of the multiple influences of major traits,
especially extraversion and anxiety/neuroticism.
Cognitive Patterning: Extraversion-Introversion
Table 1 shows an outline patterning for extraversion-
introversion. Discussion of the individual entries in the table
is beyond the scope of this article, but two general themes may
be identified. First, the temporal characteristics of tasks
may be critical. Extraverts appear to have the advantage over
introverts where speed is emphasized, as evidenced by more
risky speed-accuracy tradeoff in reaction time tasks, and faster
motor response execution. Extraverts also tend to do somewhat
better on short-term memory (STM) tasks, whereas LTM may
favor introverts. Tasks in which stimuli are presented infre-
quently, including vigilance and reflective problem solving,
tend to be better performed by introverts. Second, the coding
or representation of task stimuli may be critical. Although
extraversion is largely unrelated to verbal ability, extraverts
show greater verbal facility in tasks including fluent speech
production and retrieval from semantic memory. Indeed,
Table 1 Outline cognitive patterning for extraversion-introversion
Cognitive
function
Sample task
Result
Extra vert superiority
Divided attention
Memory search for
Extraverts faster in
single or multiple
dual-task versions
targets
conditions
Short-term
Free recall of word
Extraverts better at
memory
lists
immediate recall
Speed of motor
Choice reaction time
Extraverts show faster
response
response execution
Speech
Conversation in
Extraverts more fluent in
production
introvert superiority
a second language
speech production
Visual vigilance
Detecting line signal
Introverts show higher
detection rate
Long-term
Paired-associate
Introverts better at
memory
learning
long-term recall
Problem-solving
Problem-solving
Introverts faster and more
tasks requiring
accurate: extraverts finish
insight
impulsively
Qualitative performance differences
Vigilance
Detection of
Extraverts adopt a lower
brighter target
stimulus
response criterion
Response to
Serial choice
Extraverts are faster when
stress and
reaction time
high in arousal; introverts
arousal
are faster when low in
arousal (also depends on
time of day)
520 Cognition and Personality
studies of divided attention suggest that extraverts may have an
advantage in processing multiple symbolic (alphanumeric)
stimuli. Note also that several of these effects are moderated
by the arousal or stress level of the experimental setting. For
example, extraverts may find it especially hard to sustain atten-
tion and vigilance when the surrounding environment is unsti-
mulating and dearousing.
Cognitive Patterning: Anxiety and Neuroticism
A cognitive patterning for anxiety (or neuroticism) is shown in
Table 2. Again, some general themes are evident. Individuals
high in trait anxiety are vulnerable to overload, especially on
tasks making high demands on attention and/or working
memory. Neuroticism is also associated with difficulties in
managing transitions in workload. Second, anxiety may also
be related to compensatory effort as the person tries to cope
with the distraction associated with their own worries. Indeed,
anxiety may sometimes relate to enhanced performance on
easy tasks. Third, anxiety relates to enhanced sensitivity to
threat stimuli, as shown most directly in studies of selective
attention. The emotional Stroop test requires participants to
name the ink colors of threatening or other emotive words.
A. Mathews and C. MacLeod pioneered research showing that
Table 2 Outline cognitive patterning for anxiety/neuroticism
Cognitive function
Sample task
Result
Processing efficiency effects
Divided attention
Concurrent math
Anxiety leads to
and verbal
impairment in
memory
dual-task performance
(especially on
secondary tasks)
Working memory
Mental
Anxiety-related
transformation of
impairment increases
letter sequences
with memory load
Resisting
Comprehending text
Anxiety relates to
distraction
with background
distraction by
Cognitive bias effects
speech
irrelevant speech
Selective attention
Emotional Stroop
Anxiety subjects are
(single channel)
slow to name ink
colors of threat words
Selective attention
‘Dot-probe’ visual
Anxious subjects
(multiple
attention task
respond more quickly
channel)
to probe presented at
location of threat
Semantic
Interpreting spoken
Anxious subjects biased
processing
homophones: for
toward selecting
example, ‘die’
threatening
versus ‘dye’
interpretation
Qualitative performance differences
Response to
Performance with
Anxiety relates to
evaluative stress
evaluative
performance
instructions
impairment when
evaluated
Memory strategies
Free recall of word
Anxiety relates to
lists
reduced strategic
reorganization of
words
anxious individuals tend to respond relatively slowly on this
task, especially when the threat words match their current
concerns, implying a cognitive bias.
Other traits have been studied less extensively, so that no
detailed cognitive patterning may be presented, but there are
several promising research areas. Within the FFM, there is accu-
mulating evidence, often from an organizational perspective,
suggesting that conscientiousness tends to relate to enhanced
performance of a variety of tasks. The information-processing
basis for these findings is unclear, but conscientiousness has
been linked to effort and goal setting. Openness relates mod-
estly to cognitive ability, especially crystallized ability, and
other tests with some intellectual content, including creativity
tests. These relationships may be mediated by motivational
factors, including 'typical intelligence engagement'; that is,
more open individuals have a greater interest in intellectual
activities, supporting skill acquisition.
Cognitive Neuroscience Perspectives
The psychobiological roots of research on personality and
performance ensured that the introduction of cognitive models
to the field was accompanied by continued interest in psycho-
physiological correlates of information processing. One of the
more productive lines of research relates to evoked potential
(EP) responses derived from the electroencephalogram (EEG).
Differences between extraverts and introverts have been found
in the amplitude of the P3 (or P300) wave that may reflect
updating of working memory. These differences may be asso-
ciated with performance differences on working memory task.
In recent years, the cognitive neuroscience of personality
has been boosted by studies of brain imaging, using tech-
niques including functional magnetic resonance imaging
(fMRI). Such studies may indicate the individual differences
in neurological functioning that correspond to performance
variation. There are several promising lines of research. As
already described, traits including extraversion and anxiety
relate to attentional functioning. The cortical brain circuits
that control different attentional processes are increasingly
well-understood, and evidence from fMRI may be integrated
with behavioral studies in order to elucidate the nature of
personality effects.
Traditional biological theories of personality situated traits
in subcortical systems such as arousal circuits and reward and
punishment systems. In contemporary cognitive neuroscience,
such circuits are relevant because their activity may modulate
the cortical systems that more directly control information
processing. For example, dopaminergic afferents may influence
the areas of prefrontal cortex that support executive processing
in attention and working memory. Extraversion may be related
to reward areas of the brain and increased dopaminergic activ-
ity. Research is exploring whether this hypothesis explains the
greater efficiency of working memory in extraverts, relative to
introverts (in some task paradigms).
Similarly, structures including amygdala and anterior cin-
gulate cortex may relate to individual differences in processing
emotional information. Circuits including these structures may
mediate effects of traits on cognitive bias in attending to emo-
tional stimuli, shown by studies of the emotional Stroop test.
Finally, research is also addressing the molecular genetics of
Cognition and Personality 521
the relevant neurotransmitters; future studies may identify
polymorphisms that correspond to individual differences in
cognitive functioning and performance.
Toward Cognitive Theories of Personality Effects
The delineation of the cognitive patterning of a trait is essen-
tially a descriptive exercise, which provides the fine-grained
account of trait effects necessary for subsequent theory build-
ing. Typically, cognitive psychological theories of trait effects
reflect a progression of increasing specificity in relating traits to
information-processing mechanisms. Early studies suggest an
association with some broad cognitive function such as 'atten-
tion' or 'short-term memory,' without isolating any specific
information-processing mechanism. Further work differenti-
ates possible mechanisms and seeks to isolate those that are
responsible for trait effects. For example, the cognitive bias
associated with anxiety might be produced by either an early
'preattentive' bias in threat encoding, or by later strategic pro-
cesses. Ideally, the end-point of theory development should
be a computational model that identifies traits with quantita-
tive model parameters. Extant research has rarely attained this
goal, although promising work has been done from a connec-
tionist perspective, especially in modeling the effects of traits
linked to negative affect.
The cognitive patterning approach emphasizes that trait
effects are distributed across multiple component processes.
By contrast with traditional biological theories of personality,
there is no single key process for trait action. Thus, we may
arrive at a series of 'cognitive mini-theories' that describe the
effect of a given trait on a given class of tasks (e.g., effects of
extraversion on vigilance), but do not provide any general
synthesis of the relationship between the trait and information
processing.
Some researchers have attempted to provide general
accounts of personality effects, which will be reviewed next.
These theories relate to processing resources as a key mediating
construct to the role of executive control functions in mediat-
ing anxiety effects and to the adaptive functions of cognitive
biases associated with personality.
A Processing Resource Model
Processing resources are metaphorical reservoirs of energy
that may be required for processing that remains under volun-
tary control. The concept has been influential especially
in research on divided and sustained attention. There has
been debate over whether resources are unitary, or whether
multiple resource pools energize qualitatively different types
of processing.
M.S. Humphreys and W. Revelle proposed that variation
in resource availability may mediate the effects of both tran-
sient arousal states and a range of personality and motivational
traits. Their multiple resource model incorporates resources
for 'sustained information transfer' (SIT), required for a variety
of demanding attentional tasks, and resources for STM. High
arousal tends to increase the availability of SIT resources, con-
sistent with evidence that stimulant drugs tend to enhance
vigilance and other attentional tasks. Arousal also reduces
STM resources. The theory distinguishes on-task effort from
arousal as an energetic construct. On-task effort increases func-
tional availability of both types of resources.
The theory has inspired work in a variety of areas. The
authors' own research focused primarily on effects of extra-
version and impulsivity and their interaction with arousal
manipulations including time of day and caffeine. They inves-
tigated tasks including cognitive ability tests and more basic
information-processing tasks. They were able to show that
interactions between personality and caffeine were consistent
with the theory. A key assumption derived from the model was
that complex tasks, requiring both SIT and STM resources,
showed the inverted-U curve relating arousal to performance
assumed by traditional arousal theory. A criticism of the model
was that tasks most sensitive to interactive effects of extraversion
and arousal appear to require routine, somewhat automatic
stimulus encoding, and may not be highly resource-limited.
Researchers are continuing to elaborate the theory.
The resource model has also been applied to understanding
individual differences in vigilance. The state factor of task
engagement (combining energy, motivation, and concentra-
tion) reliably predicts performance of a range of demanding
vigilance and visual search tasks. The association between high
engagement and attention is reliable only under high levels of
task demand, as resource theory predicts, assuming task
engagement can be identified with arousal. The hypothesis
that arousal impairs STM has been less successful. Indeed,
high task engagement may facilitate working memory. The
multiple resource model also links STM deficits to reduced
on-task effort, a hypothesis that is compatible with other cog-
nitive models of anxiety.
Attentional Control Theory
Attentional control theory, developed by M.W. Eysenck and his
colleagues, aims to explain anxiety effects on a range of tasks
requiring attention and working memory in terms of specific
executive control processes. It integrates several themes of
existing anxiety research within the understanding of execu-
tive function provided by contemporary cognitive neurosci-
ence. The first theme is the generally detrimental effects of
anxiety and worry on various cognitive tasks. The initial cogni-
tive mechanisms to which these effects were attributed were
broadly described in terms such as cognitive interference and
working memory impairment. However, attention and work-
ing memory may be regulated via multiple executive control
processes, as shown by recent psychometric work in the area.
Inhibition of conflicting responses, shifting between mental
operations, and updating working memory are three separate
control processes. Evidence reviewed by Eysenck and collea-
gues suggests anxiety reliably disrupts inhibition and possibly
shifting, but not updating.
A second theme is that anxiety relates to selective atten-
tional bias, in the form of enhanced attention to threatening
stimuli. The bias effect can be demonstrated with 'subliminal'
stimuli that are masked to the point where they cannot be
consciously recognized, implying that anxiety influences early,
preattentive processes. Attentional control theory assumes that
attention is both stimulus-driven and goal-driven. Anxiety
relates to enhanced sensitivity of the stimulus-driven system
522 Cognition and Personality
to threat. A third theme is that attentional bias is controlled by
expectancy and strategy choice as well as unconscious preat-
tentive processes. Attentional control theory proposes that
weakness of executive control, especially in relation to inhibi-
tion, accentuates the tendency of anxious individuals to be
distracted by the threat stimuli prioritized by stimulus-driven
attention. A fourth theme, articulated in M.W. Eysenck's ear-
lier work, is that anxious individuals may compensate for
disruption of attention via increased effort; an assumption
that is also incorporated into the theory. It successfully inte-
grates much existing anxiety research, although it has been
developed too recently to be tested explicitly, other than in
the ongoing work of Eysenck and his colleagues. These
authors - consistent with a cognitive patterning perspective -
also point out that the theory is directed specifically toward
the effects of anxiety on attention and working memory. Other
performance correlates of anxiety, such as effects on retrieval
from LTM, are beyond its scope.
Cognitive-Adaptive Theory
Cognitive-adaptive theory, developed by G. Matthews, is an
attempt to interrelate the multiple cognitive correlates of traits
such as extraversion and neuroticism. It asks how the cognitive
correlates of the trait support adaption to the individual's life
circumstances. A key assumption is that there is a correspon-
dence between personality traits and environments. A high
level of a given trait equips the person to thrive in certain
environments (with corresponding disadvantages in others).
The cognitive correlates of a trait typically have an indirect
rather than a direct effect on adaptation by facilitating the
development of acquired skills congruent with the trait.
Extraversion-introversion effects illustrate this perspective.
Extraverts are adapted to function well in cognitively demand-
ing social environments. Cognitive correlates of extraversion
(see Table 1) such as good verbal STM, and fluent speech
production support acquisition of the relevant social skills.
Neurologically based attributes of extraversion such as toler-
ance of high arousal and stimulation, and reward-sensitivity,
facilitate adaptation in parallel with cognitive skills. Con-
versely, introverts are adapted to solitary environments in
which task stimuli and positive reinforcement may be infre-
quent. The cognitive-adaptive perspective on anxiety and neu-
roticism suggests that anxious individuals are adapted to
environments in which threats are disguised or ambiguous,
requiring high threat sensitivity and subsequent threat avoid-
ance. Individuals low in anxiety are more geared to direct
confrontation of threat after it becomes salient.
Cognitive-adaptive theory explains the distributed nature of
the cognitive correlates of traits. It is also compatible with
theories that specify information-processing mechanisms in
detail. The theory also goes farther than standard cognitive
accounts in describing the dynamic nature of interaction
between personality and cognition. The 'adaptive triangle'
refers to the interplay between (1) the stable cognitive struc-
tures and biases that provide the basis for personality effects,
(2) self-regulative processes that evaluate performance compe-
tence and coping potential, and (3) outcomes of behavior that
may provide feedback both to influence further skill acquisi-
tion, and the person's beliefs about his or her own competence.
Further Research Directions
The research topics reviewed thus far are those that address
the relationship between personality and cognition most
directly, by collecting objective data on individual differ-
ences in performance from controlled experiments. However,
the relationship between personality and cognition has also
been explored using other methodologies such as analysis of
self-reports of cognitive processes and clinical observations.
This section briefly surveys the contribution of cognitive
research to three further major areas of personality research:
self-regulation and emotion, social behavior, and clinical
disorder.
Self-Regulation and Emotion
There are many theories of emotion, but one influential
approach is to see emotion as a concomitant of self-regulation.
C.S. Carver and M.F. Scheier have suggested that avoidance
and approach behaviors are controlled by feedback loops
that aim to reduce discrepancy with some standard value.
Personal and social standards may be represented as part of
the self. Discrepancy generates negative emotion, whereas an
acceptable rate of progress toward meeting the standard pro-
duces positive affect. The later work of these authors also
describes discrepancy-enlarging feedback loops in which the
system operates to increase distance from a reference value
(e.g., avoiding ridicule). They also discuss the role of nonlin-
ear dynamics. Personality may impinge on several features
of feedback control systems, including the individual's per-
sonal standards, the availability of corrective actions, and
dynamic properties of the system such as speed of response
to feedback. Recent personality research has sometimes
equated extraversion and positive emotionality with approach
tendencies, and neuroticism and negative emotionality with
avoidance.
Self-regulation models complement traditional neuro-
logical models of personality. At a more fine-grained level,
research has identified several potential mediating processes.
Traits may bias cognitive appraisal of stimuli and events.
Neurotic individuals are more likely to appraise events as
threatening, whereas extraversion relates to challenge, elevat-
ing negative and positive affect, respectively. In addition, as
the transactional theory of stress and emotion describes, indi-
viduals manage demanding events dynamically, by imple-
menting coping strategies, and processing feedback on
outcomes of coping. Traits may influence how these cognitive
regulatory processes operate. Neuroticism has been linked to
emotion-focused coping, whereas extraversion is associated
with greater use of task- or problem-focused coping. These
associations may be demonstrated in both controlled labora-
tory settings and in real-life contexts.
Personality also relates to style of mood regulation. Indivi-
duals high in neuroticism and other traits linked to negative
emotion tend to ruminate perseveratively on negative events
and thoughts, perpetuating negative affect and worry. Con-
versely, extraverts may be better than introverts at repairing a
bad mood or maintaining positive affect. Emotions may also
be regulated through choice of activities and exposure to situa-
tions that are mood-uplifting or depressing. High neuroticism
Cognition and Personality 523
individuals are more prone to become involved in disagree-
ments with others that may elicit negative emotion, whereas
extraversion is linked to mood-enhancing social participation.
A new perspective on personality, self-regulation, and emo-
tion derives from research on the new construct of emotional
intelligence (El). Although conceptualization and assessment
of El remains controversial, emotionally intelligent individuals
may possess personality characteristics that support more
effective mood regulation. Attention to emotion, clarity of
emotions, and mood repair are aspects of trait El that may
support mood regulation. Individuals high in El may be better
able to use and express their emotions constructively, even in
difficult, potentially stressful circumstance.
Personality and Social Behavior
As previously indicated, traits may relate to the content of
self-knowledge. For example, self-esteem may relate low neu-
roticism and high extraversion. Because of the social nature of
the self, much of this knowledge refers to social relationships,
such as the extent to which the self is generally liked or
disliked by others. The person's typical social behaviors may
be influenced by this knowledge. In addition, traits may relate
to specific self-regulative processes such as self-verification
(seeking confirmation for self-beliefs) and self-enhancement.
These principles have informed experimental studies of
agreeableness. The trait relates to self-reported attributes
including empathy, cooperativeness, and liking for others.
There is indeed evidence that agreeableness relates to higher
quality of social interaction, more satisfying close relation-
ships and fewer conflicts with others. Controlled observa-
tional studies confirm that agreeableness relates to behaviors
that may promote interpersonal relationships such as greater
visual attention and more 'open' body positions. The effects
of agreeableness may be controlled by cognitive processes
including more positive perceptions of others and greater
attention to prosocial stimuli. Agreeableness also relates to
the social motivations that may be represented within the
cognitive self, including motives to accommodate to the
needs of others, and to resolve conflicts through collabora-
tion. Conversely, aggressiveness as a personality trait relates
to distorted perceptions of others that lead to false attribu-
tions of hostility, and to motives to cope with hostility via
confrontation and aggressive behaviors. At the extreme, such
cognitive distortions may contribute to externalizing disorders
and antisocial personality.
Vulnerability to Clinical Disorders
Although causal relationships between traits and episodes
of acute mental disorder may be bidirectional, it is accepted
that certain traits may elevate vulnerability to disorder. Vulner-
ability may relate to cognitive as well as to biological factors.
A.T. Beck's identification of the encoding of dysfunctional
negative self-beliefs in the self-schema as a risk factor for
depression has already been indicated. Abnormality in infor-
mation processing as well as in the content of self-knowledge
may also be important. For example, a bias toward attending
to threat may lead to a chronic overestimation of the preva-
lence and severity of threat. Indeed, recent experimental studies
conducted by C. MacLeod and colleagues show that training
participants to respond to positive or negative stimuli may
influence anxiety congruently. Memory biases also contribute
to the initiation and maintenance of disorder.
The self-referent executive function (S-REF) theory of
A. Wells and G. Matthews aims to provide a comprehensive
account of the various cognitive factors that may shape
vulnerability to anxiety and mood disorders and their perpetu-
ation. A key assumption is that multiple cognitive processes
contribute to vulnerability. Cognitive content, information-
processing, and dynamic person-situation interaction are all
important. Critical to the symptoms of emotional disorder
are cycles of perseverative worry or rumination that reflect
misdirected executive processing. Several cognitive factors
increase the likelihood of such self-referent processing includ-
ing dysfunctional declarative and procedural self-knowledge,
and excessive attention to threat and to the self. Metacogni-
tions that worry is effective for problem-solving are also
harmful. Emotion-regulation strategies intended to reduce
awareness of disturbing thoughts and images may, paradoxi-
cally, strengthen dysfunctional self-knowledge. Avoidance of
feared situations may produce a similar, harmful outcome, by
preventing opportunities for disconfirming irrational negative
self-beliefs. Both neuroticism and more narrowly defined per-
sonality traits related to metacognitive style may tend to
enhance these maladaptive styles of interaction with the exter-
nal environment. These insights have supported cognitive-
behavioral therapies directed toward the person's style of
attention and metacognition.
Abnormal personality traits may relate to attentional
impairments that contribute to additional mental disorders.
The strongest body of research surrounds the schizotypy trait.
Schizotypy refers to traits such as unusual and disorganized
patterns of thinking, together with interpersonal difficulties,
that may raise vulnerability to schizophrenia. Several studies
show that schizotypy is associated with impairments in selec-
tive attention, associated with weaknesses in inhibition of
task-irrelevant stimuli. P. Venables proposed that the selective
attention deficit tended to flood the person's consciousness
with confusing and sometimes bizarre ideas and images, sim-
ilar to the positive symptoms of schizophrenia. Additional
traits that have been linked to maladaptive attention include
sensation-seeking, impulsivity, and tendencies to cognitive
failures.
An Integrative Perspective on Personality and
Cognition
One of the difficulties of cognitive personality research is that
of seeing the forest for the trees. Personality appears to be
associated with often small-magnitude biases in a wide variety
of cognitive processes. G. Matthews has suggested that it is
useful to place cognitive expressions of personality within the
trilevel explanatory framework for cognitive psychology
advanced by Z. Pylyshyn. Personality effects on behavior may
variously be attributed to individual differences in
(1) parameters of the physical, neural architecture for infor-
mation processing (biological level),
524 Cognition and Personality
(2) parameters of the virtual processing architecture associated
with constructs such as attentional resources and working
memory (symbol-processing level), and
(3) self-knowledge, personal goals, and beliefs about strategies
for goal attainment (semantic level).
Thus, different problems in personality call for different
types of cognitive explanation. Cognitive-adaptive theory is
an attempt to show how the information-processing correlates
of personality traits may cohere around individual differences
in adaptation. At the most basic level, traits may represent
different styles of adaptation to human social environments,
adaptations that call for different cognitive strengths.
See also: Anxiety and Fear; Attention; Clinical Psychology:
An Information Processing Approach; Cognitive Bias; Extraversion-
Introversion; Personality, Structure.
Further Reading
Canli T (2009) Neuroimaging of personality. In: Corr PJ and Matthews G (eds.)
Cambridge Handbook of Personality, pp. 305-322. Cambridge: Cambridge
University Press.
Caprara GV and Cervone D (2008) Personality: Determinants, Dynamics, and Potentials.
New York, NY: Cambridge University Press.
Carver CS and Scheier MF (1998) On the Selt-Regulation of Behavior. New York, NY:
Cambridge University Press.
Eysenck MW (1997 ) Anxiety and Cognition: A Unified Theory. Hove, England:
Psychology Press.
Eysenck MW, Derakshan N, Santos R, and Calvo MG (2007) Anxiety and cognitive
performance: Attentional control theory. Emotion 7: 336-353.
Humphreys MS and Revelle W (1984) Personality, motivation, and performance:
A theory of the relationship between individual differences and
information-processing. Psychological Review 91: 153-184.
Jensen-Campbell LA, Knack JM, and Rex-Lear M (2009) Personality and social
relations. In: Corr PL and Matthews G (eds.) Cambridge Handbook of Personality,
pp. 506-523. Cambridge: Cambridge University Press.
Lubow RE and Kaplan 0 (2000) Psychopathology and individual differences in latent
inhibition: Schizophrenia and schizotypality. in: Matthews G and Gruszka A (eds.)
The Handbook of Individual Differences in Cognition: Attention, Memory, and
Cognitive Control. New York: Springer.
Mathews A and MacLeod C (2005) Cognitive vulnerability to emotional disorders.
Annual Review of Clinical Psychology 1 : 1 67-1 95.
Matthews G (2008) Personality and information processing: A cognitive-adaptive
theory. In: Boyle GJ, Matthews G, and Saklofske DH (eds.) Personality Theories and
Models. Handbook of Personality Theory and Testing, voi. 1 , pp. 56-79. Thousand
Oaks, CA: Sage.
Matthews G, Deary IJ, and Whiteman MC (2009) Personality Traits, 3rd edn.
Cambridge: Cambridge University Press.
Robinson MD and Sedikides C (2009) Traits and the self: Toward an integration. In:
Corr PL and Matthews G (eds.) Cambridge Handbook of Personality, pp. 457-472.
Cambridge: Cambridge University Press.
Schnabel K, Asendorpf JB, and Greenwald AG (2008) Using implicit association
tests for the assessment of implicit personality self-concept. In: Boyle GJ,
Matthews G, and Saklofske DH (eds.) Personality Measurement and Testing.
Handbook of Personality Theory and Testing, voi. 1, pp. 508-528. Thousand Oaks,
CA: Sage.
Wells A and Matthews G (1 994) Attention and Emotion: A Clinical Perspective.
Hove, Sussex: Lawrence Erlbaum.
Zeidner M (1998) Test Anxiety: The State of the Art. New York: Plenum.
Zeidner M and Matthews G (2010) Anxiety 101. New York: Springer.
Relevant Websites
http://dev2.affinitynewmedia.com/uea_psyc/links-to-other-personality-labs-
pg82.html - Philip Corr, of the University of East Anglia, maintains this list of
personality labs.
http://ipip.ori.org/ipip/index.htm - The International Personality Item Pool (IPIP) is a
useful open-source resource for obtaining scales for numerous personality traits.
IPIP scales may not be validated as extensively as the personality guestionnaires
developed by test publishers.
http://www.personality-project.org/personality.html - The Personality Project, directed
by William Revelle of Northwestern University, provides information on numerous
aspects of contemporary personality research, including cognitive models.
Cognitive Behavior Therapy
R Hagen and 0 Hjemdal, Norwegian University of Science and Technology, Trondheim, Norway
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behavioral experiments Planned experimental activities
based on experimentation or observation, which are
undertaken between patients between therapy sessions. The
primary purpose is to help patients to test the validity of
their maladaptive beliefs, in order to obtain new
information about themselves and the world.
Cognitive restructuring A technique used in cognitive-
behavioral therapy to identify and correct negative thinking
patterns, by replacing them with more rational beliefs.
Exposure A technique used in cognitive-behavioral therapy
which involves presenting a patient with anxiety-provoking
stimuli for a long enough time to decrease the intensity of
his or her emotional reaction.
Metacognitions The knowledge you have of your own
thoughts and the factors that influence your thinking.
Mindfulness A kind of nonelaboradve, nonjudgmental,
present-centered awareness in which each thought, feeling,
or sensation that arises in the mind is acknowledged and
accepted as it is.
Introduction
The development of cognitive behavior therapy could be
divided into three generations. Behavior therapy (BT) could
be considered as the first wave, where behavioral principles
developed in the laboratory were used in a clinical setting in
treating psychological problems (e.g., phobias). The second
wave was cognitive therapy (CT) developed by A.T. Beck in
the early 1970s, which initially was developed to treat depres-
sion, but later expanded to the treatment of other psychiatric
disorders as, for example, anxiety and personality disorders.
From the late 1980s to early 1990s, we could notice a synthesis
between behavior therapy and cognitive therapy, into cognitive
behavior therapy, which today has a large evidence base related
to its treatment efficacy. During the two last decades, a number
of new treatments have evolved from the cognitive therapy
foundation, which many have classified as the 'third wave' of
cognitive behavior therapy. This article describes the theory
behind and therapeutic techniques used in traditional cogni-
tive behavioral therapy (CBT) and newer developments in the
third wave of cognitive therapy. In addition to treatment prin-
ciples and cognitive theory, the efficacy of cognitive behavioral
therapy is also discussed.
What Is CBT?
CBT has rapidly grown as a therapy approach in the last
20 years. It acknowledges that the processes between cognitions,
emotions, and behaviors are influenced by complex interac-
tions between biological processes and environmental and
interpersonal influences. The relation between cognitions,
emotions, and behavior is accentuated in this approach and
particularly how change in cognitions could influence altera-
tions in emotions and behavior. Departing from psychoanalytic
concepts, several post-Freudian theorists like Adler, Sullivan,
and fforney have influenced the development of CBT. Core
beliefs and self-schemas from Kelly's personal construct, and
the rational-emotive therapy from Ellis contributed to the
theory and methods used in CBT. Beck and others extended
CBT to a wide array of conditions like depression, anxiety dis-
orders, eating disorders, schizophrenia, bipolar disorder,
chronic pain, personality disorder, and substance abuse. More
than 300 controlled studies of CBT have been completed for a
variety of psychiatric disorders.
CBT integrates two originally separate theoretical under-
standings and treatment approaches, namely, the behavioral
and the cognitive approach. The most stringent behavioral
approach exclusively focuses on observable and measurable
behavior and does not take into account the mental processes.
The cognitive approach, according to Beck, asserts that emo-
tional disorders are maintained by a 'thinking disorder' where
cognitions are viewed as antecedents and determinants of feel-
ings and behaviors. Negative automatic thoughts are essential
in CBT and manifest themselves as rapid content-specific
thoughts. For example, in depression, these thoughts have a
content that reflects negative ideas about the self, the world,
and the future, while for anxiety, they usually reflect the
theme of danger and an underestimation of coping. The cogni-
tive theory further states that negative automatic thoughts
are rooted in early negative experiences that may facilitate
the formation of more permanent negative beliefs (core
beliefs). Particular situations relevant to a negative belief may
activate it. Once the negative belief has been activated, negative
automatic thoughts will occur in the stream of consciousness.
Beck therefore proposed a cognitively based therapy aiming
to reverse dysfunctional cognitions and related behaviors.
Panic disorder is one particular example where cognitive and
behavior theories come together. According to Clark's cognitive
model of panic disorder, patients suffering from panic attacks
typically have cognitive symptoms like catastrophic fears of
physical calamites or loss of control combined with behavioral
impulse to escape or avoid. Altering dysfunctional cognitions
and encouraging the reversal of maladaptive behavioral coping
like avoidance would therefore be a focus in CBT.
The principal element of CBT is that cognitive processing is
given a prominent role, because humans continually appraise
the significance of the events they observe and take part in.
Depending on the appraisal, emotional and physical reactions
often form the basis for behavioral responses. According to the
525
526 Cognitive Behavior Therapy
Heimberg's cognitive model of social phobia, patients with this
disorder may have the following thoughts related to an invita-
tion to a social event: "I don't know what to say . . . Everyone
can see that I am nervous . . . Others will think I'm stupid ... It
is best if I don't go at all ..." These thoughts that are typical
in social phobia could give emotional and physiological
responses like severe anxiety, physical tension, and autonomic
arousal like dry mouth, increased heart rate, and increased
sweating. Avoidance of a feared situation strengthens the rela-
tion between negative thinking and avoidance, primarily
because the avoidance gives an immediate reduction of stress-
ful emotional and physical reactions. However, the avoidance
also limits or eliminates alternative learning experiences be-
cause an individual avoids situations that may contribute to
new information, and therefore strengthens his or her dysfunc-
tional assumptions.
Basic concepts
In CBT, it is not the event in itself that affects our behavior and
emotion, but it is how the event is appraised that influences
our responses. CBT has identified three levels of cognitive
processing, namely (1) conscious processing, (2) automatic
thoughts, and (3) schemas (core beliefs). The first process
allows us to monitor and assess our interaction with the envi-
ronment, integrate past memories with present experiences,
and plan and control the future. The two latter concepts are
parts of automatic information processing and thus not part of
a controlled and voluntary information process. In this article,
the concept of automatic thoughts and schemas will be
described.
Automatic thoughts
Automatic thoughts are cognitions that flow rapidly through
our minds while we are remembering a situation or are in the
midst of one. These thoughts are typically subliminal and not
subjected to careful rational analysis, but they do influence
how we evaluate the significance of particular events. All
humans have automatic thoughts, but it is only when the
situation triggers negative automatic thoughts that these
thoughts form the basis for psychological disorders. Research
has shown that individuals with psychological disorders often
experience a flood of maladaptive automatic thoughts that
contribute to generate stressful emotional reactions and dys-
functional behaviors. For depression, the negative automatic
thoughts often spin around themes like failure, hopelessness,
and low self-esteem. For anxiety disorders, such thoughts often
focus on the prediction of danger, harm, uncontrollability, and
inability to manage threats. In CBT, increased awareness of
primarily negative automatic thoughts and subsequently the
emotions and behaviors they contribute to create is among
the initial steps of the treatment.
Cognitive errors
Initially, in CBT, it was theorized that there would be charac-
teristic errors of logic in the automatic thoughts typical for
individuals with emotional disorders. Research has later con-
firmed that there are pathological styles of information proces-
sing where characteristic cognitive errors are typical. Beck and
colleagues have identified six main categories of cognitive
errors: arbitrary inferences, selective abstraction, dichotomous
or all-or-nothing thinking, magnification and minimization,
overgeneralization, and personalization. The CBT therapist
typically educates the patients about these cognitive errors
which thereby increase the patients' ability to identify when
they make such cognitive errors, and make new evaluations
less influenced by them.
Schemas and core beliefs
In CBT theory, schemas are core beliefs that underlie or govern
information processing, and underlie the more superficial
layer of automatic thoughts. These core beliefs are enduring
principles of thinking that are typically shaped in early child-
hood. A multitude of life experiences influence the develop-
ment of these schemas, like parental modeling and teaching,
educational activities along with peer experiences and traumas.
It is theorized that these schemas develop in order to facilitate
the processing of the multitude of information that individuals
face in their everyday life. It is thought that they serve as a
screening, filtering, or coding function in assigning meaning to
information for the individual. Some researchers differentiate
between nuances of the levels of schemas; however, it has been
found that most patients benefit from recognizing the general
concepts of maladaptive schemas and that these have a strong
influence on self-esteem and behaviors. The goal in CBT is to
identify and develop adaptive schemas and simultaneously
modify and reduce the influence of the maladaptive schemas.
Overview of CBT Therapy Approach
There are some fundamental principles for administering CBT
treatment. The cognitive processes involved in appraisal of
information described earlier are equally relevant for clinical
and nonclinical individuals and as such relevant for all
humans. The CBT is at present, problem- and content-oriented.
In addition to a thorough diagnostics, it is very important to
develop an individualized case formulation at the beginning of
therapy, as this will serve as a guideline throughout the therapy
and direct therapist interventions. An early identification of a
shared goal of therapy is important and can facilitate the thera-
peutic alliance. The CBT sessions follow a particular pattern for
each session, starting with establishing an agenda, establishing
the main themes of therapy are discussed, designing content
appropriate homework tasks, and at the end of the session,
letting clients summarize the session and learning points.
The general approach is founded on scientific principles,
and the therapist aims to view the client's assumptions, auto-
matic thoughts, and schemas as testable scientific hypothesis
that ideally can generate behavioral experiments in order to
check out the validity of the patient's thoughts. This is often
labeled collaborative-empirical therapeutic relationship. One
particular often used technique called Socratic questioning
involves formulating questions in order to stimulate curiosity
and inquisitiveness and aiming to get the patient involved in
a learning process. This technique has been shown to facilitate
guided discovery, where the therapist leads the patient to dis-
cover the dysfunctional patterns and behaviors and thereby
generate alternative cognitions or operationalize their assump-
tions to make them testable in behavioral experiments.
Psychoeducation is also an important part of CBT in
Cognitive Behavior Therapy 527
normalization of symptoms, also in addition to deepening the
understanding of the relation between cognition, emotion,
and behavior.
Particular techniques
Identifying negative automatic thoughts and subsequently the
underlying core beliefs is a primary objective of CBT. Once
the content of the negative automatic thoughts and schemas is
identified, these assumptions are later challenged. This is often
labeled cognitive restructuring, which basically involves iden-
tifying cognitive errors, examining the evidence (pro-con ana-
lyses), reattribution, exploring alternative explanations or ways
of understanding situations, and listing rational alternatives
and cognitive rehearsal and behavioral experiments to test
hypothesis based on either negative automatic thoughts or
schemas. One example of a particular technique in CBT is the
ABC technique. Listing Antecedents (situations), Beliefs, and
(emotional) Consequences, in three columns, familiarizes the
clients with how particular situations may activate different
automatic beliefs, which again elicits emotional reactions. Pro-
posing and exploring alternative beliefs may alter the emo-
tional reaction.
Behavioral techniques are primarily designed to help clients
break patterns of avoidance and helplessness, gradually face
feared situations, build adaptive coping skills, and reduce mal-
adaptive emotional and autonomic arousal. A behavioral
experiment is designed to explore the validity of the patient's
existing beliefs about themselves, others, and the world, and
to construct and test new more adaptive beliefs. Behavioral
experiments take a scientific approach and tests whether a
particular automatic thought or a core belief could predict
what happens in the real world. It is therefore important to
generate one or more precisely operationalized predictions
based on the theory the client and therapist want to test. If
the theory of a patient suffering from social phobia is 'I blush
when I talk in social situation and therefore others think I am
stupid,' then it should be possible to observe the blushing on a
video, and further we should check if others really think some-
one is stupid for blushing. Behavioral consequences of partic-
ular sets of cognitive errors may lead to behavioral avoidance
and safety-seeking behaviors, which in itself contributes to
maintain the psychological disorder. Following each behav-
ioral test, it is important to evaluate the outcome, and explore
if the predictions were true and if there is reason to maintain or
abandon previous behavior and negative automatic thoughts
and schemas.
Efficacy of CBT
CBT has played a very important role in establishing the
evidence-based approach within psychotherapy, by using
meticulous scientific methods to explore the validity of theory
and therapy. Initially a treatment for depression, CBT has
become a treatment approach to very many different psychiat-
ric disorders. With time, many studies have evaluated the
efficacy of CBT, so meta-studies are rather common. One such
study undertaken in 2006 by Butler and colleagues summarized
16 methodologically rigorous meta-analyses. Large effect sizes
were found for CBT for unipolar depression, generalized anxiety
disorder, panic disorder with and without agoraphobia, social
phobia, posttraumatic stress disorder, and childhood depressive
and anxiety disorders. Moderate effect sizes were found for
marital disttess, anger, childhood somatic disorder, and chronic
pain. CBT and behavioral therapy was equally effective for
adult depression and obsessive-compulsive disorder. For
bulimia nervosa and schizophrenia, there were found large
uncontrolled effect sizes. Another meta-analysis of 108 treat-
ment trials across anxiety disorders in 2007 by Norton and
colleagues found that treatment was superior to no treatment
and expectancy control treatments.
In summary, there are many randomized controlled treat-
ment trials for CBT with large numbers of participants that
collectively support the efficacy of CBT across a wide variety
of diagnoses. Collectively, these studies suggest that CBT could
be considered the treatment of choice for anxiety and depres-
sive disorders. Despite extensive evidence, the scientific and
hypotheses-driven approach of CBT has resulted in the devel-
opment of newer cognitive treatment approaches, which we
describe next.
The Third Wave of Cognitive Behavior Therapy
The third wave represents a new approach to CBT, and
examples of such so-called third wave CBT therapies
include Acceptance and Commitment Therapy (ACT), Dialec-
tical Behavior Therapy (DBT), Metacognitive Therapy (MCT),
Mindfulness-Based Cognitive Therapy (MBST), Cognitive
Behavioral Analysis System of Psychotherapy (CBASP), Func-
tional Analytic Psychotherapy (FAP), and Integrative Behav-
ioral Couple Therapy (IBCT).
Compared to traditional CBT, the aforementioned types of
third wave cognitive behavior therapies have important differ-
ences between them and some shared features. While tradi-
tional CBT is primarily engaged in working with the content of
thoughts (disputing, change and restructuring content of the
person's thoughts), these new forms of cognitive behavioral
therapies try to work to change the relationship between the
person and his or her thoughts and feelings. The therapeutic
methods used for change have also a more experiential charac-
ter than the didactic approach which is well known in tradi-
tional cognitive behavior therapy. In many of the third wave
therapies developing mindfulness and acceptance of your
thoughts play a leading role, while disputing the content in
the thoughts is emphasized in naditional CBT. It would be
impossible to give a full description of all the third wave
cognitive behavioral therapies in this article. As ACT, DBT,
and MCT are treatment approaches developed to treat a
broad variety of psychological problems and also have been
tested out empirically, these treatments are described in some
detail in relation to their theoretical basis, treatment principles,
and treatment efficacy.
Acceptance and Commitment Therapy
ACT described their theoretical basis for understanding psy-
chopathology, and their model for psychological treatment is
embedded in what is called Relational Frame Theory (RFT).
The details of RFT go beyond the scope of this article, so just a
very short version of the theory is given here. From a RFT
528 Cognitive Behavior Therapy
perspective, psychopathology develops mainly because indivi-
duals have problems in separating the process of thinking from
the products of thinking. This cognitive fusion between the
process and products of thinking leads to an excessive or
improper regulation of behavior by verbal processes. The beha-
viors of individuals will then be guided more by their inflexible
verbal networks than by contingencies of reinforcement in
their environment. The process of cognitive fusion leads to
other processes in developing psychopathology. As individuals
have problems in separating the content of thoughts from
the process of thinking (thinking the thought), many people
start to fear and avoid their own thoughts, feelings, and bodily
sensations as a way to cope with and regulate their negative
emotions. This process of escaping thoughts and emotions is
destructive, because it leads to what is called 'experiential
avoidance.' It will trigger people to think that certain emotions,
thoughts, and bodily sensations are dangerous to experience
and therefore must be avoided. All in all, these psychological
processes will cause people to miss out of their experiences of
their present moment. They will disappear in what Stephen
Hayes describes as "a cacophony of human thinking and its
reasons, explanations and justifications for behavior." In accor-
dance with RFT, it would therefore be of importance to down-
regulate these negative processes so that people could start
clarifying their personal values and goals and let them guide
their behavior instead.
In ACT, psychological suffering is seen as being caused by
experiential avoidance and cognitive fusion. This would result
in psychological inflexibility that leads to a failure in taking the
needed behavioral steps into acting according to a person's
core values of life. If people's actions are directed by their
personal values, it would bring more vitality and meaning
into their lives. Making people clarify their values and act
according to these goals would increase their psychological
flexibility and make them less vulnerable to psychological
problems.
ACTs principles of treatment is focused around six core
treatment principles, where all of them are aimed at helping
the client to develop an increased psychological flexibility,
making the client experience the present moment more fully,
and trying to let the clients' behavior be guided by their per-
sonal values and goals. A short description of these six treat-
ment processes in ACT is given in the following paragraphs.
• Cognitive defusion: Cognitive defusion techniques try to
make the client to change the way one interacts with or
relates to one's thoughts. Instead of trying to alter the form
of the thoughts and their frequency, the client is taught to
perceive thoughts and emotions as they are, not as what
they appear to be.
• Acceptance : Acceptance of the thoughts and emotions is
taught as an alternative to experiential avoidance. Instead
of trying to avoid specific thoughts and emotions, the client
is taught to allow emotions and thoughts to come and go,
without struggling to suppress and avoid them.
• Contact with the present moment: ACT tries to increase the
client's skill of having a nonjudgmental contact with his or
her private events and events in the environment as they
occur. In order to make this happen, the client is trained in
developing skills to develop a better awareness of the 'here
and now.' If the experience of the 'here and now' is guided
by interest and openness instead of an experiential avoid-
ance, it would increase psychological flexibility in the
individual.
• Observing the self: Since human language leads to a sense of
self as a locus of perspective, the self cannot be contacted
consciously since it is at the core of consciousness itself. It is
therefore essential that the client gets access to his or her
transcendent sense of the self. A transcendent sense of the
self instead of having a picture of the self as a fixed entity,
would help the client to experience his or her thoughts and
feelings as ideas of themselves, instead of permanent and
predetermined characteristics.
• Values: Early in treatment, the client is asked to list values in
different life domains, in order to make them discover what
is important to oneself and life. When the client chooses
directions in life which are based on their personal values
instead of choices based on avoidance or cognitive fusion,
it could help the client to develop a more vital and
meaningful life.
• Committed action: While personal values never really can be
fully achieved, committed actions involving these values
must be broken down to make concrete short- and long-
term goals. Based on these goals, committed actions to
achieve their values are carried out by clients.
These core treatment processes in ACT are overlapping and
interrelated, and could be chunked into two groups of inter-
ventions according to ACT. Mindfulness and acceptance pro-
cesses involve acceptance and cognitive defusion which will
result in a better contact with the present moment, and the
development of the client's ability to see their personal self in a
context. The processes of commitment and behavior change
involve contact with the present moment, where the clients
could see themselves in a context, and make their values
explicit and committed actions in accordance with these
values. All of these treatment processes use different exercises,
homework, and metaphors in order to work and change the
client's psychological problems.
There are a substantial number of controlled trials investi-
gating the efficacy of ACT on a diversity of psychiatric disor-
ders. A meta-analysis by Ost done in 2008 showed that ACT
compared to waiting list conditions reported a large effect size
both at treatment and follow-up. Powers followed up with a
new meta-analysis in 2009 where further new ACT studies were
included. The meta-analysis from Powers showed that ACT
outperformed control conditions as waiting list, treatment as
usual (TAU), and psychological placebo both at posttreatment
and follow-up with a medium effect size. However, when ACT
was compared to established active treatments as for example
CBT, ACT showed no distinct advantage over these treatments.
Dialectical Behavior Therapy
DBT is a psychotherapeutic intervention originally developed
to treat persons with borderline personality disorder (BPD).
Although it was originally developed for chronically sui-
cidal individuals with BPD, DBT has since then been adapted
for other psychiatric disorders characterized by emotional
dysregulation problems. These disorders include substance
Cognitive Behavior Therapy 529
dependence, binge-eating disorders, and depression in elderly
and suicidal adolescents.
DBT came into being mostly because traditional cognitive
behavior treatment showed important shortcomings in the
treatment of BPD, and combines the basic strategies for behav-
ior therapy with eastern mindfulness practices. DBT tries to
integrate both the concepts of acceptance and change in their
treatment philosophy. The therapist is focused on accepting
and validating people's feelings, thoughts, and behaviors as
reactions of patients' suffering and problems, but at the same
time trying to educate patients that some of these feelings and
behaviors are maladaptive and in need of change. The DBT
treatment could be broken down in phases where in the first
phase self-injurious and suicidal behavior takes the main con-
cern and needs to be taken under control, which is then fol-
lowed by a focus on therapy-interfering behaviors. The next
phase is directed on how to enhance the quality of life and
make the clients feel better, to resolve problems in living and
residual disorders, and finally to find joy and meaning in their
lives. Acceptance procedures in DBT include mindfulness and a
variety of other validation and acceptance-based strategies.
Change strategies in DBT include behavioral analysis of mal-
adaptive behaviors and the introduction of new problem-
solving techniques, including skills training, contingency
management, cognitive modification, and exposure-based stra-
tegies. The goal of the treatment is to help the patient to engage
in functional life-enhancing behavior even when intense
emotions are present. The therapy could be split into several
components. Among the components is the use of highly
structured individual or group skill training, individual psy-
chotherapy, and telephone contact between the patient and the
individual therapist. In both group and individual therapy,
learning and improvement of skills is emphasized. These spe-
cific skills are divided into four modules: mindfulness skills,
interpersonal skills, emotional regulation skills, and distress
tolerance skills. In addition to all these aforementioned tech-
niques, a team regularly meets the therapists to keep them
motivated and supervise them in order to provide effective
treatment for their patients. For a further in-depth description
on the treatment principles of BPD, the books listed in the
Further Reading section is recommended.
Since DBT is developed for the treatment of BPD, it is
understandable that most of the studies done on the efficacy
of DBT are focused on BPD, although there are a few studies
related to the use of DBT in treating eating disorders and on
depression. The original DBT treatment lasts for lyear, but
studies with shorter treatment duration has been published.
Ost's meta-analysis from 2008, showed an overall effect-size of
a mean of 0.58 the DBT studies included in this meta-analysis.
In line with the results of ACT, DBT showed large effect sizes
compared to a waiting list condition, but in comparison to
treatment as usual and other active Ueatment, the effect size
was moderate.
Metacognitive Therapy
Metacognition refers to the cognition applied to cognition and
may be defined as any knowledge or cognitive processes that is
involved in the appraisal, control, and monitoring of thinking.
According to metacognitive theory, there is a distinction
between metacognitive knowledge, which is information that
individuals have about their own thinking and about the stra-
tegies that affect it, and metacognitive regulation, which com-
prises the strategies used to change the status of thinking. As
many people experience negative thoughts but do not develop
psychological disorders, MCT proposes that thought content is
not so important as suggested in CBT, but it is the reactions to
the thoughts which play a prominent role in the development
and maintenance of psychological disorders. MCT is focused
on the core question: Why do people have problems regulating
their preservative and fixative style of thinking?
The theoretical grounding of MCT is the Self-Regulatory
Executive Function (S-REF) model, which emphasizes the simi-
larities in maladaptive cognitive processing across psychologi-
cal disorders. The S-REF model postulates a thinking style
called the Cognitive Attentional Syndrome (CAS). The CAS is
a universal feature of psychiatric disorder and is responsible
for prolonging and intensifying distressing emotions within
the individual. The CAS consists of inflexible self-focused
attention, preservative thinking style in the form of worry
and rumination, and threat monitoring and coping behaviors
that backfire and fail to modify erroneous beliefs about your
thinking and the world around you. The CAS is considered
to be a problem for psychological well-being because it
focuses appraisal on threat, it fails to provide information
that can modify erroneous appraisals and beliefs, it drains
attentional resources for more adaptive responses, it also
biases perception and automatic processing in a negative and
skewed manner, and consequently it maintains the psycho-
logical problems. Threat monitoring is a particular process of
the CAS where a person focuses his or her attention to external
and internal threats as a way to cope. This way of coping causes
the individual to become hypervigilant and aware of potential
internal and external threats. Such a focus often leads to
negative thoughts, worry, and rumination, causing the person
to become depressed and anxious. Avoidance and thought
suppression is the last template of the CAS. Such cognitive
processes block new information processing and thus preserve
erroneous beliefs about themselves, others, the world, the
past, and the future. There is a large evidence-base to prove
that the CAS is associated with the vulnerability and serves
as preserving factor of emotional disorders.
The CAS is driven by metacognitive beliefs and knowledge
stored in long-term memory, and exists as programs for guid-
ing cognitive information processing and coping behaviors.
MCT implies that all disorders could be linked to a higher
level of metacognitive beliefs about thinking and that these
fall into either positive or negative categories. Positive metabe-
liefs concern the advantages of worrying, ruminating, threat
monitoring, and controlling of cognition. An example of such
positive metabeliefs could be 'Worrying about the future helps
me to be prepared for forthcoming negative events.' Negative
metabeliefs are linked to the uncontrollability of worries and
rumination or the dangerous consequences of such thinking
processes. An example of such negative metabeliefs could be
'Worrying too much could make me lose control, and make
me go mad.' MCT suggests that positive metabeliefs make a
person vulnerable to develop emotional disorders and that
negative metabeliefs serve as a maintaining factor for distress.
The problem is that the metabeliefs lock the individual into a
530 Cognitive Behavior Therapy
processing style filled with worry and rumination which pro-
longs anxiety and depressive symptoms, because the person
focuses on metabeliefs and cognitive processes that strengthen
dysfunctional patterns. CAS occurs in response to initial nega-
tive thoughts or intrusions and the persons use worry and
rumination as a way to cope with these thoughts.
A metacognitive therapy normally starts on building a
case conceptualization that is grounded in a disorder-specific
model. When doing metacognitive therapy, the therapist needs
to work at the metacognitive level. In order to do this, the MCT
therapist focuses on detecting and modifying the aspects of the
CAS; preservative thinking, maladaptive attentional strategies,
and unhelpful coping strategies. Clients are helped to see that
problems are not the occurrence of negative thoughts, but that
their problems lie in how they react to the thoughts them-
selves. It is essential that the client shifts from a content focus
to a metacognitive way of processing. This will facilitate the
development of a new awareness to their cognitive processes.
Techniques from the standard CBT, such as Socratic ques-
tioning and behavioral experiments, could be used, but in
MCT, these techniques are aimed at the metacognitive level
rather than at the cognitive level. Metacognitive beliefs are
restructured through experiencing new types of relationships
with cognitions, so the CAS could be modified. The MCT
therapist incorporates techniques such as detached mindful-
ness, attentional training, and situational attentional refocus-
ing in order to manipulate and change metacognitive beliefs
and the CAS. It is essential for the client to develop new plans
for processing by challenging the patient's negative and posi-
tive metabeliefs. Such changes are typically changed through
new corrective experiences of their thoughts and thinking, with
very little focus on the actual content of thoughts.
There is a large evidence base supporting the role of the
CAS and the importance of metacognitions in psychological
disorders. Some studies on the effectiveness of MCT have
been completed, and other studies are ongoing. Several open
trials indicate that MCT is an effective treatment for generalized
anxiety disorders, social phobia, posttraumatic disorder, obses-
sive compulsive disorder, and major depressive disorders.
However, there is a lack of large randomized controlled trials
comparing MCT to established evidence-based treatments,
which implicates a need for such studies before conclusions
regarding the efficacy of metacognitive therapy could be drawn.
Conclusions
This article has described what characterizes different forms of
cognitive behavioral therapies. While traditional CBT is pri-
marily engaged in working with the content of thoughts, the
third wave of cognitive behavioral therapies tries to work to
change the relationship between people and their thoughts
and feelings, and with less focus on the content of thoughts.
In the last four decades, CBT has shown that it is an effective
psychotherapeutic treatment approach for mostly all of
the psychiatric disorders listed in the diagnostic manuals.
Although the results from clinical trials are promising, third
wave cognitive behavioral therapies need to be compared
to already evidence-based treatment for different psychiatric
disorders in larger controlled trials, to prove if this is a
more effective treatment than traditional CBT. Although there
are few clinical studies related to the effect of these new
cognitive behavioral approaches, it is important to acknowl-
edge that the results are promising based on the fact that
these types of therapeutic interventions are rather new.
See also: Anxiety Disorders; Cognition and Personality; Depression;
Psychotherapy; Social Anxiety Disorder; Social Cognition.
Further Reading
Beck AT, Rector NA, Stolar N, and Grant P (2008) Schizophrenia: Cognitive Theory,
Research, and Therapy. New York: Guilford Press.
Beck AT, Rush AJ, Shaw BF, and Emery G (1987) Cognitive Therapy of Depression.
New York: Guilford Press.
Bennett-Levy J, Butler G, Fennell M, Hackmann A, Mueller M, and Westbrook D (2004)
Oxford Guide to Behavioral Experiments in Cognitive Therapy. New York:
Oxford University Press.
Butler AC, Chapman JE, Forman EM, and Beck AT (2006) The empirical status of
cognitive-behavioral therapy: A review of meta-analyses. Clinical Psychology
Review 36: 1 7-31 .
Flaxman PE, Blackledge JT, and Bond FW (2010) Acceptance and Commitment
Therapy: Distinctive Features. New York: Routledge.
Hayes SC, Strosahl K, and Wilson KG (1 999) Acceptance and Commitment Therapy:
An Experiential Approach to Behavior Change. New York: Guilford Press.
Leahy RL (2003) Cognitive Therapy Techniques: A Practitioners Guide. New York:
Guildford Press.
Leidley DR, Marx BP, and Heimberg RG (2005) Making Cognitive-Behavioral Therapy
Work: Clinical Process for New Practitioners. New York: Guildford Press.
Linehan MM (1993a) Cognitive-Behavioral Treatment of Borderline Personality
Disorder. New York: Guilford Press.
Linehan MM (1 993b) Skills Training Manual tor Treating Borderline Personality
Disorder. New York: Guilford Press.
Norton PJ and Price EC (2007) A meta-analytic review of adult cognitive-behavioral
treatment outcome across the anxiety disorders. Journal ot Nervous and Mental
Disease 195(6): 521-531.
Ost LG (2008) Efficacy of the third wave of behavioral therapies: A systematic review and
meta-analysis. Behaviour Research and Therapy 46: 296-321.
Powers MB, et al. (2009) Acceptance and commitment therapy: A meta-analytic review.
Psychotherapy and Psychosomatics 78: 73-80.
Swales MA and Heard HL (2008) Dialectical Behavior Therapy: Distinctive Features.
New York: Routledge.
Wells A (2009) Metacognitive Therapy tor Anxiety and Depression. New York:
Guilford Press.
Relevant Websites
http://contextualpsychology.org/act - Acceptance and Commitment Therapy.
http://www.academyofct.org/ - Cognitive Behavioral Therapy.
http://behavioraltech.org/ - Dialectical Behavior Therapy.
http://www.mct-institute.com/ - Metacognitive Therapy.
Cognitive Bias
A Wilke, Clarkson University, Potsdam, NY, USA
R Mata, University of Basel, Basel, Switzerland
© 2012 Elsevier Inc. All rights reserved.
Glossary
Bounded rationality The principle that organisms have
limited resources, such as time, information, and cognitive
capacity, with which to find solutions to the problems
they face.
Cognitive bias Systematic error in judgment and
decision-making common to all human beings which can be
due to cognitive limitations, motivational factors, and/or
adaptations to natural environments.
Ecological rationality The principle that there is a match
between the structure of information in the environment
and the judgment and decision-making strategies of humans
and other organisms.
Heuristic Judgment or decision-making mechanism or
cognitive shortcut that relies on little information and
modest cognitive resources.
Heuristics and Biases: A Short History of Cognitive Bias
In the early 1970s, Amos Tversky and Daniel Kahneman intro-
duced the term 'cognitive bias' to describe people's systematic
but purportedly flawed patterns of responses to judgment
and decision problems. A term search for 'cognitive bias' in
the Social Sciences Database of ISI Web of Knowledge reveals
close to 4000 hits covering the past 35-year period and an
exponential increase in the usage over time, suggesting that
the term 'cognitive bias' has since gained significant influence
in the psychological and social science literatures.
Tversky and Kahneman's research program - the heuristics
and biases program - addressed the question of how people
make decisions given their limited resources. The program was
inspired by Herbert Simon's principle of bounded rationality.
In the late 1950s, Simon attempted to oppose the idea of
classical rationality, which was concerned mostly with the for-
malization of normative solutions to judgment and decision-
making problems through probability theory and statistics,
with the idea of bounded rationality, which addressed the
specific constraints faced by agents in their environments. For
example, humans have only limited time, information, and
cognitive capacity to decide which mate to choose, food to
eat, or house to buy, and so may have to rely on simple
decision strategies or heuristics to make their decisions. The
heuristics and biases program followed the bounded rational-
ity principle by attempting to identify the specific constraints or
biases associated with human judgment and decision-making.
The heuristics and biases program was inspired by research
on perceptual biases, and proposed that the human cognitive
system like the perceptual system was designed to make infer-
ences about the external world based on imperfect cues that
could lead to errors in some situations. The program thus
generated a straightforward and productive research paradigm,
which can be described as follows. First, participants were
presented with a reasoning problem to which corresponded
a normative answer from probability theory or statistics.
Next, participants' responses were compared with the solution
entailed by these norms, and the systematic deviations (biases)
found between the responses and the normative solutions were
listed. Finally, the biases were explained as the consequence
of the use of heuristics or simple cognitive principles. Using
this strategy, researchers in the heuristics and biases program
have produced an extensive catalog of norm violations.
We present a partial list in Table 1 that spans the judgment
and decision-making, social, and memory research domains.
Naturally, the goal was to provide explanations of these viola-
tions due to reliance on a small set of cognitive principles, the
most popular judgment and decision mechanisms proposed
being representativeness (a judgment is based on how much the
hypothesis resembles available data), availability (a judgment
is based on how easily an example can be brought to mind),
and anchoring-and-adjustment (a judgment is based on a specific
value or anchor and then adjusted to account for other factors) .
The heuristics and biases program represents the most
influential psychological research program to emerge in the
last 40 years, and its merit lies in showing the shortcomings
of classical economic approaches and the value of a bounded
rationality perspective on understanding human judgment.
The heuristics and biases program has, however, been criti-
cized. First, researchers have argued that there are no unequiv-
ocal norms for defining rational judgments and decisions. For
example, there are different concepts of probability espoused
by statisticians and philosophers that imply different norms,
which makes deviations from one hard to interpret as error
or bias. Second, the program has been criticized for presenting
only vague models of human reasoning. For example, the
representativeness, availability, and anchoring-and-adjustment
heuristics proposed by Tversky and Kahneman do not provide
quantitative predictions of people's judgments and it is often
unclear which heuristic is applied under which condition.
Third, the heuristics and biases program has been criticized
for focusing on people's initial responses to judgment problems
rather than providing opportunity for learning from experi-
ence. For example, some anomalies to classical decision theory
are eliminated if people have substantial experience with a
decision problem. Similarly, many classic paradigms in this
tradition involve participants' responses to situations described
in word vignettes, which are not ecologically valid and thus
may offer inadequate insights about everyday decision-making.
531
532 Cognitive Bias
Table 1 Examples of common cognitive biases
Cognitive bias Short description
Confirmation bias
Conjunction fallacy
Endowment effect
Fundamental
attribution error
Gambler’s fallacy
Halo effect
Hindsight bias*
Hot-hand fallacy*
Illusory correlation
In-group bias
Mere exposure effect
The tendency to selectively search for or interpret information in a way that confirms one’s preconceptions or hypotheses
The tendency to assume that specific conditions are more probable than a single general one
The tendency that people often demand more to give up on an object than they would be willing to pay to acquire it
The tendency to overemphasize personal factors and underestimate situational factors when explaining other people’s
behavior
The tendency to think that future probabilities are changed by past events, when in reality they are unchanged (e.g., series of
roulette wheel spins)
The tendency for a person’s positive or negative traits to extend from one area of their personality to another in others’
perceptions of them
A memory distortion phenomenon by which with the benefit of feedback about the outcome of an event, people’s recalled
judgments of the likelihood of that event are typically closer to the actual outcome than their original judgments were
The expectation of streaks in sequences of hits and misses whose probabilities are, in fact, independent (e.g., coin tosses,
basketball shots)
The tendency to identify a correlation between a certain type of action and effect when no such correlation exists
The tendency for people to give preferential treatment to others they perceive to be members of their own group
The tendency by which people develop a preference for things merely because they are familiar with them
Asterisks refer to examples that are discussed in the main text.
This view echoes well Egon Brunswik's argument for the study
of the mind by relying on the informational cues present in
natural environments.
Ecological Rationality: Putting Cognitive Biases in an
Environmental Context
One fundamental criticism of the heuristics and biases pro-
gram is that it has severely neglected the ecology of judgment
and decision processes. The principle of bounded rationality is
deeply associated with the idea that cognitive systems are
fundamentally adapted to their environments - either through
individual learning or by design through natural selection.
Simon illustrated this with a metaphor: mind and environ-
ment as blades of a pair of scissors. Similar thoughts have
been espoused by a number of other theorists. For example,
Roger Shepard saw human vision as reflecting regularities of
the physical world. John Anderson advanced the idea that
memory is structured so as to mimic the probability of infor-
mation occurring in the world and thus being needed by the
organism.
In the late 1990s, Gerd Gigerenzer, Peter Todd, and the
ABC Research Group presented a research program - the fast
and frugal heuristics program - that extended the principle
of bounded rationality and gave new breadth to the idea of
cognitive bias. The fast and frugal heuristics program empha-
sized the principle of ecological rationality, that is, how the
success of reasoning strategies depends on the structure of
the environment. A good example of this principle is demon-
strated by the United Parcel Service (UPS) Right Turn Policy:
UPS, an international shipping company, delivers millions
of packages every year in numerous delivery trucks. The right
turn policy involves carefully mapping out routes for all deliv-
eries to reduce the number of left-hand turns each truck makes,
which helps reduce accidents as well as save fuel, thus max-
imizing overall profits. Naturally, this strategy works well in
the United States and other countries where traffic keeps to the
right. One would predict, however, that the right turn policy
would have the opposite results in countries, such as England,
India, or Hong Kong, where people drive on the left.
The fast and frugal heuristics program has proposed an
alternative research paradigm to the heuristics and biases'
one. The program starts by analyzing the statistical structure
of a specific task environment people face and then - based on
the analysis - derives attributes of the cognitive models of
reasoning that perform well in that environment. In sum, this
program holds that exploring the characteristics of the environ-
ment will contribute to our understanding of what reasoning
processes people follow and when and why these processes
work well.
According to the fast and frugal heuristics program, a cog-
nitive bias is the tendency to solve problems using a particular
cognitive tool or heuristic. Crucially, it sees the selection of a
particular heuristic not necessarily as the product of cognitive
limitations but rather as a bet on the part of the organism
about the structure of the environment in which it finds itself.
One metaphor that guides the fast and frugal heuristics pro-
gram is that of the mind as an adaptive toolbox of simple
decision mechanisms, a repertoire of strategies, with each strat-
egy tuned to a particular environment. A model of mind based
on an adaptive toolbox is, therefore, boundedly rational in the
sense of relying on few cognitive resources, and ecologically
rational in the sense of being tuned to characteristics of natural
environments.
Some have suggested that the differences between the
Heuristics and Biases and the Fast and Frugal Heuristics pro-
grams are not substantive, boiling down to a disagreement
between those that stress that the human mind is fallible and
those who claim that it is often accurate. One clear contribu-
tion of the Fast and Frugal Heuristics program has been, how-
ever, to emphasize the role of environment and specify the
statistical properties of environments that make particular cog-
nitive biases or heuristics successful. In addition, the focus
on ecological rationality has spumed new approaches that
emphasize the role of environment and sampling in deter-
mining adaptive behavior. Specifically, recent approaches are
devoted to understanding the role of sampling in generating
Cognitive Bias 533
bias with less focus on the cognitive apparatus and more on
environmental stimuli. For example, people's risk judgments
of low probability events are often inflated. One possibility is
that such bias is due to selective memory retrieval. However,
an unbiased memory may also produce inflated judgments
of risk due to biased media coverage of natural catastrophes
and accidents. Current and future work on cognitive bias is
concerned with the role of biased sampling in both the external
environment and the internal cognitive apparatus.
Evolutionary Rationality: Understanding Why
Cognitive Biases Occur
The concept of ecological rationality describes the match
between structure and representation of information in the
environment on one side, and the simple decision-making
algorithms such as heuristics on the other. Whenever this
match exists, heuristics can perform well. Evolutionary ratio-
nality holds, however, that it is important to consider the
match between mind and the past environments in which
the mind evolved. In other words, evolutionary rationality
attempts to sketch the evolutionary origins of cognitive bias.
Some evolutionary scientists have followed the Heuristics
and Biases program approach of using errors to study cognitive
bias. The underlying principle behind such research strategy is
that while people can make rapid adaptive decisions using
simple and reliable cues, they are still at risk of making errors.
However, these researchers have tried to introduce the role
of costs to theories of cognitive biases. The argument goes
that eliminating errors altogether is rare, if ever possible, but
the costs associated with certain errors may lead organisms to
systematically commit one type of error over another. This
principle is at the heart of error management theory - a theory
that applies evolutionary logic to signal detection theory. Imag-
ine the problem of reliably identifying a recurrent danger in the
environment such as poisonous snakes. For any given relevant
percept (e.g., a long slender object on the ground), one must
make a decision: snake present or snake absent. Because of the
dire consequences of being bitten by a venomous snake, it is
better to have a low evidentiary threshold for inferring that
long slithering objects are snakes so as to identify every snake
you encounter, than to require too much evidence and occa-
sionally incur a costly surprise. Because both types of errors
cannot be minimized at the same time, asymmetries in the
costs of two types of errors (false positives and false negatives)
should lead systems to be biased in the direction of the least
costly error.
Examples of such biases can be found in auditory percep-
tion. For example, listeners perceive tones with rising intensity
to change faster than equivalent tones falling in intensity -
an effect termed auditory looming. Auditory looming has
also been found to occur in nonhuman primates and is
well explained in an error management theory framework.
The enhanced saliency of rising intensities associated with
approaching objects causes listeners to reliably underestimate
object arrival time. The bias occurs with tones but not broad-
band noise showing some specificity for sound that provides
reliable single-source information and made almost exclusively
by biological organisms. Of course, any time a bias affects
perception of the physical environment, there are risks of mis-
applying it to irrelevant objects that could lead to any variety of
costly errors. The degree to which this is true will largely
determine how advantageous the bias will be, and thus its
impact over evolutionary time. In the case of auditory loom-
ing, the costs of false alarms (e.g., wasting time by being ready
too early) are relatively low compared to the costs of misses
(i.e., not being prepared for an approaching object). The dif-
ference in these costs allows for the selection of a bias that
causes people to systematically overestimate a reliable auditory
cue of movement toward a listener.
Examples of Research on Cognitive Biases
In this section, we introduce two examples of research on
cognitive bias. The first example focuses on search in the exter-
nal world and how people's perceptions of events or their
co-occurrence may be biased toward frequent, natural distribu-
tions. In this example, cognitive bias arises from experimenters
observing an organism's behavior or judgments in environ-
ments that are very atypical compared to those experienced
across phylogenetic and/or ontogenetic time. The second
example focuses on biases in internal search from memory
and emphasizes that cognitive bias may occur both due to
cognitive limitations and motivational factors. For example,
an individual's inaccurate recall of poor past performance
may be due to poor memory and/or a motivation to preserve
a positive view of the self.
Foraging, Hot Hands, and the Structure of the Environment
The work of Andreas Wilke and colleagues on human foraging
behavior in patchy environments, illustrates that an awareness
of ancestral conditions can be key to understanding human
decision-making strategies. When resources are distributed
in patches (i.e., areas with a high density of the resource
surrounded by areas with low density), animals are required
not only to make decisions on where to forage, but also on
how long they should forage in a particular patch as resources
diminish. Biologists have studied simple decision mechanisms
that solve this problem of patch time allocation and identified
resource environments where these mechanisms work well.
Different patch-leaving strategies are necessary because resource
environments differ in how resources are distributed across
patches. The number of resource items within a patch can either
be similar (evenly dispersed distributions), completely random
(Poisson distributions), or some patches may only contain a
few items while others will be very resource rich (aggregated
distributions). Wilke and colleagues tested how well humans
can adapt their patch-leaving behavior when faced with such
resource distributions in a computerized foraging game. The
results showed that participants applied patch-leaving rules
that were particularly appropriate for aggregated environments
also in other types of environments (e.g., those with evenly
dispersed and Poisson distributions). Were research partici-
pants ecologically irrational' 1 .
This finding is less puzzling once one considers that aggre-
gation in space and time, rather than dispersion, is likely to
have been the norm for most of the natural resources humans
534 Cognitive Bias
encountered over evolutionary time. Species of plants and
animals rarely, if ever, distribute themselves in a purely ran-
dom manner in their natural environment, because individual
organisms are not independent of one another: Whereas
mutual attraction leads to aggregation for some species, mutual
repulsion leads to regularity (dispersed environments) in
others. Most often, these deviations from randomness are in
the direction of aggregation, because aggregation offers consid-
erable benefits such as a common habitat, mating and parent-
ing, or the benefits of group foraging. Since humans have been
hunters and gatherers for a very long part of their history, it
could well be that our evolved psychology is adapted to assume
such aggregated resource distributions as the default. Thus,
participants in the foraging experiment might have behaved
evolutionarily rationally.
The idea that humans expect aggregation in space and time
also helps to explain why apparent misconceptions of proba-
bility, such as the hot-hand fallacy, may not reflect fundamen-
tal shortcomings of the human mind but rather adaptation to
the statistical structure of natural environments. The hot-hand
fallacy occurs when research subjects expect lucky streaks in
hits and misses in everything from basketball to coin tosses
when in fact the probabilities of events are independent. When
a basketball player hits many shots in a row, for instance, the
natural expectation is that he has a 'hot hand' and will shoot
another successfully. People are often surprised to discover that
this strong intuition does not square with the reality that the
success of the next shot is determined independently from
the shot before it.
The foraging example presented above hints at an
explanation for the hot-hand phenomenon based on limited
experience with evolutionarily novel events like coin tosses,
and gambling that involve random events. Instead, one can
ask about the structure of objects and events surrounding
important adaptive problems faced by our ancestors, and
what kinds of adaptations might have been shaped by selec-
tion. Evolutionary behavioral scientists would argue that
many of these - plants, animals, human settlements, and
even weather - would have been organized in an aggregated,
clumpy fashion - not perfectly at random (independent) like
those in Las Vegas. Thus, the default human expectation is
aggregation, dumpiness, and nonindependence. To explore
this hypothesis, Wilke devised additional computer tasks in
which the subject could forage for fruits, coin tosses, and
several other kinds of resources, and present them to American
undergraduates and a South American indigenous population
of hunter-horticulturalists (the Shuar). In each population,
subjects exhibited the hot-hand phenomenon for all resource
types, despite the fact that the resources were distributed
randomly by the computer. The one exception found was for
coin tosses for the American students only for which the hot-
hand expectation was reduced though not altogether elimi-
nated. This suggests that the expectation of aggregation in
space and time may be the psychological default that is over-
come only through extensive experience with truly independent
random phenomena like coin tosses. This runs in contrast to the
original explanation offered for the hot-hand phenomenon -
that it is attributable to biased sampling by the mind - and
instead suggests it is a consequence of the minds' adaptation to
the distribution of resources in the natural environment.
Memory Biases: Cognitive and Motivational Determinants
Would humans be better off if we had been blessed with
superior cognitive abilities, such as unfailing memories? One
view on the rather limited cognitive capacities of the human
mind is that limitations, such as forgetting, have functional
significance. Some researchers, like John Anderson, have sug-
gested that the function of memory is not simply to store
information, but rather provide relevant information in spe-
cific situations. According to this view, the human memory
system is organized such that it facilitates the retrieval of infor-
mation that is recent, frequent, and relevant to the current
context. In other words, memory is designed to provide the
information we are most likely to need. Many man-made
information systems are built in such way. For example, com-
puter applications usually incorporate a timesaving feature as
follows: When a user tries to open a document file, the applica-
tions presents a 'file buffer,' a list of recently opened files from
which the user can select. Whenever the desired file is included
on the list, the user is spared the effort of searching through the
file hierarchy. For this device to work efficiently, the applica-
tion must provide the user with the desired file. It does so by
'forgetting' files that are considered unlikely to be needed on
the basis of the assumption that the time since a file was last
opened is negatively correlated with its likelihood of being
needed now. In other words, such a system has a bias toward
information that is likely to be needed.
Although memory systems are very often efficient, they can
sometimes fail because forgetting and sensitivity to contextual
knowledge may lead to systematic error. The hindsight bias is
one of the most frequently cited and researched cognitive
biases in the psychological literature. Hindsight bias is a type
of memory distortion in which, with the benefit of feedback
about the outcome of an event, people's recalled judgments are
typically closer to the outcome of the event than their original
judgments were. Research on hindsight bias is particularly
important because it is a ubiquitous phenomenon and one
with potentially detrimental consequences in applied settings,
such as law and medicine.
In the 1970s, Baruch Fischoff was concerned with profes-
sionals such as clinicians' or politicians exaggerated feeling of
having known all along how patients' recovery or elections were
going to turn out. To study this issue empirically, Fischhoff
asked participants to assess the probabilities of various possible
outcomes concerning upcoming events, for example, President
Nixon's historic trips to China and the Soviet Union (e.g., Pres.
Nixon will meet Chairman Mao; Pres. Nixon will announce that
the trip was a success). After the trips, participants were asked to
recall their predictions. Results showed that participants tended
to exaggerate what they had known in foresight.
There are two common experimental designs that have been
used in the psychological literature. In the memory design, par-
ticipants first make judgments concerning some stimuli, then
receive feedback on some or all of the items, and are finally
asked to recall the original judgments. In the hypothetical design,
participants first receive feedback concerning some or all of the
items and are then asked to say what they would have estimated
had they not been given feedback. Empirical results using either
design have shown that recalled or hypothetical estimates are
commonly biased toward the feedback information.
Cognitive Bias 535
At present, there is no single theory that can explain all
patterns of data and moderator variables that have been
studied in laboratory or real-world settings (e.g., expertise,
experimental materials). One potential reason for this is that
multiple processes are involved in producing the effect. In fact,
there is largely consensus that the bias is multiply determined,
and involves both cognitive and motivational factors.
Regarding cognitive factors, the prevalent idea is that both
processes of retrieval and reconstruction play a role. For exam-
ple, when reporting the original judgment participants are
likely to both try to retrieve the specific memory of the event
as well as reconstruct the original judgment process. Accord-
ingly, the hindsight bias effect can occur by new information
(feedback) biasing ( 1 ) the retrieval cues used to query memory
for the original judgment, (2) the reconstruction of the judg-
ment process, (3) or both. This view also suggests a prominent
role for inhibition processes. Accordingly, research shows that
individuals with strong inhibitory deficits have more difficul-
ties inhibiting feedback about the outcome of an event from
entering working memory and thus show increased hindsight
bias. As expected, this is particularly the case when the correct
response is either in sight or accessible in working memory at
the time of the attempt to recall one's original response.
In addition, there is evidence that hindsight bias may serve
motivational goals. For example, people seem to change the
perceived probabilities of events so that negative events appear
inevitable as a way to mitigate disappointment and personal
blame. However, this seems to occur mostly in situations
people can control and in situations that are unexpected,
suggesting that such phenomena should be interpreted in the
light of people's attempts at preparing for future events. In
other words, these forms of hindsight bias can be seen as
arising from the use of a sense-making process, whereby people
integrate all they know about a topic into a coherent mental
model. In this light, human memory is not so much designed
to accurately reconstruct the past as it is to make sense of it to
better deal with the future.
See also: Cognition and Personality; Defense Mechanisms; Judgment.
Further Reading
Anderson JR and Milson R (1989) Human memory: An adaptive perspective.
Psychological Review 96: 703-719.
Blank H, Musch J, and Pohl RF (2007) Hindsight bias: On being wise after the event.
Social Cognition 25: 1-9.
Fiedler K and Juslin P (2006) Information Sampling and Adaptive Cognition.
Cambridge: Cambridge University Press.
Gigerenzer G, Todd PM, and the ABC Research Group (1999) Simple Heuristics that
Make Us Smart. New York: Oxford University Press.
Gilovich T, Griffin DW, and Kahneman D (2002) The Psychology of Intuitive Judgment:
Heuristics and Biases. Cambridge: Cambridge University Press.
Gilovich T, Vallone R, and Tversky A (1985) The hot hand in basketball: On the
misperception of random sequences. Cognitive Psychology 17: 295-314.
Haselton MG and Buss DM (2000) Error management theory: A new perspective on
biases in cross-sex mind reading. Journal of Personality and Social Psychology
78: 81-91.
Kahneman D, Slovic P, and Tversky A (1982) Judgment Under Uncertainty: Heuristics
and Biases. New York: Cambridge University Press.
Scheibehenne B, Wilke A, and Todd PM (2010) Expectations of clumpy resources
influence predictions of sequential events. Evolution and Human Behavior.
Shepard RN (2001) Perceptual-cognitive universals as reflections of the world.
Behavioral and Brain Sciences 24: 581-601 .
Simon HA (1956) Rational choice and the structure of the environment. Psychological
Review 63: 129-138.
Todd PM, Gigerenzer G, and the ABC Research Group (in press) Ecological Rationality:
intelligence in the World. New York: Oxford University Press.
Tversky A and Kahneman D (1974) Judgment under uncertainty: Heuristics and biases.
Science 185: 1124-1131.
Wilke A and Barrett HC (2009a) The hot hand phenomenon as a cognitive adaptation to
clumped resources. Evolution and Human Behavior 30: 161-169.
Wilke A, Hutchinson JMC, Todd PM, and Czienskowski U (2009b) Fishing for the right
words: Decision rules for human foraging behavior in internal search tasks.
Cognitive Science 33: 497-529.
Cognitive Disorders
M E Berryhill, D Peterson, K Jones, and R Tanoue, University of Nevada, Reno, NV, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cognitive disorders Illnesses in which the primary
symptoms are related to thought, attention, and perception.
Consciousness Awareness of surroundings.
Disorder Root cause of a syndrome.
Disoriented person Someone who is unable to report where
he or she is, why he or she is there, and/or what time it is.
Syndrome Cluster of commonly co-occurring symptoms.
What Are Cognitive Disorders?
Although we never fully know someone else's experience, in
general, people agree on many aspects of the world around us.
In other words, where you are, what day it is, who you are, and
so on are questions that participants in a conversation can
typically agree on. In contrast, when consciousness and thought
processes themselves - problem solving, attention, perception,
and language - are disrupted, patients are said to have cognitive
disorders. Patients with these disorders may be disoriented,
inattentive, have memory and language disturbances, and sen-
sory irregularities including hallucinations. The cognitive disor-
ders, delirium, amnesia, and dementia, are some of the most
devastating diagnoses for at least two reasons. First, these dis-
orders separate patients from his or her normal conscious
states. This makes communication very challenging and it can
make it difficult for caregivers to know if the patient's symp-
toms have been alleviated. Second, diagnoses such as dementia
are progressive and unrelenting indicating that the patient will
not return to his or her former state. In the paragraphs below,
we detail aspects of each of the primary cognitive disorders,
delirium, amnestic disorders, and dementia. These descrip-
tions delineate the criteria set forth in the diagnostic and
statistical manual (DSM-IV-TR), which mental health clini-
cians use to assess psychiatric illness.
How Do We Recognize and Treat Delirium?
Delirium
Delirium is a group of conditions characterized by distur-
bances of consciousness and changes in cognition that occur
over short periods of time, on the order of hours or days.
Delirium can have a rapid-onset and resolve itself so that the
person returns to normal rather quickly. However, because
delirium is secondary to an underlying cause it requires
immediate medical attention. To more precisely distinguish
between delirium brought on by different causes, there are
several different classifications: delirium due to a general med-
ical condition (disease basis such as inflammation, fever, or
infection, malnutrition, dehydration, major stress), substance-
induced delirium (drug withdrawal), delirium due to multiple
etiologies (disease and substance basis), or delirium not other-
wise specified (where the underlying cause cannot be deter-
mined). In delirium, these subclassifications are important in
determining the appropriate course of treatment.
Symptoms
There is no laboratory test for delirium. This makes diagnosis
challenging and attention to the symptomology is critical.
The classic symptoms of delirium include a disturbance of
consciousness accompanied by a change in cognition. In other
words, the patient may be confused, and unable to respond to
simple questions. These symptoms develop over a short period
of time, usually hours to days. Delirium symptoms may fluctu-
ate during the course of the day. If medical attention is quickly
sought, delirium is a temporary condition that will last up to
a few weeks in duration. Patients with delirium must be
monitored for noncompliance with medical directives, such
as removing intravenous tubing. During the delirious episode,
a patient may have moments of lucidity, in which he or she is
aware of what is happening around him or her. Disorientation
is often the first symptom to appear in cases of mild delirium,
and is exhibited by an individual being unable to report
time or place. For example, the patient may not know the day,
the month, or even the year. In addition, attention, and particu-
larly sustained attention, is abnormal in delirium patients. Irrel-
evant objects easily distract individuals with delirium. For
example, they may stare at the flowers in the room rather than
look at the person speaking to them. Cognition also changes in
delirium. In particular, new memory formation is impaired.
A simple test for memory impairment is to ask the individual
to remember several unrelated objects (e.g., apple, clock, tur-
tle) and then to ask him or her to report the objects after several
minutes of conversation. Communication through speech
and language may be disrupted for a variety of reasons. An
individual may have a difficulty articulating words (dysar-
thria), a deficit in naming objects (dysnomia), an impaired
ability to write (dysgraphia), or even language production or
comprehension deficits (aphasia). Symptoms affecting atten-
tion, memory, and orientation may be compounded by sen-
sory abnormalities including hallucinations. For example,
delirium patients may perceive imaginary people in the room
with them. Hallucinations may be in any sensory domain.
Delirium is also associated with disturbances in the sleep-
wake cycle. Sleep-wake cycle disturbances mean that the regu-
lar pattern of daytime alertness and nighttime rest is abnormal.
Symptoms may include daytime sleepiness, nighttime agita-
tion, and difficulty in falling asleep. Patients may sleep during
much of the day and be awake or agitated during the night to
the point where a complete reversal of the night-day sleep-wake
cycle occurs. Patients often report experiencing nightmares.
536
Cognitive Disorders 537
Psychomotor abnormalities are often associated with delirium
such that diagnosed individuals are hyperactive or hypoactive.
It is more common to observe hyperactivity in delirium patients
but decreased activity including sluggishness and lethargy asso-
ciated with hypoactivity may also be witnessed. In the same
individual, psychomotor activity may even shift from one
extreme to the other over the course of a day. The hyperactive
state is more likely to prompt symptoms such as hallucinations.
Finally, emotional states may be disturbed in delirious states.
Individuals diagnosed with delirium may display intense
emotional disturbances associated with anxiety or fear. Other
emotional states include depression, irritability, anger, eupho-
ria, and apathy. There can be rapid and unpredictable shifts
from one emotional state to another. Threatening hallucina-
tions or paranoid delusions often induce a strong fearful
response in the patient and such individuals may attack those
who are falsely perceived as threatening. Alterations of emo-
tional state are most common at night or in the dark because
of the lack of other environmental cues.
When delirium is suspected, electroencephalogram (EEG)
may be used as a confirmatory tool. The EEG signal in delirium
patients reveals slowed processing to sensory inputs (e.g.,
lights, sounds). EEG can also be used to follow recovery over
time. As the patient improves, the EEG signal should return to
normal.
Prevalence
As with all of the cognitive disorders, the prevalence of delir-
ium is not consistent across the population. Indeed, the preva-
lence of delirium in the general population is normally very
low, less than 1% of young adults over 18 years of age and
rising slightly in those over the age of 55. Although rare, certain
populations including the elderly and hospitalized individuals
have a much higher incidence of delirium. In hospitalized
individuals the prevalence of delirium ranges from 10% to
40%, with the elderly being the most likely to be affected.
Disease can also increase the likelihood of delirium. In hospi-
talized cancer patients, as many as one-quarter may be affected.
The numbers are even higher for those with AIDS (30-40%).
Populations with the highest prevalence of delirium are
residents of nursing homes (60%) and individuals diagnosed
with a terminal illness (up to 80%). It is important to note that
the prevalence of delirium depends on many aspects of their
overall health.
Time Course
Delirium usually develops quickly over hours to days. The
disorder usually presents with symptoms such as restlessness,
anxiety, irritability, disorientation, and disturbed sleep. An
individual's delirium may resolve in a few hours to days,
or symptoms may persist for weeks. Elderly individuals who
also are diagnosed with dementia are especially susceptible
to longer lasting delirium and they are less likely to fully
recover. However, treatment is important and failure to treat
the underlying cause of the delirium (e.g., withdrawal, nutri-
tional problems, or disease) may lead to more severe condi-
tions including coma, seizures, and even death. Delirium
can also prompt additional medical complications in elderly
individuals. Complications due to delirium may include
pneumonia and bedsores. These types of complications can
lengthen hospital stays. Patients who develop delirium during
a hospitalization also have a high mortality rate in the months
following discharge.
Treatment
Most individuals diagnosed with delirium achieve a full recov-
ery. However, successful treatment ultimately depends on
the appropriate diagnosis and amelioration of the underlying
cause. Treatment most often involves pharmacological treat-
ment in conjunction with appropriate interventions to relieve
symptoms. Neuroleptics (antipsychotic medications) are the
most common medication used to treat delirium when the
patient is agitated and possibly aggressive. Benzodiazepines
(sleep-inducing, antianxiety medications) are used to combat
withdrawal symptoms. In addition to medications, supportive
therapy involves environmental modifications such as ensur-
ing balanced nutrition, hydration, and creating a controlled
and supportive environment for the patient. Finally, support
from health care professionals and family members are neces-
sary in achieving treatment goals and ensuring patient health.
Current Research Directions
Current research in delirium treatment focuses on two fronts:
the prevalence and consequences of delirium in hospitalized
patients, and new treatments. Recently, a multicenter investi-
gation including over one hundred intensive care units revealed
that delirium was associated with increased mortality rates and
longer hospital stays. The large scale of this study provided an
important step toward assessing the widespread effects of
delirium. Future research focused on delirium as an indepen-
dent risk factor may provide health care professionals with
additional diagnosis criteria and will also substantiate delirium
as a critical health concern.
Progress is being made in therapeutic treatments to im-
prove outcomes following delirium. The use of cognitively
stimulating activities (e.g., reading, writing, word games, puz-
zles, etc.) in the attempt to reduce delirium severity, duration,
and functional loss was studied in postacute care settings. Their
promising results suggest that cognitive stimulation may pro-
mote faster recovery and improve physical function and mental
status. This type of research will help improve our understand-
ing of a disorder whose cause and long-term effects are not yet
fully understood.
How Do We Recognize and Treat Amnestic Disorders?
Symptoms
A second group of cognitive disorders are called amnestic
disorders. These disorders are all associated with impaired
memory processes. Symptoms of amnestic disorders include
specific types of memory impairment, which can be serious
enough to dramatically impair a person's quality of life. Spe-
cifically, certain amnestic disorders prevent patients from explic-
itly learning any new information. This symptom is referred
to as anterograde amnesia. Other patients may have trouble
538 Cognitive Disorders
remembering information from their own past. This deficit is
called retrograde amnesia. Additionally, both types of amnestic
impairments may occur simultaneously. In some instances,
people will spontaneously create stories to fill in missing
details from their memory. This process of adding supporting
details, which never occurred, to memory is known as confab-
ulation. The person does not intend to mislead; he or she is
genuinely not aware that the information is not true. Serious
cases of these disorders can involve impairments in maintain-
ing knowledge of time and space. People suffering from
these disorders may forget where they are or exhibit confusion
of time. In amnestic disorders, however, confusion of time
extends beyond momentary lapses of memory for matters
like the current day of the week to more severe impairments
such as failing to know the current calendar year. Other char-
acteristics of the amnestic disorders may include an apparent
lack of caring about anything, limited emotion when commu-
nicating with others, and general confusion.
Furthermore, in order to be categorized as an amnestic
disorder, social or work-related problems must occur alongside
memory deficits. The manner of measuring the impact these
disorders have on a person's life involves comparing their
current state with their functioning prior to onset of the amnes-
tic disorder. Another important requirement for reaching a
diagnosis of amnestic disorder exists. The amnestic symptoms
must continue beyond a single episode. This distinguishes the
disorder from delirium. The amnestic disorders do not include
disruptions of other cognitive functions, which are related to
dementia. It is also important to note that cultural differences
that place emphasis on memories for certain types of events
should be taken into account. Types of information that a
person ought to know in one culture may have little meaning
for people from a different culture. As such, lack of memories
for culture-specific information outside an individual's cultural
background should not be taken as an indication of the pres-
ence of an amnestic disorder. Finally, in spite of the serious
memory deficits in these patients, other cognitive processes
remain intact. In other words, symptoms are restricted to the
mnemonic realm.
Causes
As with delirium, there are multiple underlying causes of
amnestic disorders, but all causes result in permanent brain
damage. Specifically, amnestic disorders can result from a vari-
ety of physical events affecting the integrity of the brain itself.
Physical causes typically result from three types of circum-
stances. The first of these is related to medical conditions
often resulting from, but not limited to, physical trauma to
the brain. For example, the person may have been involved in
an accident that caused damage to specific parts of the brain.
Vascular problems can also cause brain damage. An example
is stroke, which is caused by insults to blood vessels in the
brain. A second cause of amnestic disorder is due to metabolic
damage to the brain. In these cases, severe vitamin deficiency
can lead to brain damage. This phenomenon is associated with
substance-induced amnestic disorders (see below). A third
cause is disease. Medical conditions causing amnestic disorders
include temporary lack of oxygen (hypoxia), virus (herpes sim-
plex encephalitis), or bacterial infection (bacterial meningitis).
The characteristics of the amnestic disorder depend on the
parts of the brain that are affected and how severely they are
damaged. One of the best known cases of an amnestic disorder
is that of the neuropsychological patient H.M. This patient
suffered from a seizure disorder presumably brought on by a
head injury. To minimize the detrimental effects of this condi-
tion on the brain, the portions thought to be triggering seizures
were surgically removed. The excised structures were located in
regions on the lower left and right sides of the brain known as
the medial temporal lobes. The medial temporal lobes and
related structures include the hippocampi, parahippocampal
gyrus, entorhinal cortex, mammillary bodies, and fornix. Some
of these stmctures, particularly the hippocampi, are necessary
for the formation of long-term memories. On removal of por-
tions of the medial temporal lobe, H.M. experienced dramatic,
but specific memory deficits. He could no longer form new
long-term memories, although short-term and procedural
memory remained intact. In other words, he could remember
his childhood, he could remember the question you just asked
him, and he could remember how to do things like riding a
bicycle, but he could not remember what happened that morn-
ing or a few moments ago. The case of H.M. established that
damage to the medial temporal lobe and surrounding regions
can cause an amnestic disorder as he lost the ability to form
new long-term memories.
Substance-Induced Amnestic Disorders
As noted, amnestic disorders can be caused by substance abuse
and either recreational or prescription medications. Excessive
consumption of alcohol over many years can cause permanent
brain damage. One well-known example is Korsakoff’s syn-
drome, in which severe thiamine deficiency develops because
the patient's calories come primarily from alcohol. Alcohol
does not contain necessary vitamins. This chronic thiamine
deficiency leads to cell death in the mamillary bodies, and
results in permanent memory deficits. People with Korsakoff s
syndrome will confabulate their personal histories in part
because they are unable to remember the past and they cannot
form new memories. These patients have both anterograde and
retrograde amnesia.
Some other substances such as medications (e.g., anticon-
vulsants) may produce side effects that can create amnestic
deficits. Additionally, exposure to environmental toxins such
as industrial solvents, lead, mercury, and carbon monoxide
may alter the typical functioning of the brain in a manner
consistent with amnestic symptoms. In substance-related types
of amnestic disorder, it is essential to prevent reexposure to the
substances.
Time Course
Depending on the cause of the amnestic disorder, the time
course of symptom onset can vary. If the cause is a traumatic
head injury, for example a car accident, the onset of amnestic
symptoms may be instantaneous. In these cases, symptoms
observed after the acute stage immediately after trauma are
generally permanent. In some cases, some symptoms may
subside. On the other hand, the time course of amnestic dis-
orders caused by substances is more variable. In some cases,
Cognitive Disorders 539
the memory symptoms may be temporary, although if the
brain has been permanently damaged, the amnestic disorder
is also permanent. In the case of Korsakoff s syndrome, it may
take many years to develop and only with prolonged alcohol
abuse. Disease processes such as meningitis and encephalitis
can progress very rapidly - over the course of hours and days,
and with early intervention they can be treated with some
success. If treatment is sought early, greater chance of recovery
is possible. If treatment is delayed, these rapidly developing
diseases may quickly lead to permanent amnestic symptoms.
Treatment
Prevention is the best treatment for amnestic disorders. Trau-
matic injuries can be prevented by use of helmets and taking
extra safety precautions. Brain damage due to substance expo-
sure can be entirely avoided. Once a person has experienced
brain damage, the prognosis is very poor for alleviating
amnestic disorders. If the amnesia is caused by a substance
(e.g., alcohol or drug), it is important to prevent continued
exposure. In some cases, treatment in the form of teaching
people suffering from amnestic disorders memory strategies
for improving their day to day functioning may help. Essen-
tially, this type of treatment is designed to establish coping
strategies to maintain function and independence in these
individuals. These may include making lists and altering their
environment to provide external memory cues. For example,
labels placed around the house may help remind the patient
where commonly used objects are found and how to use
them. Additionally, counseling may help people diagnosed
with amnestic disorders to regain a sense of purpose and focus
on existing positive aspects of their life. Substance-induced
amnesties may benefit from drug treatments typically used to
reduce irritability or confusion associated with memory deficits.
Current Research Directions
While no effective treatment currently exists to cure amnestic
disorders, recent research has examined various prospective
treatments. A series of recent endeavors have used animal
models, mainly mice, to investigate the efficacy of newly
identified chemical compounds. Some of these compounds
lead to improved memory performance in animals including
methylene blue (a diaminophenothiazine drug) and others
that are found in nature, including neferine (from Nelumbo
nucifera seeds), Marapuama extract (from a type of wood in
the Amazon), Ficus religiosa Finn (from a fig tree), Murraya
koenigii leaves, Asparagus recemosus, and Bacopa monniera
(a perennial herb). These compounds now require extensive
testing before any drugs will be appropriate for humans with
amnestic disorders.
There are advances in improving the independence of those
with amnestic disorders as well. Recent approaches include
the use of technology and external cues to improve memory
function. The SenseCam'-’ (Microsoft) automatically takes dig-
ital pictures after designated intervals (e.g., every 2 min). The
amnestic user can therefore review the sequence of images as
a reminder of recent events. Similarly, the implementation
of SmartPhone technology allows people with amnestic symp-
toms support for managing everyday tasks. These tasks can be
accomplished by creating reminders of frequent upcoming
events (via alert messages) and memo options available within
most SmartPhone technologies. For example, a person with
amnesia could enable an auditory alert system on his or her
phone signaling him or her to take a dosage of an important
medicine or of an upcoming appointment such as a visit
to the doctor. Another external cue approach involves the
Memory Support System, a calendar and organizational system
shown to help people with memory impairments to be more
self-sufficient.
How Do We Recognize and Treat Dementia?
The third member in the cognitive disorder grouping is demen-
tia. Dementia consists of memory impairments accompanied
by at least one additional cognitive problem. The second cog-
nitive deficit may include dysfunction of language (aphasia),
the loss of motor planning (apraxia), or loss of the ability to
recognize certain objects (agnosia). The cognitive deficits result-
ing from dementia must be severe enough to cause impairment
in social or occupational ability. Dementia is generally a pro-
gressive condition, but it may also be static and rarely in
remittance. Dementia also disrupts cognitive and social func-
tioning in patients.
Prevalence
The overall prevalence of dementia in the adult population
increases with age. The estimates are in the range of <2% for
adults 65-69 years of age. The incidence of dementia increases
rapidly after the age of 85 to somewhere between 16 and 25%.
A diagnosis of dementia in children is rare and difficult
because multiple cognitive deficits must demonstrate docu-
mented deterioration.
Time Course
The DSM-IV criterion for a diagnosis of dementia is on
the basis of a pattern of cognitive deficits beginning with
absentmindedness that becomes more notable over time.
As previously discussed, dementia can be progressive, static,
or in remittance. The degree of disability may depend largely
on the amount of social support available to the patient. This
support allows the patient greater independence for longer
periods of time. An additional factor is the type of cognitive
deficits observed in the patient. Finally, in demented indivi-
duals, severe infections and accidents can be fatal. Dementia is
typically associated with progressive incapacitation, which
requires fulltime monitoring. Individuals with advanced
dementia are typically in residential nursing homes.
Differential Diagnosis
To some extent, dementia treatment depends on the appropri-
ate diagnosis of the underlying cause of symptoms. However,
as there is no cure for dementia, knowledge of the cause
provides for better management of symptoms. The mini-
mental state exam (MMSE) is a commonly used neuropsycho-
logical test that can be used at the patient's bedside to quickly
identify the possibility of dementia. The MMSE asks common
540 Cognitive Disorders
knowledge questions everyone should be able to answer easily.
For example, 'Where are you?' Individuals are scored on a
30-point scale; scores of 25 or higher are considered to be in
the normal range and lower scores are strongly correlated with
dementia. The most important aspect of diagnosis is to ensure
that the patient is actually demented and not suffering from
another cognitive disorder with similar symptoms.
Dementia Arising from Alzheimer’s Disease
The difficulty in diagnosing dementia in patients with Alzhei-
mer's disease (AD) arises from the gradual onset and continual
cognitive decline characteristic of AD. AD is assumed if no
other cause can be found, and when it is clear that the patient
is not suffering from delirium. Other causes such as central
nervous system conditions (e.g., Parkinson's and Huntington's
disease), systemic conditions (e.g., HIV), and other causes
such as head trauma must be ruled out first to receive this
diagnosis. Thus, the majority of dementia cases are of the AD
type. AD patients have great difficulty with explicit memory
tasks, for example repeating a story or remembering a picture.
Visuomotor skills are less affected. The course of AD is progres-
sive; those with the diagnosis generally lose 3-4 points (out of
30-points) on the MMSE per year. The prevalence of AD
increases substantially after the age of 65. The rates increase
rapidly after the age of 65, from ~11%, to more than 26% by
the age of 85. There are also genetic factors related to AD.
Relatives of AD patients are more likely to develop AD than
the general population. In postmortem analysis, AD patients
have neurofibrilary tangles (tangled protein masses) and senile
plaques (protein and cell deposits). These signs of AD are
found throughout the brain, especially in the hippocampus.
In addition, neuronal cell loss is particularly noted in the
temporal and parietal lobes of the brain. Patients with AD are
likely to die within ten years of diagnosis.
Dementia Due to Frontal Lobe Dementia
In frontal lobe dementia, the frontal lobes show atrophy and
cell loss within portions of the brain as in AD. Instead of
dominant memory symptoms, these patients present with
symptoms typical of frontal lobe dysfunction such as problem
solving and planning. Frontal lobe dementia is thought to have
many causes and several variants, including Pick's disease.
Pick's disease is a degenerative disease that affects the frontal
and temporal lobes. Difficulties with memory, motor function,
and other features of dementia materialize later in the course of
the disease. Pick's disease typically begins with personality
changes before memory loss. Individuals may become more
distractible, socially inappropriate, and impulsive. At autopsy,
Pick's disease is confirmed by the presence of Pick bodies (char-
acteristic protein bundles found inside neurons). Frontal lobe
dementias account for an estimated 10% of disease-based
dementia cases. Patients are diagnosed at an earlier age, generally
in the mid-50s and people live for ~ 1 0 years following diagnosis.
Vascular Dementia
Vascular dementia is the second most common form of
dementia after AD. Vascular dementia results from interrupted
blood flow to the brain. Commonly, vascular dementia occurs
after a stroke. Strokes that cause a blockage in arteries within
the brain are called infarctions. Multi-infarct dementia (MID)
is caused by a large number of small strokes, which often go
unnoticed. MID develops gradually because of the cumulative
effects of multiple mini-strokes. A major stroke can lead to
rapid-onset vascular dementia. When vascular dementia is
diagnosed in conjunction with AD, it is referred to as mixed
dementia. Symptoms of delirium and depressed mood can
occur simultaneously in vascular dementia.
Dementia from Other Medical Conditions
A wide range of other conditions can cause dementia. Demen-
tia can result from medical conditions such as HIV, Parkinson's
disease, Huntington's disease, Pick's disease, Creutzfeldt- Jakob
disease (colloquially known as 'mad cow disease'), traumatic
brain injuries, hypothyroidism, brain tumors, or vitamin B12
deficiency. For example, in HIV-related dementia, the virus
destroys the myelinated white matter pathways throughout
the brain. Myelinated white matter tracts enable faster con-
nectivity between regions of the brain and damage to these
tracts disrupts proper intercortical communication. Behavioral
changes resulting from HIV-related dementia include apathy,
social withdrawal, delirium, delusions, forgetfulness, imbal-
ance, slowness, poor concentration, and difficulties with prob-
lem solving. Within AIDS patient populations, more than a
third develop HIV-related dementia by the time of their death.
In these patients, the cerebral cortex is generally intact but
deeper brain structures including the basal ganglia and white
matter tracts show damage.
Dementia commonly follows from Parkinson's disease.
Parkinson's related dementia is characterized by reduced cog-
nitive abilities, executive dysfunction, and impaired memory.
Parkinson's disease is a result of a progressive deterioration of
certain cells in the nervous system as dopamine producing cells
in the substania nigra die. Additionally, dementia can result
from the genetic disorder Huntington's disease. Creutzfeldt-
Jakob disease is caused by a viral infection that causes the
misfolding of proteins. These patients typically develop symp-
toms of dementia as well as involuntary movements. The virus
is transmittable, and can be exchanged between people and
even humans and animals.
Finally, head trauma is a growing public health concern.
The severity of head trauma is correlated with the severity of
the subsequent dementia. Generally speaking, in traumatic
brain injuries there is memory loss. Dementia resulting from
a single head injury is usually not progressive; instead, after the
acute stages of treatment, the dementia symptoms remain
constant. Dementia resulting from multiple head traumas
(e.g., in the sports of boxing or football) leads to more notable
progressive dementias. Repeated traumatic brain injury has
cumulative detrimental effects on the brain and often causes
memory loss and depression.
Treatment
Dementia cannot be reversed as it results from tissue damage
in the brain. The primary treatment is to improve the quality
of life of patients by making life more simple and comfort-
able. Medical treatments are aimed at improving symptoms
Cognitive Disorders 541
Table 1
Comparison between the cognitive disorders:
delirium, amnestic disorders, and dementia
Delirium
Amnestic disorders
Dementia
Symptoms
•
Disturbed consciousness
•
Impaired memory
•
Memory impairments and other
•
Disorientation
•
Failure to remember old or new information
symptoms
•
Speech and language
•
Agnosia
disturbances
•
Apraxia
•
Memory impairment for recent
•
Aphasia
events
Age of onset
•
All ages
•
Trauma: all ages
•
Elderly, especially over 85 years old
•
Substance abuse: older middle age
Time course
•
Rapid-onset
•
Following head trauma
•
Progressive over decades
•
Resolution after hours or days
•
Prolonged onset when due to substance abuse
•
Permanent
•
Permanent
Causes
•
Medical condition
•
Physiological general medical condition
•
Disease process
•
Substance induced (e.g., toxin
(e.g., head trauma)
•
Reduced blood flow
exposure)
•
Substance induced (e.g., toxin exposure)
•
Vitamin B12 deficiency
•
Hypothyroidism
Treatment
•
Supportive therapy
•
Technology (e.g., smartphones, SenseCam®)
•
Treat symptoms
•
Pharmacological
•
Memory strategies
•
Supportive therapy
•
Pharmacological
in dementia patients. Recently, several cholinesterase inhibi-
tor drugs have been approved for AD treatment and are used
in other dementia populations. This class of drugs prevents
the decomposition of the neurotransmitter acetylcholine
and shows evidence of slowing the memory decline in AD.
In vascular dementia, prevention of further vascular pro-
blems is a priority. Treatment targets conditions such as
high blood pressure, diabetes, and high cholesterol. Antide-
pressants are often prescribed for those who also have
depressive episodes; sedatives are prescribed to treat agita-
tion. Education of family members or caregivers toward
strategies that make life more comfortable for those with
dementia is beneficial.
Current Research Directions
Dementia is a topic of considerable research, in particular
because as our population ages, the number of individuals
diagnosed with dementia will also increase. Currently, five
approved drugs inhibit acetylcholinesterase to elevate the
level of acetylcholine in the synapse. However, a recent review
of these drugs reveals that the effectiveness of these drugs is
questionable. Patients on these drugs improve only marginally.
Selegiline, a drug that is used to treat Parkinson's, is often used
as a treatment for dementia although studies have found no
long-term benefit. Several new compounds, including soybean-
derived phosphatidylserine, show some improvement in those
with mild cognitive impairments. The improvements were seen
mostly in tests of delayed verbal recall, which may be effective at
improving the quality of life of elderly dementia patients. Recent
findings suggest that alternative treatments may also be benefi-
cial in treating dementia. In one study, elderly patients suffering
from dementia were incorporated in everyday Montessori clas-
ses with preschool-aged children. Dementia patients were
involved in teaching the normal lessons and playing along
with the children. Taking part in the activities was positively
correlated with positive engagement. Music therapy seems to
help treat some symptoms of dementia, although statistically
significant approaches have not yet been confirmed.
Differential Diagnoses of Cognitive Disorders
It can often be difficult to distinguish between the cognitive
disorders (see Table 1). All of them include disorientation and
confusion. Yet, proper diagnosis is essential for appropriate
treatment. There are important guidelines that can improve
diagnostic accuracy. For example, if the patient does not have
disturbed consciousness, the patient does not have delirium.
The diagnosis of delirium follows when a patient has a rapid-
onset and episodes of disturbed consciousness and fluctuating
cognitive symptoms. Although symptoms of delirium fluctu-
ate regularly, symptoms of dementia and amnestic disorder
are more stable. If the patient is conscious, and only has
memory problems, the diagnosis is amnestic disorder. Amnes-
tic disorder involves severe memory impairment, which can
be seen in dementia but does not involve cognitive deficits
such as apraxia, agnosia, aphasia, or other executive function
deficits. Finally, if the patient is conscious, shows evidence of
memory problems, and has other cognitive disruptions, the
diagnosis is dementia. Symptoms of dementia may persist for
months unlike those seen in delirium. Unfortunately, there
are no magic cures for the cognitive disorders. Future research
is needed to treat these burgeoning public health problems
and to improve the quality of life for patients and their
families.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of Mental
Disorders, 4th edn., text revised. Washington, DC: American Psychiatric
Association.
Morgan JE and Ricker JE (eds.) (2008) Textbook of Clinical Neuropsychology (Studies
on Neuropsychology, Neurology, and Cognition). New York: Psychology Press.
542 Cognitive Disorders
Trzepacz P, Breitbart W, Levenson J, Franklin J, and Martini DR (1999) Practice
guideline for the treatment of patients with delirium. American Journal of Psychiatry
156(5 Supplement): 1-20.
Relevant Websites
http://www.alz.org - Alzheimer’s Association.
http://www.alz.org/professionals_and_researchers_11176.asp - Alzheimer’s
Association Dementia Care Training.
http://www.europeandeliriumassociation.com/ - European Delirium Association.
http://www.mayoclinic.com - Mayo Clinic.
http://www.nlm.nih.gov/medlineplus- Medline Plus.
http://www.stroke.org - National Stroke Association.
http://www.ninds.nih.gov/disorders/dementias/dementia.htm - NINDS - Dementia
Information Page.
Cognitive Dissonance Theory
E Harmon-Jones, Texas A&M University, College Station, TX, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Action-based model A theoretical reinterpretation of
the original theory of cognitive dissonance that posits
that the cognitions most likely to cause dissonance are those
cognitions with opposing action implications. Inconsistent
cognitions that have no likelihood of influencing the
individual's behavior should not arouse dissonance.
Effort justification According to dissonance theory,
individuals should experience dissonance whenever they
engage in an effortful activity to obtain an outcome.
Dissonance occurs because the cognition that the activity is
unpleasantly effortful is inconsistent with engaging in the
activity. Dissonance can be reduced by exaggerating the
desirability of the outcome, which would add consonant
cognitions.
Induced compliance paradigm Experimental paradigm
used by psychological researchers interested in studying
cognitive dissonance processes. This paradigm has been
used to test the hypothesis that dissonance should be
aroused when a person acts in a way that is contrary to his or
her attitudes. The amount of dissonance experienced will be
inversely related to the amount of external pressure or
justification for behaving in that manner. For example, an
individual who is forced by an authority figure to behave
contrary to his/her attitude should experience less
dissonance than an individual who was subtly coerced into
behaving contrary to the attitude. If the external pressure is
low, individuals may change their attitudes to be more
consistent with the counter-attitudinal behavior.
Misattribution paradigm An experimental paradigm used
by cognitive dissonance researchers to test whether
dissonance-related attitude change was indeed motivated by
negative emotional feelings, as the theory proposed. In this
paradigm, dissonance is created in standard ways but some
subjects are led to believe that their negative emotional
feelings created by dissonance are actually caused by another
source, such as the bright lights in the room or a pill they
had just consumed. When individuals believe that their
dissonance-created negative feelings are caused by another
source, they do not show the typical attitude change that
occurs as a result of dissonance.
New look A theoretical reinterpretation of the original
theory of cognitive dissonance that posited that cognitive
inconsistency was not the cause of dissonance but that
feeling personally responsible for producing a foreseeable
negative consequence was the cause. This reinterpretation
was applied primarily to the induced compliance paradigm,
which is the main experimental paradigm used to test
dissonance theory.
Self-affirmation A theoretical reinterpretation of the
original theory of cognitive dissonance. This reinterpretation
posits that dissonance is not the result of cognitive
inconsistency or self-inconsistency, but is instead the result
of a threat to one's self image.
Self-consistency A theoretical reinterpretation of the
original theory of cognitive dissonance. Instead of positing
that any type of cognition can be involved in creating
dissonance, this reinterpretation posits that only cognitions
involving the self-concept will arouse dissonance. According
to this view, dissonance only occurs when one behaves
contrary to his/her self-view.
Spreading of alternatives According to research derived
from cognitive dissonance theory, after individuals make
difficult decisions, they often perceive their chosen
option as more positive and their rejected option as more
negative than they perceived these options prior to the
decision, when the options were perceived as similar in
value. In other words, the individuals spread apart their
attitudes toward the decision options from predecision to
postdecision.
Cognitive dissonance theory and research dominated social
psychology from the 1950s until the 1970s. The theory revolu-
tionized thinking about psychological processes, particularly
regarding how rewards influence attitudes and behavior, and
how behavior and motivation influence perception and cogni-
tion. In the 1990s, research on the theory was revived and it
has since been gaining in interest. In addition to the explicit,
renewed interest in the dissonance theory itself, the theory has
had implicit influence on much of contemporary theorizing.
The Original Version of the Theory
Leon Festinger formulated the original theory of cognitive dis-
sonance in the mid-1950s, and the first formal and complete
presentation of the theory appeared in 1957. The theory was
presented in abstract terms and, consequently, it has been used
to understand a variety of phenomena, as is shown below.
Festinger theorized that when an individual holds two or more
elements of knowledge that are relevant to each other but incon-
sistent with one another, a state of discomfort is created.
He called this unpleasant state 'dissonance.' Festinger theo-
rized that the degree of dissonance in relation to a cognition =
D/(D + C), where D is the sum of cognitions dissonant with a
particular cognition and C is the sum of cognitions consonant
with that same particular cognition, with each cognition
weighted for importance. Several theorists have proposed that
the dissonance between cognitions could be determined by
assessing whether a person expects one event to follow from
another.
543
544 Cognitive Dissonance Theory
Festinger theorized that persons are motivated by the un-
pleasant state of dissonance and that they may engage in 'psy-
chological work' to reduce the inconsistency. This work will
typically be oriented around supporting the cognition most
resistant to change. To reduce the dissonance, individuals
could add consonant cognitions, subtract dissonant cognitions,
increase the importance of consonant cognitions, or decrease the
importance of dissonant cognitions. One of the most often
assessed ways of reducing dissonance is change in attitudes.
Attitude change is expected to be in the direction of the cognition
that is most resistant to change. In tests of the theory, it is often
assumed that the knowledge about recent behavior is usually
most resistant to change, because if a person behaved in a certain
way, it is often very difficult to undo that behavior. Thus, attitude
change would be consistent with the recent behavior.
Festinger presented dissonance theory during the heyday of
cognitive consistency theories, and he discussed dissonance as
a cognitive consistency theory. However, as several subsequent
theorists noted, dissonance theory contained an important
element not present in other consistency theories. That is,
dissonant and consonant cognitions are defined in relation to
a particular focal or generative cognition. Moreover, this focal
cognition is usually related to a behavior, a point to which we
return later.
Experimental Paradigms Used to Test the Theory
Free Choice
In 1956, Jack Brehm examined dissonance theory's predictions
for postdecision processing. According to the theory, after a
decision, all of the cognitions that favor the chosen alternative
are consonant with the decision, whereas all the cognitions
that favor the rejected alternative are dissonant. The greater the
number and importance of dissonant cognitions and the lesser
the number and importance of consonant cognitions, the
greater is the degree of dissonance experienced by the individ-
ual. In a decision situation, dissonance is typically greater the
closer the alternatives are in attractiveness, as long as each
alternative has several distinguishing characteristics. Disso-
nance caused by a decision can be reduced by viewing the
chosen alternative as more attractive and/or viewing the
rejected alternative as less attractive; this particular pattern of
attitude change has been referred to as spreading of alternatives.
Brehm conducted an experiment in which participants
made either an easy or a difficult decision between two alter-
natives. The difficult decision was one in which the alternatives
were close in attractiveness, whereas the easy decision was one
in which one alternative was much more attractive than the
other. Participants were asked to evaluate the decision options
before and after the decision. Brehm found that, after persons
made a difficult decision, they changed their attitudes to
become more negative toward the rejected alternative (and
slightly more positive toward the chosen alternative). After an
easy decision, participants did not change their attitudes. Some
scientists have questioned whether this paradigm actually pro-
vides support for a dissonance theory interpretation. However,
research has dealt with these challenges and provided evidence
that difficult decisions cause spreading of alternatives for the
reasons outlined by dissonance theory.
Induced Compliance
In 1959, Leon Festinger and Merrill Carlsmith reported an
experiment to test the hypothesis that dissonance should be
aroused when a person acts in a way that is contrary to his or
her attitudes. To test this prediction, they brought participants
into the laboratory and asked them to perform a boring task.
Then, participants were paid either $1 or $20 to tell 'another
participant' that the task was interesting. According to disso-
nance theory, lying for a payment of $20 should not arouse
much dissonance, because $20 provides sufficient justification
for the counter-attitudinal behavior (i.e., it adds 20 cognitions
consonant with the behavior). However, being paid $1 for
performing the same behavior should arouse much disso-
nance, because $1 provides just enough justification for the
behavior (i.e., it adds only one consonant cognition). As
expected, participants in the $1 (low-justification) condition
changed their attitudes to be more positive toward the task,
whereas participants in the $20 (high-justification) condition
did not change their attitudes.
Effort Justification
Dissonance is aroused whenever a person engages in an
unpleasant activity to obtain some desirable outcome. From
the cognition that the activity is unpleasant, it follows that one
would not engage in the activity; the cognition that the activity
is unpleasant is dissonant with engaging in the activity. Disso-
nance should be greater, the greater the unpleasant effort
required to obtain the outcome. Dissonance can be reduced
by exaggerating the desirability of the outcome, which would
add consonant cognitions.
In the first experiment designed to test these ideas, in 1959,
Aronson and Mills had women undergo a severe or mild
'initiation' to become a member of a group. In the severe
initiation condition, the women engaged in an embarrassing
activity to join the group, whereas in the mild initiation condi-
tion, the women engaged in an activity that was not very
embarrassing to join the group. The group turned out to be
dull and boring. The women in the severe initiation condition
evaluated the group more favorably than the women in the
mild initiation condition. This paradigm continues to be used
fruitfully in research, and it has been usefully applied in
psychotherapy.
Other experimental paradigms have been used to test the
theory but they are used less frequently and because of space
limitations are not described here.
Challenges to the Research and Original Theory
After these and other dissonance results appeared in the litera-
ture, some theorists began to question whether the results were
due to motivation. Some theorists hypothesized that the effects
were due to nonmotivational, cognitive processes or impres-
sion management concerns. However, subsequent research
confirmed that dissonance is a motivated process. That is,
research revealed that during the state of dissonance, indivi-
duals evidence heightened electrodermal activity (which is
associated with activation of the sympathetic nervous system)
and report increased negative affect. After discrepancy is
Cognitive Dissonance Theory 545
reduced (attitude change occurs), self-reported negative affect
is reduced. However, electrodermal activity does not decrease
at this point unless individuals are distracted from the cogni-
tive discrepancy. It is possible that the arousal following atti-
tude change is the result of a motivation to follow through with
the commitment.
Moreover, research using the misattribution paradigm has
revealed that discrepancy reduction is motivated by the need to
reduce negative affect. In the misattribution paradigm, partici-
pants are provided a stimulus (e.g., a placebo) that is said to
cause specific side effects. It is assumed, following Schachter and
Singer's two-factor theory of emotion, that individuals may mis-
takenly attribute their dissonance arousal to this other source,
but only when the expected side effects of the other source are
similar to the state produced by dissonance. The nature of the
internal state can then be inferred indirectly by determining
the type of stimuli to which individuals misattribute the state
aroused by dissonance. In this paradigm, participants are
exposed to treatments that will or will not arouse dissonance,
and then they are either provided or not provided a possible
external cause for their experienced state. Research has indicated
that participants in the dissonance arousing conditions will
misattribute their arousal to stimuli that are said to cause nega-
tive affective side effects and that once this misattribution is
made, dissonance reduction (attitude change) does not occur.
Beginning in the late 1960s, researchers began to propose
motivational explanations for dissonance effects that differed
from Festinger's originally proposed theory. Three revisions
of dissonance theory have been proposed, and their origina-
tors have provided evidence to support these conceptions.
These include Aronson's self-consistency theory, Steele's
self-affirmation theory, and Cooper and Fazio's new look at
dissonance.
Self-Consistency
In his self-consistency theory, Elliot Aronson proposed that dis-
sonance is due not merely to an inconsistency between cogni-
tions. Instead, he posited that dissonance occurs when a person
acts in a way that violates his or her self-concept, that is, when
a person performs a behavior inconsistent with his or her sense
of self. Because most persons have a positive self-concept, disso-
nance is most often experienced when a person behaves nega-
tively, behaving in an incompetent, irrational, or immoral
manner. One of the primary predictions derived from this revi-
sion is that low and high self-esteem individuals should respond
with less and more dissonance reduction (e.g., attitude change),
respectively, because in dissonance experiments high self-esteem
individuals are induced to act in ways that are more discrepant
from their positive self-views. Experiments testing this prediction
have produced mixed results. Also, Beauvois and Joule obtained
results that are difficult to explain with this revision.
Self-Affirmation
Claude Steele proposed a different alternative to Festinger's
dissonance theory. He proposed that persons possess a motive
to maintain an overall self-image of moral and adaptive ade-
quacy. He stated that dissonance-induced attitude change
occurs because dissonance threatens this positive self-image.
Whereas Festinger's dissonance theory posited that individuals
are motivated to reconcile inconsistent cognitions, Steele pro-
posed that, instead, individuals are merely motivated to affirm
the integrity of the self. In support of this idea, Steele presented
experiments, where, following a dissonance induction, partici-
pants either were, or were not, presented with an opportunity
to affirm an important value. When participants were allowed
to affirm an important value, dissonance-related attitude
change did not occur.
However, Simon, Greenberg, and Brehm presented evidence
supporting an alternative explanation for Steele's findings that
was in line with the original theory of dissonance. Festinger's
original theory proposed that the degree of dissonance experi-
enced depended upon the importance of the dissonant and
consonant cognitions. Simon et al. hypothesized that making
an important value salient could reduce dissonance by reducing
the individual's perception of the importance of the dissonant
act. They conducted an experiment in which participants who
opposed a tuition increase were given a high choice to write
essays in support of a tuition increase (a counter-attitudinal
statement). After writing the essay, participants were either
given an opportunity to affirm an important value (self-
affirmation condition), or asked to write about a value that
was not important to them personally but was of general impor-
tance (value salient condition, e.g., world hunger), or neither
(control condition). Participants were then asked to rate the
degree to which they supported a tuition increase. Participants
in the control condition changed their attitudes to be more
favorable toward a tuition increase, as expected. Participants
in both the self-affirmation and value-salient conditions did
not change their attitudes. They had trivialized, or reduced the
importance of, the tuition increase issue by thinking about
other important values, even when these values were not
personally important and thus not self-affirming. Other evi-
dence has been presented that is difficult to interpret in
self-affirmation theory terms.
In more recent work testing his self-standards model of disso-
nance, Jeff Stone has found that individuals with low self-esteem
show less attitude change following induced compliance if their
personal self-standards were primed (by rating their personal
ideal for themselves on untrustworthy, precise, and ethical traits)
immediately after the writing of the counter-attitudinal essay.
When normative standards (by rating what their peers thought
they ought to be on untrustworthy, precise, and ethical traits)
or no particular standards were primed, participants with low
self-esteem showed the same amount of attitude change as parti-
cipants with high self-esteem. Stone suggested that "for self-
consistency to operate in dissonance, something in the context
must make idiosyncratic self-knowledge accessible. Otherwise,
dissonance processes are not necessarily moderated by individ-
ual differences in the structure and content of self-knowledge"
(2003: 852). Stone suggested that these results cast doubt on
both self-affirmation and self-consistency theories, and he pro-
posed that both the self-affirmation and self-consistency models
are correct, but under different conditions.
New Look
Joel Cooper and Russell Fazio proposed the idea that the
discomfort experienced in dissonance experiments was not
546 Cognitive Dissonance Theory
due to an inconsistency between the individual's cognitions,
but rather to feeling personally responsible for producing an
aversive consequence. They stated, "Dissonance has precious
little to do with the inconsistency among cognitions per se,
but rather with the production of a consequence that is
unwanted." In support of this idea, in 1970, Cooper and
Worchel replicated and extended Festinger and Carlsmith's
classic experiment. In addition to the conditions of the original
experiment. Cooper and Worchel added conditions in which,
when the participant tells the confederate that the boring
task is interesting, the confederate is not convinced by the lie.
They found that attitude change occurred only in the low-
justification condition where the confederate believed the lie.
This result and others have been interpreted as indicating that
dissonance-related attitude change occurs only when indivi-
duals feel personally responsible for producing an aversive
consequence. The new look, or aversive consequences, revision
of cognitive dissonance theory was widely accepted.
However, the results obtained in paradigms other than the
counter-attitudinal action paradigm are not consistent with
the aversive consequences model. Dissonance research using
a selective-exposure paradigm has demonstrated that persons
are more willing to examine materials that confirm their beliefs
than materials that dispute their beliefs. Research using a belief
disconfirmation paradigm has shown that, when persons are
exposed to information that challenges their beliefs, they often
strengthen their original belief. Research using a hypocrisy
paradigm has shown that persons change their behavior to be
more in line with their beliefs when they are reminded of times
when they did not live up to their beliefs. It is difficult to
reconcile any of these lines of dissonance research with a
conception of dissonance theory in which the production of
an aversive consequence is the only motivator of dissonance-
related attitude change.
According to the original theory of cognitive dissonance,
the production of aversive consequences would be expected
to increase the amount of dissonance produced because an
aversive consequence may be an important dissonant cogni-
tion. However, the original theory would deny that an aver-
sive consequence is necessary to produce dissonance. In the
induced-compliance experiments testing the necessity of aver-
sive consequences, attitude change may have occurred only
when participants caused aversive consequences for a number
of reasons. First, the lack of attitude change in the no-aversive
consequences conditions is a null effect. Null effects are diffi-
cult to explain and subject to multiple alternatives. Second,
attitude change may have been produced, but may have been
too slight to be detected with the small sample size of these
experiments. Third, not enough dissonance may have been
aroused in these experiments to produce attitude change with-
out the additional help of an aversive consequence. For exam-
ple, too much justification for the counter-attitudinal behavior
may have been provided. Fourth, in these experiments, disso-
nance may have been produced in the no-aversive conse-
quences conditions, but may have been reduced by a route
other than attitude change.
To examine whether attitude change could occur in an
induced compliance setting in which aversive consequences
were not produced, Harmon-Jones and colleagues conducted
several experiments. Under the pretext of participating in an
experiment on memory, participants were exposed to an atti-
tudinal object. They were assured of privacy and anonymity,
and then given high or low choice to write a counter-attitudinal
statement about the object (to manipulate justification). They
were asked to discard the statement in the trash after writing
it so that there was no chance of the statement causing an
aversive consequence. This manipulation was on the basis of
Cooper and Fazio's statement, "making a statement contrary
to one's attitude while in solitude does not have the potential
for bringing about an aversive event" (1984: 232). Other
experiments revealed that dissonance produced in this experi-
mental paradigm caused more nonspecific skin conductance
responses and greater self-reported negative affect.
These results demonstrate that dissonance affect and
dissonance-related attitude change can occur in situations in
which a cognitive inconsistency is present but the production
of aversive consequences is not present. They also demonstrate
that the experience of cognitive dissonance evokes an unpleas-
ant motivational state that motivates dissonance reduction.
These experiments have supported the original conception of
dissonance theory over the revisions. But why does dissonance
evoke this negative motivational state? Why is inconsistency
aversive? Festinger simply proposed that inconsistency was
aversive but never thoroughly explained or tested why incon-
sistency was aversive.
Action-Based Model of Dissonance
To address these questions, Harmon-Jones proposed an action-
based model of dissonance. The model concurs with other areas
of psychological research in proposing that perceptions and
cognitions can serve as action tendencies. The model further
proposes that dissonance between 'cognitions' evokes an aver-
sive state because it has the potential to interfere with effective
and unconflicted action. Dissonance reduction, by bringing
'cognitions' into consonance, serves the function of facilitating
the execution of effective and unconflicted action. The action-
based model uses the term 'cognitions' to be consistent with the
language of the original theory. However, for the model, the
more accurate term for the psychological construct involved in
dissonance is action tendency. That is, it is the inconsistency
between important action tendencies that causes dissonance.
This line of thinking is consistent with the thinking of past
dissonance theorists who emphasized behavior as the focal or
generative cognition. The action-based model extends these
past views by suggesting that the other (nongenerative) cogni-
tions most likely to arouse dissonance are those that have
action tendencies inconsistent with the generative cognition.
The action-based model proposes that inconsistency be-
tween cognitions makes persons uncomfortable because incon-
sistency has the potential to interfere with effective action. From
the viewpoint of the action-based model, cognitions are impor-
tant because they guide the actions of an organism. When an
individual holds two relatively important cognitions that are
inconsistent, the potential to act in accord with them is under-
mined. To reduce the inconsistency and resulting negative
affect, individuals engage in a variety of cognitive strategies.
For example, dissonance results when one 'freely chooses'
to engage in behavior that is inconsistent with an attitude or
Cognitive Dissonance Theory 547
belief. The 'free choice' is subtly induced by the experimenter
in experimental research. Experiments have shown that when
individuals engage in such behavior, they often change their
attitudes to be consistent with their recent behavior, as discussed
previously. After making difficult decisions, individuals value the
chosen alternative and devalue the rejected alternative more than
they did prior to the decision. In both of these dissonance-
evoking situations, dissonance occurs because there are cogni-
tions that are inconsistent with a chosen course of action. That is,
in the former situation, the past attitude is inconsistent with the
current behavior. In the latter situation, the positive aspects of
the rejected alternative and the negative aspects of the chosen
alternative are inconsistent with the decision. The dissonance
thus has the potential of interfering with the translation of the
decision into effective action. According to the action-based
model, attitude change produced by dissonance is the result of
following through with the commitment to the behavior. The
attitude change is posited to be one of a number of processes that
would assist with the translation of the commitment into effec-
tive and unconflicted action. Thus, according to the action-based
model, dissonance evokes a negative affective state that signals
the organism that something is wrong and motivates the organ-
ism to engage in behavior to correct the problem. The correction
of the problem often involves following through with the com-
mitment to the behavior or decision. This view of dissonance is
consistent with past as well as present theorizing on the function
of dissonance and dissonance reduction.
Other scientists have advanced similar but not identical
conceptions. For instance, in 1951, Lewin discussed the organ-
ism's capacity to 'freeze' upon an action tendency following
a decision. In 1967, Jones and Gerard discussed the concept
of an unequivocal behavior orientation that was described as
an adaptive strategy that forced the individuals to bring their
relevant cognitions into harmony with each other. Another
perspective consistent with the present model is Julius Kuhl's
theory of action control. He proposed that to insure that the
intended action rather than a competing action tendency will
be executed, the intended action tendency has to be selectively
strengthened and protected against interference until the
action is executed. The postdecisional spreading of decision
alternatives may serve the function of putting the decision into
action. When one considers that dissonance is primarily a
theory about postdecisional processing, it is easy to see how
these theories fit with the present conception of the function
of the dissonance process.
Experimental Tests of the Action-Based Model
Action-Orientations
An action-oriented state is a state that often occurs following a
decision. When a person is in an action-oriented state, imple-
mentation of decisions is enhanced. In 2002, Harmon-Jones
and Harmon-Jones integrated these ideas with dissonance
theory to propose that this action-oriented state that follows
decision-making is equivalent to the state in which dissonance
motivation operates and dissonance reduction occurs. They
hypothesized that experimentally manipulating the degree of
action-orientation experienced following a decision should
affect the degree of dissonance reduction.
Harmon-Jones and Harmon-Jones conducted an experi-
ment to test these ideas. Participants were asked to make either
an easy decision or a difficult decision, choosing to perform
one of two physical exercises that the participant rated attrac-
tive and unattractive or rated as very similar in attractiveness.
Participants were asked to fill out a mindset questionnaire after
the decision. The neutral mindset questionnaire asked partici-
pants to list seven things they did in a typical day, while the
action-oriented mindset questionnaire asked participants to
list seven things they could do to perform well on the exercise
they had chosen. Participants were then asked to re-evaluate
the exercises. Results indicated that participants in the high-
choice, action-oriented condition changed their attitudes to
prefer the chosen exercise more than participants in the other
conditions.
In a second experiment testing the action-based model,
Harmon-Jones and Harmon-Jones replicated the results of
the first experiment using a different manipulation of action-
orientation. In this experiment, action-orientation was in-
duced by asking participants to think of an important decision
that they had made and to list the steps they intended to use to
successfully follow through with their decision. The partici-
pants in the action-orientation condition engaged in more
attitude change following a difficult decision than did partici-
pants in the comparable conditions. This study replicated the
results of the previous study, but provided stronger support
for the model because it used an action-orientation induction
unrelated to the decision at hand.
Neural Processes Involved in Dissonance
To assist in translating the behavioral commitment or intention
into effective action, approach-motivational processes should
be activated, as the individual works to successfully implement
the new commitment. Thus, the increase in approach motiva-
tion should activate the left frontal cortex, as much past re-
search has found this cortical region to be involved in
approach-motivational processes.
Other research has suggested that activity in the anterior
cingulate cortex is involved in monitoring the occurrence of
errors or the presence of response conflict. Importantly, research
has found increased anterior cingulate cortex activity, as
measured by the event-related potential known as the error-
related negativity, when behavior conflicts with the self-concept.
This finding suggests that even higher level conflicts, the type
with which dissonance theory has been most concerned, also
activate the anterior cingulate cortex. More recently, Vincent van
Veen and colleagues found increased anterior cingulate cortex
activation in the induced compliance paradigm.
Based on this past research, dissonance (or potential
response conflict) first activates the anterior cingulate cortex,
and then activates the left dorsolateral prefrontal cortex, which
assists in resolving the conflict. To test the prediction that
dissonance is associated with increased left frontal cortical
activity, university students who were opposed to a tuition
increase participated in a study ostensibly concerned with
attitudes and personality. They were randomly assigned to
one of two choice conditions. In the low-choice condition,
participants were told they were to write an essay supporting
a 10% tuition increase at their university. In the high-choice
548 Cognitive Dissonance Theory
condition, participants were told that writing the essay in favor
of the tuition increase was their choice and completely volun-
tary. However, the instructions subtly encouraged them to
write such an essay. EEG was assessed for 1 min following the
beginning of the writing of the counter-attitudinal essay, as
past research has revealed that dissonance is greatest at this
point in time. Moreover, the commitment alone (and not the
complete essay writing) is sufficient to evoke dissonance. Then,
participants completed an attitude measure. Replicating past
research, results revealed that high-choice participants changed
their attitudes more than low-choice participants. Supporting
the primary prediction, results also revealed that high-choice
participants evidenced greater relative left frontal activity than
low-choice participants.
Another experiment extended the above research by testing
the hypothesis that an action-oriented mindset would not only
facilitate discrepancy reduction following a decision, but
would also increase relative left frontal cortical activity. Repli-
cating past results, the current experiment demonstrated that
the action-oriented mindset caused greater spreading of alter-
natives than the neutral mindset condition. Moreover, the
action-oriented mindset caused greater relative left frontal acti-
vation than the other condition. These results supported our
predictions and suggest that the dissonance reduction involves
activation of the left frontal cortex and approach-motivational
processes.
In the previous experiment, the psychological process
(action-orientation) was manipulated and the proposed phys-
iological substrate was measured (left frontal cortical activa-
tion). Studies of this type are limited in the causal inferences
that can be drawn. Because the measured physiological activa-
tion may be only one of a number of physiological activations
that occur in response to the psychological manipulation, it
is possible that one of the other unmeasured physiological
activations is more responsible for the psychological process.
To provide stronger causal inferences regarding the role of a
particular neural structure's involvement in a particular psy-
chological process, it is important to reverse the direction and
manipulate the physiology and measure the psychology. Such
measurement also provides stronger causal evidence than sim-
ply correlating the proposed mediator with the outcome. Thus,
an experiment used neurofeedback to manipulate left frontal
cortical activity. After neurofeedback training, participants
were given a difficult decision and following the decision
their attitudinal spreading of alternatives was measured. The
manipulated decrease in relative left frontal activity led to a
decrease in spreading of alternatives.
Individual and Cultural Differences
Research has suggested that individual and cultural differences
may moderate dissonance processes. For instance, individuals
with greater preferences for consistency show greater attitude
change after being given high choice in an induced compliance
situation, and individuals from Eastern cultures show greater
dissonance-related attitude change as compared to individuals
from Western cultures when interdependence is salient. As
noted by Wicklund and Brehm, individual (or cultural) differ-
ences in dissonance-related attitude change could emerge
because of differences in the initial perception of discrepant
cognitions, the awareness of dissonance, the tolerance of dis-
sonance, and/or the mode of dissonance reduction. If attitude
change is the only measure in a standard dissonance experi-
ment examining individual differences, it is impossible to
determine why a particular individual difference may be
related to a pattern of attitude change. In order to determine
why a particular individual, or cultural, difference relates to a
pattern of attitude change, it would be necessary to measure
the relationship of this difference to factors influencing
dissonance.
Heine and Lehman found that North Americans and East
Asians differ in their attitudinal responses to difficult decisions.
While North Americans showed the typical spreading of
alternatives following the difficult decision (regarding choice
over popular compact disk music selections), East Asians
did not. This observed effect was not consistent with earlier
observations by Haruki Sakai and colleagues who had found
dissonance-related attitude change following public but not
private induced compliance.
However, Hoshino-Browne and colleagues noted this dis-
crepancy between results and suggested that the experiments
by Sakai and colleagues may have produced dissonance-related
attitude change because participants were concerned about the
interpersonal consequences of their actions. That is, the parti-
cipants, who were typically motivated to be interdependent
with others and avoidant of interpersonal conflict, experienced
dissonance because they had acted inconsistently with those
cultural ideals. To address these issues and others, Hoshino-
Browne and colleagues conducted four studies in which
European Canadians and Asian Canadians made difficult deci-
sions for themselves or for a friend. Results indicated that
whereas European Canadians spread alternatives more for
self than friend decisions, Asian Canadians spread alternatives
more for friend than self decisions. These results serve as a
reminder that the importance of the cognitions was one of
the factors affecting the magnitude of dissonance in Festinger's
original theory. Cultural values would be expected to relate
to the importance of cognitions, and thus, to the amount of
dissonance these behaviors would evoke.
Conclusion
Several experiments have challenged the revisions of disso-
nance theory and have provided support for Festinger's origi-
nal conception of dissonance. Clearly, dissonance has much to
do with inconsistency and is not due to such limiting condi-
tions as a self-threat or the production of an aversive conse-
quence. As Aronson noted, a number of social psychological
theories, such as self-affirmation theory, could be thought of as
dissonance in other guises. In addition to the theories noted by
Aronson, much research and theory concerned with guilt and
self-regulation over prejudiced and other impulses may be
understood from the perspective of dissonance. Incorporation
of the key variables and past research of dissonance theory into
these and other research enterprises will likely benefit those
enterprises and lead to a more cumulative psychological
theory.
Festinger did not propose why cognitive inconsistency pro-
duces discomfort and motivates perceptual, cognitive, and
Cognitive Dissonance Theory 549
behavioral changes. However, the action-based model of dis-
sonance does propose an underlying motivation. Research on
the action-based model suggests that dissonance reduction
may serve the function of assisting in the successful execution
of a commitment, which may facilitate effective and uncon-
flicted action. Incorporation of these ideas derived from action
control thinking may assist in stimulating new research on
dissonance theory and assist in connecting the large body of
dissonance theory evidence with other research literatures
concerned with action orientation, behavioral regulation,
emotion regulation, and the neural processes that underlie
these important psychological processes.
As dissonance theory goes on to its sixth decade, it clearly
has weathered many challenges but still provides much explan-
atory, integrative, and generative power.
See also: Altitude Change; Attitude Formation; Motivation;
Self-Esteem.
Further Reading
Aronson E (1992) The return of the repressed: Dissonance theory makes a comeback.
Psychological Inquiry 3: 303-31 1 .
Aronson E and Mills J (1959) The effect of severity of initiation on liking for a group.
Journal of Abnormal and Social Psychology 59: 1 77-1 81 .
Beauvois JL and Joule RV (1996)71 Radical Dissonance Theory. London: Taylor and
Francis.
Brehm JW (1956) Postdecision changes in the desirability of alternatives. Journal of
Abnormal and Social Psychology 52: 384-389.
Brehm JW and Cohen AR (1962) Explorations in Cognitive Dissonance.
New York: Wiley.
Cooper J and Fazio RH (1984) A new look at dissonance theory. In: Berkowitz L (ed.)
Advances in Experimental Social Psychology, vol. 17, pp. 229-264.
Orlando, FL: Academic Press.
Draycott S and Dabbs A (1998) Cognitive dissonance 1 : An overview of the literature
and its integration into theory and practice of clinical psychology. British
Journal of Clinical Psychology 37: 341-353.
Festinger L (1957)4 Theory of Cognitive Dissonance. Stanford, CA: Stanford University
Press.
Festinger L (1964) Conflict, Decision, and Dissonance. Oxford, England: Stanford
University Press.
Festinger L and Carlsmith JM (1959) Cognitive consequences of forced compliance.
Journal of Abnormal and Social Psychology 58: 203-21 0.
Harmon-Jones E, Amodio DM, and Harmon-Jones C (2009) Action-based model of
dissonance: A review, integration, and expansion of conceptions of cognitive
conflict. In: Zanna MP (ed.) Advances in Experimental Social Psychology, vol. 41 ,
pp. 119-166. San Diego, CA: Academic Press.
Harmon-Jones E and Mills J (1999) Cognitive Dissonance: Progress on a Pivotal
Theory in Social Psychology. Washington, DC: American Psychological
Association.
Steele CM (1988) The psychology of self-affirmation: Sustaining the integrity of the self.
In: Berkowitz L (ed.) Advances in Experimental Social Psychology, vol. 21,
pp. 261-302. San Diego, CA: Academic Press.
Stone J (2003) Self-consistency for low self-esteem in dissonance processes:
The role of self-standards. Personality and Social Psychology Bulletin 29:
846-858.
Wicklund RA and Brehm JW (1976) Perspectives on Cognitive Dissonance. Hillsdale,
NJ: Lawrence Erlbaum.
Comparative Primate Psychology
D Maestripieri, The University of Chicago, Chicago, IL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Adaptation A genetically controlled trait that has evolved by
natural selection.
Analogous traits Similar traits that evolved independently
in different species in response to similar environmental
pressures.
Comparative psychology The study of mind and behavior
in nonhuman animals.
Convergent evolution The process by which natural
selection produces similar adaptations in different species
that live in similar environments.
Homologous traits Similar traits possessed by
different species as a result of inheritance from a common
ancestor.
Phylogeny The evolutionary development and
history of a species or higher taxonomic grouping of
organisms.
Primatology The discipline that studies nonhuman
primates.
Trait A morphological, physiological, or behavioral
characteristic of an organism.
Evolutionary Foundations of Comparative
Primate Psychology
Psychology is the discipline that studies mind and behavior.
Within psychology, there are many subfields including, among
others, biological, clinical, cognitive, comparative, develop-
mental, educational, evolutionary, industrial/organizational,
personality and individual differences, and social psychology.
Some of these subfields are often recognized as independent
disciplines. Comparative psychology, which involves the study
of mental processes and behavior in other animals, is also
known as ethology or behavioral biology. Within comparative
psychology, there are subfields specializing in the study of
particular groups of animals. For example, the comparative
psychology of nonhuman primates is sometimes called behav-
ioral primatology, or simply primatology.
All of these multilevel disciplinary subdivisions have re-
sulted from the growth of psychological research, its expansion
in many different directions, and the increasing specialization
of researchers addressing different types of questions about
mind and behavior. Although specialization is both welcome
and inevitable, the risk involved in these sub-fields turning into
independent disciplines is that they may become conceptually
disconnected from psychology and from each other. With
regard to the study of mental and behavioral processes in non-
human primates, it is important to discuss how this area of
research is conceptually linked both to comparative psychology
and to psychology in general.
A frequently used rationale for conducting comparative
psychological research is the 'animal model' argument. This
argument is that similarities in brain structure and function, or
physiological processes, or learning abilities in animals and
humans allow researchers to use animals as models for a specific
aspect of human behavior, or cognitive process, or its underlying
neurobiological or physiological regulation. For example, ac-
cording to Gottlieb and Lickliter, the 'animal model' approach
entails finding "nonhuman species with behavioral and psycho-
logical repertoires that are similar to humans so that the results
of experiments with the model may throw light on seemingly
related behavior in human beings" (pp. 3 1 1 -3 1 2) . Evolutionary
arguments do not figure prominently in comparative research
conducted with the animal model approach. Particular organ-
isms are selected mainly on the basis of practical criteria, such as
they are small and cheap to maintain in the laboratory, they
reproduce frequently, and it is easier to conduct experimental
studies that may require highly invasive procedures. The most
common organisms that fit these criteria and are used in com-
parative psychological research include fruit flies, cockroaches,
some frogs and reptiles, pigeons and some other birds, and of
course, rats, mice, and other rodents. Although comparative
research with these 'model organisms' can make some beneficial
contributions to psychology, there are risks involved in selecting
organisms based solely on practical criteria. One of these risks is
that the extrapolation of findings or conclusions from animal to
human research may be inappropriate.
Comparative psychologists who use the 'animal model'
argument as rationale for their research not only recognize the
limitations of this approach but, surprisingly, also minimize
the contributions comparative research can make to psychol-
ogy. This position is exemplified by Gottlieb and Lickliter, who
argued that "the certainty that animal models are faithfully
mimicking their presumed human counterparts in the arena
of psychological, social, and behavioral function is always open
to even greater question" (p. 312). As a result, they conclude
that the main contribution of comparative research to psychol-
ogy is to provide food for thought, that is, hypotheses but not
data, general principles but not facts.
In a critique of Gottlieb & Lickliter's article, I argued that
a logical corollary of their position is that the rationale for
doing comparative animal research in general is very weak.
Computer simulations can be a good source of testable
hypotheses about human behavior. Studying human behavior
can be an excellent source of testable hypotheses about human
behavior. Why bother with animals? Given the economic
costs, the logistic difficulties, and the ethical issues associated
with animal research, such research would no longer be war-
ranted if its contribution were only to provide hypotheses or
food for thought. Even animal studies with highly invasive
550
Comparative Primate Psychology 551
experimental procedures that are impossible in humans would
be difficult to justify if their main contribution were only to
provide hypotheses. I argued instead that animal studies pro-
vide much more than food for thought, and discussed many
cases in which animal data have clear cross-species validity and
direct extrapolability to the human condition. To give just one
example, in his 1969 formulation of attachment theory, British
psychoanalyst John Bowlby did not use information from
nonhuman primate research just to develop the hypothesis
that there they might be an attachment system in humans or
some general principles about its functioning. He showed that
there were 'formal similarities' in the infant response to sepa-
ration, the development of fear of strangers, and the use of the
mother as a secure base in monkeys and in humans. Based on
these and other similarities, he argued that the development,
regulation, and adaptive function of the attachment system are
very similar in humans and some other primates, and that this
similarity is probably due to common descent, that is, humans
and closely related primates inherited the infant attachment
system from their common ancestors. Thus, Bowlby provided a
clear example of how data from nonhuman animals, and not
just hypotheses or general principles, can be directly extrapo-
lated to humans and how one can develop a theory of behavior
that has a strong evolutionary foundation and cross-species
validity.
Evolutionary biologist Theodosius Dobzhansky once said
"Nothing in biology makes sense except in light of evolution."
Many evolutionary and comparative psychologists believe that
the same should be said about psychology as well. Compara-
tive psychological research can be strongly justified on evolu-
tionary grounds, and evolutionary arguments can also help
select the best study organisms and the research questions
that can be best addressed from a comparative perspective.
One evolutionary rationale for conducting comparative re-
search is that similarities in behavior between animals and
humans are the result of common phylogenetic history. In
this view, certain traits are similar in animals and humans
because they were inherited from a common ancestor, that is,
they are 'homologous.' Studying these traits in animals helps
one elucidate the phylogenetic history of human behavior.
In general, the probability that two species have homologous
traits is higher the closer the phylogenetic relationship between
the species. Thus, human behavior is more likely to be homol-
ogous to the behavior of other primates than to the behavior of
nonprimate animals. This provides a strong evolutionary ratio-
nale for using nonhuman primates, especially primates that
are closely related to humans, in comparative psychological
research.
Another evolutionary rationale for conducting comparative
research is that similarities in behavior between animals and
humans may be the result of convergent evolution. This means
that similar traits in animals and humans evolved by natural
selection in response to similar pressures from the environ-
ment, but independently in different species. These traits are
considered 'analogous' and studying them can help one under-
stand how the environment has shaped human behavior
through the action of natural selection. Similar adaptations
to the environment can occur in species that are distantly
related and, therefore, in theory, to investigate the adaptive
aspects of human behavior from a comparative perspective,
honeybees can be as good as models as chimpanzees are. In
reality, however, there are many constraints on the type of
adaptations to the environment that organisms can evolve
through natural selection. Therefore, similarities in genetic,
anatomical, physiological, and cognitive constraints increase
the probability that organisms will evolve similar adaptations
to the environment. Thus, human behavior is generally more
likely to be analogous to the behavior of other primates than to
the behavior of nonprimate species. This provides another
strong evolutionary rationale for using nonhuman primates
in comparative psychological research.
Studying the mind and behavior of nonhuman primates
that are closely related to man can potentially enhance the
understanding of many human mental and behavioral pro-
cesses for clear evolutionary reasons. Although primatology
and psychology are sometimes considered different disciplines,
there is no conceptual reason why this should be the case. In
fact, the history of research on primate minds and behavior
illustrates that, from the very beginning, studies of primate and
human minds and behavior were closely interwoven.
History of Comparative Primate Psychology
The implications of primate behavior research for understand-
ing human behavior were first explicitly recognized by Darwin,
who in his 1872 book on 'The Expression of Emotions in Animals
and Man' drew several parallels between the facial expressions
of nonhuman primates and those of human beings. However,
it was psychologists rather than evolutionary biologists who
began the systematic study of primate behavior and cognition
at the beginning of the twentieth century.
One of the first psychologists to conduct systematic primate
behavioral research was Wolfgang Kohler. As a Gestalt psychol-
ogist, Kohler was interested in cognitive processes other than
learning and was curious to see if apes could use 'insight' to
solve novel cognitive tasks. In a research station established at
Tenerife on the Canary Islands, Kohler conducted many ele-
gant experiments with chimpanzees during the period 1913—
1917. Many of these experiments involved the manipulation of
the environment to obtain food rewards and the use of previ-
ously familiar objects in novel and instrumental ways. Kohler's
research questions and some of its procedures were very inno-
vative and some of its findings are still highly cited in contem-
porary research. He can certainly be considered one of the
founders of modern research on primate cognition.
In the United States, the systematic study of primate behav-
ior was pioneered by Robert Yerkes, a Harvard-trained psychol-
ogist who established a primate research facility in Orange
Park, Florida, with the goal of making primates available to
many different kinds of scientific inquiry, most notably, psy-
chological research. Yerkes felt that research on the behavior
and cognitive abilities of primates, and in particular the great
apes, would help answer some questions in psychology that
had historically been very difficult to address. His contribu-
tions to primate behavior research were many and ranged from
studies of spatial cognition and problem-solving to research on
social and maternal behavior.
In addition to the pioneer efforts by Kohler and Yerkes,
other early attempts to study primate behavior and cognition
552 Comparative Primate Psychology
were made in Russia, France, Cuba, and other parts of the
world. The 1920s and 1930s also saw the beginning of
attempts to teach language to chimpanzees. The first of such
attempts was made in 1930 by Kellogg and Kellogg who raised
a young chimpanzee named Gua along with their son Donald.
The Kelloggs' experiment turned out to be a failure but it was
followed by many others, using similar or different strategies.
Along with the growing recognition that primate behavior
could be useful to understand human behavior, the years
before World War II were characterized by increasing interest
in studying primates in their natural habitat and understand-
ing the basic principles regulating their social organization.
With World War II, research on primate behavior was inter-
rupted for almost a decade, but in the early 1950s and espe-
cially in the 1960s, there was renewed interest in studies of
primate behavior all around the world. Japanese primatolo-
gists' intensive and long-term studies of social behavior led to
the discovery of kinship systems and cultural traditions in
macaque societies. Primate behavior research in Japan, how-
ever, was originally conducted within the tradition of anthro-
pology, and it is only later that such research established a
strong connection with psychological science. The 1950s also
witnessed the resumption of research with rhesus macaques on
the island of Cayo Santiago, in Puerto Rico, where an American
zoologist, Clarence Ray Carpenter, had established a colony of
these monkeys prior to the War. The availability of genealogi-
cal information on the animals and the long-term observations
of their behavior contributed, along with the work of Japanese
primatologists, to the identification of the matrilineal structure
of macaque society and the mechanisms underlying the acqui-
sition of dominance.
As more information on primate social behavior became
available. Harvard anthropologist Sherwood Washburn devel-
oped the conviction that extant primate species could provide
important information on human origins and social evolution.
He and his graduate students pioneered field studies of primate
behavior in Africa and Asia, placing particular emphasis on
aggressive and maternal behavior. These two topics dominated
much of primate behavior research through the 1960s and
1970s. Interest in anthropology and human origins also moti-
vated paleontologist Robert Leakey to begin long-term studies
of chimpanzees, gorillas, and orangutans, which were led by
Jane Goodall, Diane Fossey, and Birute Galdikas, respectively.
Psychologists' interest in primate behavior rose dramati-
cally with the resumption of research in captivity after World
War II. Harry Harlow's research at the University of Wisconsin
played a pivotal role in this process. After making important
contributions to the study of primate learning, Harlow con-
centrated his efforts on elucidating the nature of infant attach-
ment and social development in rhesus monkeys. Harlow's
well-known experiments with surrogate mothers demonstrated
that the mother's ability to provide contact comfort is a more
important determinant of infant attachment than her ability to
provide milk, thus providing a fatal blow to secondary-drive
theories of attachment. Because Harlow's work touched upon
many areas of research that were very important to psycholo-
gists at that time (e.g., learning and motivation, attachment,
normal and abnormal social development, the social origin of
affective disorders), and because Harlow's academic career
took place within psychology, during the years in which most
of his work was conducted and published, primate behavior
research was very well known among psychologists.
Although Harlow was very effective in promoting the
importance of primate behavior research in the scientific com-
munity and the general public, the person who made the most
systematic effort to conceptually integrate primatology and
psychology was probably the British ethologist Robert Hinde.
Hinde's interest in primate research was sparked by John
Bowlby, who encouraged him to set up a colony of rhesus
monkeys in Cambridge and investigate mother-infant attach-
ment processes. From the study of social influences on the
mother-infant relationship, the scope of Hinde's research
was gradually broadened and elaborated into a conceptual
framework for the study of social processes, which distin-
guished three main levels of complexity: interactions, relation-
ships, and social structure. Hinde made important conceptual
contributions to the science of social relationships, and for
decades was one of the most articulate proponents of the
conceptual integration between biological and psychological
approaches to the study of behavior.
Thanks to the efforts of talented and charismatic scientists
such as Harlow and Hinde and the success of field studies of
primate behavior begun in the 1960s, primatology reached a
peak in popularity in the 1960s and early 1970s. In the mid- to
late 1960s, in particular, behavioral research in the United
States thrived at the newly established Regional Primate
Research Centers, and most research proposals to study pri-
mate behavior were readily funded by federal and private agen-
cies. In the 1960s and 1970s, a large number of articles and
books on primate behavior were published, and primate behav-
ior research was probably well represented in all branches of
scientific psychology, including developmental, social, cogni-
tive, and clinical. The heyday of primatology, however, did not
last long. In the early 1970s, that is, only ten years after the
establishment of the NIH-funded Regional Primate Research
Centers, there were already significant cuts to research funding.
The realization that evolutionary theory could be effectively
applied to the study of social behavior in the 1960s and 1970s
gave a great boost to primate research in the field. Although
anthropology and psychology had been the disciplines dom-
inating primate behavior research up to the 1970s, ecology and
evolutionary biology acquired a leading role in most sub-
sequent research. The fact that behavioral ecologists were
mostly interested in questions of adaptive function whereas
psychologists were mostly interested in questions of proxi-
mate causation or development of behavior was one of the
several factors that contributed to the growing separation
between primate behavior research and psychological science
that occurred in the 1980s and early 1990s. Another impor-
tant factor was the rapid progress of biological disciplines
such as genetics, molecular biology, and neuroscience and
the growing popularity of scientific reductionism. In particu-
lar, the success of neuroscience led to the optimistic view that
many important questions about behavior would eventually
be answered by studies of brain anatomy and function, thus
rendering behavioral research unnecessary. One corollary of
this view was the belief that comparative research with pri-
mates may not be as useful as research with other species,
given the difficulty of conducting molecular work with
primates.
Comparative Primate Psychology 553
Despite the current weakness of primatological research, in
the last decade, some favorable conditions have emerged for
renewed cross-fertilization between primatology and psychol-
ogy. The cognitive revolution that occurred in psychology in
the middle of the century was followed, a few decades later, by
a similar cognitive revolution in the field of animal behavior,
and primate behavior in particular. Thanks to the efforts of
pioneers such as Donald Griffin and Gordon Gallup, once it
became scientifically acceptable to ask whether animals have
a sense of self and understand other individuals as having a
mental life of their own, the field of primate cognition
boomed. It became apparent that many of the questions tradi-
tionally addressed by cognitive psychologists could also be
addressed, with similar or new experimental procedures, in
primates as well. Therefore, research on primate cognition
shifted from the study of learned behavior to the study of
mental representations of the self and of the physical and
social environment. Today, cognition is the branch of scientific
psychology in which primate behavior research is best known
and represented. Interestingly, and in line with the notion that
comparative psychological research should have strong evolu-
tionary foundations, a further impetus for research on primate
cognition has been provided by the framing of cognitive inves-
tigations within ecology and evolutionary biology, which has
led to a new understanding of primate cognitive adaptations,
their ecological significance, and evolutionary origins.
Comparative Primate Cognition
Primate cognitive adaptations can be conceptualized as com-
plex "behavioral adaptations in which perceptual and behav-
ioral processes are ( 1 ) organized flexibly, with the individual
organism making decisions among possible courses of action
based on an assessment of the current situation in relation to
its current goal; and (2) involve some kind of mental represen-
tation that goes beyond the information given to direct per-
ception." Cognitive adaptations, and their underlying neural
substrates, evolve by natural selection in response to recurrent
problems posed by the physical, ecological, or social environ-
ment. Within this framework, the question can be raised as to
whether the primate order as a whole exhibits cognitive adap-
tations that are different from those of other animals, and
whether primates that are phylogenetically closest to humans
show evidence of cognitive specializations similar to those of
the human species.
The study of primate cognitive adaptations has involved
many aspects of physical and social cognition. Primate re-
search in the domain of physical cognition has addressed how
monkeys and apes acquire information about the physical
space in which they live and the inanimate objects in it, how
information is mentally represented and processed, and how it
is retrieved and used to make decisions. Free-ranging primates
form spatial maps that represent the environment in which
they live and use them to make travel decisions as they search
for food within their home range. In the laboratory, primates
exhibit knowledge of movements of objects through space and
understanding of object permanence, that is, the notion that
objects continue to exist and maintain their features and
properties if they have been moved or hidden from view.
For example, primates search for hidden objects and can
solve tasks that require mental rotation of object orientation.
Though primates are proficient at these tasks, there is no evi-
dence that primates have greater understanding of space and
objects relative to other mammals, nor is there evidence of
significant differences among primate species (e.g., between
monkeys and apes).
Other research in the domain of physical cognition has
involved object manipulation tasks, in which objects are used
in relation to other objects, and which require an understand-
ing of causality (e.g., the relation between the use of the tool
and the goal to be accomplished with it). Many species of
primates, and especially capuchin monkeys and the great
apes, are proficient tool users and also show some evidence
of understanding of causality. However, primates' tool using
skills have been matched or even surpassed by those of some
corvid birds. Discrimination learning studies have addressed
whether primates learn to discriminate particular features
of objects and assign these objects to categories on the basis
of similarities and differences in these features. These studies
have shown that primates can not only discriminate and cate-
gorize objects but also understand complex rules underlying
categorization, for example, the notion that categories of
objects can be formed on rules such as identity, oddity, same-
ness, or difference. Similar to birds and other mammals (e.g.,
laboratory rats), primates also possess the ability to make
accurate estimates of small quantities of items as well as the
ability to solve simple tasks involving quantity conservation or
summation. The exact perceptual or conceptual mechanisms
underlying these skills remain unclear.
Taken together, studies of primate physical cognition have
shown that monkeys and apes possess the ability to form
mental representations of their space and objects, including
hidden ones, but show little evidence of greater learning skills
or greater understanding of the physical world and its proper-
ties than other vertebrate animals do. The strongest evidence
for a potentially unique primate cognitive specialization in the
realm of physical cognition involves the use of tools and the
understanding of relational properties of objects including
causality. This is particularly strong in large-brained primate
species that face strong ecological pressures for complex food
processing such as capuchin monkeys, and for all species of
great apes.
In the domain of social cognition, early studies of primates'
ability to recognize themselves in a mirror were driven by
the hypotheses that mirror self-recognition indicates self-
awareness and that knowledge of the self forms the basis for
theory of mind. On the mark test of self-recognition, primates
as a whole perform better than other animals, and apes per-
form better than monkeys. However, the notion that learning
how to use a mirror to inspect inaccessible aspects of one's
body necessarily entails possessing a concept of self has been
questioned. Experimental evidence for theory of mind skills
including intentionality or attribution of knowledge or igno-
rance to others is scarce but stronger for chimpanzees and
other apes than for monkeys. Theory of mind is clearly a
unique primate cognitive specialization but could be limited
to the human species, and possibly the great apes. Complex
forms of social learning involving emulation, imitation, or
teaching have also been documented mainly in great apes,
554 Comparative Primate Psychology
where they are believed to form the basis for the origin and
spread of behavioral traditions that may approximate elemen-
tary forms of human culture. The question of whether or not
culture can be considered a cognitive specialization unique to
humans or shared by other primates rests on how culture is
defined and what operational criteria are used for its identifi-
cation across species.
While the question of whether nonhuman primates have the
ability to think about other individuals' mental states remains
unanswered, it is well recognized that they excel at the task of
observing other individuals' behavior, remembering past inter-
actions, and making predictions about the future. Primates form
complex social relationships with others and have knowledge
and memory not only of their own relationships, but also of
relationships between other individuals. Studies investigating
this aspect of social cognition have assessed primates' ability
to recognize kinship, dominance rank relationships, or friend-
ships between individuals that reside in their social group.
Knowledge of social relationships is used in complex cooperative
and competitive strategies involving exchange of favors, alliance
formation, opportunistic exploitation of social situations, and
manipulation of other individuals with deceitful tactics.
Complex social strategies in group-living monkeys and
apes invariably entail the exchange of vocal or visual signals
between individuals. Communication can therefore provide a
window into the primate social mind. Specifically, many
aspects of communication can provide insights into social
cognition, including the role of social learning in the develop-
mental acquisition of signal production, comprehension, and
usage; the extent to which signals are under volitional control
and can be used flexibly; the complexity in the structure of
signals; the information content or meaning of signals; the
extent to which signals are combined with other signals within
the same modality or across different modalities to accomplish
different functions; the extent to which combinations of sig-
nals exhibit properties of human languages such as grammar or
syntax; and the extent to which the production of signals is
modified in relation to the presence of particular individuals
(audience effects), their attentional states or current behavior,
and possibly also their mental states.
The search for cognitive complexity in primate communi-
cation has often focused on vocalizations, in part because of
possible direct parallels between monkey vocalizations and
human speech. For example, great emphasis has been placed
on the finding that vervet monkeys and other primates possess
different alarm calls for aerial and terrestrial predators, and are
therefore capable of semantic communication. It is now recog-
nized, however, that this ability is shared by a number of birds
and other mammals. Food calls have also been given as exam-
ples of referential signals, as according to some researchers,
they convey information about the type, quantity, and location
of food to other conspecifics. It is unlikely, however, that
primate vocalizations about predators or food require a higher
degree of cognitive complexity than similar vocalizations used
by other mammals, birds, and other animals. This is because
the problems faced by most primates during foraging or escap-
ing predators are simply no different in complexity from those
faced by most other animal species. Therefore, it is difficult to
argue that these activities posed a special pressure to evolve
higher cognitive or communicative abilities in primates.
The agonistic screams of macaques appear to elicit different
responses from other group members in relation to character-
istics of opponents such as their dominance rank, and repre-
sentational signaling in the context of recruitment of agonistic
support is an ability that might have been strongly selected for
in the social environment of group-living primates. Vocaliza-
tions that are emitted in order to coordinate the behavior of
group members during travel or to facilitate affiliative and
bonding interactions are an even more promising area of
investigation because, unlike antipredator calls and recruit-
ment screams, these signals are not obviously associated with
states of high arousal. Contact vocalizations that facilitate
coordination of group movements and close-range interac-
tions are particularly well developed in arboreal species such
as New World monkeys. The complexity of vocal structure and
vocal sequences in New World monkeys, however, is likely to
be the result of the pressures of arboreal life rather than those
of social variables. Moreover, the referential nature of agonistic
screams, grunts, or other short-range contact calls has been
called into question even for the Old World monkeys and apes.
Regardless of how primate calls are interpreted, there is
little or no evidence that primates as a whole show complex
cognitive specializations in their vocal communication abil-
ities relative to other animals. Moreover, there is no trend
toward increasing complexity in the structure, function, and
use of vocal signals from the prosimians to the New World
monkeys, the Old World monkeys, and the great apes, suggest-
ing that the evolutionary increase in brain size that occurred in
the Cercopithecoids and the ape lineage was not associated
with increasing complexity in vocal exchanges or their cogni-
tive substrates. Such an evolutionary trend, however, is observ-
able in the use of nonvocal signals. In the Old World monkeys
and in the great apes, there is a clear increase in the role played
by facial expressions (associated with the development of com-
plex facial musculature) relative to vocalizations. Moreover, in
the great apes, there is an involvement of the arms and hands
in making social gestures to a degree that is not observed in
other nonhuman primates or other animals.
Although there is no good evidence that manual gestures
are socially learned from others, the flexibility with which great
apes use gestures in different contexts likely requires consider-
able developmental learning and experience. If gestures de-
velop through a process of ontogenetic ritualization, it would
require many exposures to and opportunities to perform a
particular action before it becomes a gesture. If the forms of
gestures are largely innate, then a period of learning when
different gestures are likely to be effective (e.g., visual only
when visible) would be required. Thus, it is possible that pri-
mates, and especially great apes, have evolved cognitive specia-
lizations to attend to and learn to use social variables (such as
identity, visual attention, and dominance) during communica-
tion, and particularly during gesture.
The importance of development in shaping primate com-
munication has been investigated in only a limited number of
studies, and almost all of them have focused on vocal commu-
nication. Additionally, little is known about the relationships
between cognitive abilities such as understanding visual atten-
tion, gaze following, and individual recognition and the struc-
ture and use of communication systems in nonhuman
primates. In humans, language develops alongside a whole
Comparative Primate Psychology 555
range of cognitive abilities - building upon some and providing
the foundation for others. The relationship between the emer-
gence of language and other cognitive abilities is well studied in
humans: similar work is needed in nonhuman primates. Com-
parative developmental studies are needed to understand
whether and how development of one ability (e.g. : understand-
ing social hierarchies) affects or is affected by the development
of communicative behaviors (e.g., using different strategies
when communicating with a dominant vs. a subordinate indi-
vidual). Within this framework of research, developmental
studies of gesture acquisition and experimental studies of ges-
ture use have the potential to shed new light on the social-
cognitive specializations of great apes and, more generally,
provide a window into the nonhuman primate mind.
See also: Evolutionary Psychology; Primate Cognition.
Further Reading
Blum D (2002) Love at Goon Park: Harry Harlow and the Science of Affection.
Cambridge, MA: Perseus.
Call J and Tomasello M (eds.) (2007) The Gestural Communication of Apes and
Monkeys. Mahwah, NJ: Lawrence Erlbaum.
Cheney DL and Seyfarth RM (2007) Baboon Metaphysics: The Evolution of a Social
Mind. Chicago: The University of Chicago Press.
Gallup GG (1 985) Do minds exist in species other than our own? Neuroscience and
Biobehavioral Reviews 9: 631-641.
Gottlieb G and Lickliter R (2004) The various roles of animal models in understanding
human development. Social Development 13: 311-325.
Hauser MD (1996) The Evolution of Communication. Cambridge, MA: MIT Press.
Hinde RA (1987) Individuals, Relationship’s and Culture. Links Between Ethology
and the Social Sciences. Cambridge: Cambridge University Press.
Maestripieri D (ed.) (2003) Primate Psychology. Cambridge, MA: Harvard
University Press.
Maestripieri D (2003a) The past, present, and future of primate psychology. In:
Maestripieri D (ed.) Primate Psychology, pp. 1-16. Cambridge, MA: Harvard
University Press.
Maestripieri D (2003b) Attachment. In: Maestripieri D (ed.) Primate Psychology,
pp. 108-143. Cambridge, MA: Harvard University Press.
Maestripieri D (2005) On the importance of comparative research for the understanding
of human behavior and development: A reply to Gottlieb & Lickliter (2004).
Social Development 14: 181-186.
Maestripieri D (2007) Macachiavellian Intelligence: How Rhesus Macaques and
Humans Have Conquered the World. Chicago: The University of Chicago Press.
Matsuzawa T (2001) Primate Origins of Human Cognition and Behavior.
Berlin: Springer.
Rendall D and Di Fiore A (2007) Homoplasy, homology, and the perceived special
status of behavior in evolution. Journal of Human Evolution 52: 504-521.
Tomasello M and Call J (1997) Primate Cognition. Oxford: Oxford University Press.
Competition
S H Liening, The University of Texas at Austin, Austin, TX, USA
P H Mehta, The University of Oregon, Eugene, OR, USA
R A Josephs, The University of Texas at Austin, Austin, TX, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Basal levels Baseline or resting state levels of hormones.
Behavioral economics A field of research interested in
economic decision making with an emphasis on whether to
cooperate for collective gain or defect for individual gain.
Cortisol A steroid hormone that is part of the
physiological stress response that can also act as a
behavioral inhibitor.
Dominance An aspect of personality characterized by a
drive for attaining and maintaining high social status.
Moderator A variable that controls the nature of the effect
that an independent variable has on a dependent variable.
Status An individual's position in a social hierarchy relative
to other members of the hierarchy.
Testosterone A steroid hormone that has masculinizing
effects on an individual that can also influence behavior.
Competition
Whether it is for a significant other, a promotion, or the admi-
ration of peers, social competition dictates a large amount
of human behavior. This article will provide a review of the
scientific research on social competition in humans. We will
discuss the social function of competition, dispositional dom-
inance, and the underlying physiology that drives dominance
behaviors, focusing on testosterone. To that end, we will dis-
cuss the dynamic relationship between testosterone, personal-
ity traits, situational forces, and cortisol. We will also discuss
sex differences in competitive behaviors, and conclude with a
discussion of the field of behavioral economics with an eye
toward how this research has informed our understanding of
competitive behaviors.
The Social Function of Competition
Across all species, individuals must compete for access to lim-
ited resources. Yet, rarely do individuals compete over every
uneaten apple or available mate. Rather, competitions are used
to establish a social hierarchy, and this hierarchy is the primary
means by which limited resources are allocated throughout the
group. High status, though risky to pursue, grants the individ-
ual increased access to limited group resources.
High status within the hierarchy gives an individual numer-
ous benefits. For humans, high status grants the possessor the
power to direct groups (i.e., leadership), control inner-group
functioning, influence lower status group members, and gain
privileged access to group resources. From an evolutionary per-
spective, competition exists to facilitate the perpetuation of those
genetic lines that are most fit to reproduce. By competing for
high status, the most able individuals are most likely to repro-
duce. By virtue of having high status, individuals who win social
competitions are healthier, have better diets, less stress, are more
likely to copulate, and their offspring are more likely to live to a
reproductively viable age. In short, high status yields a life style
that is generally more pleasant, healthier, and more productive,
and thus is generally more desirable than a life of low status.
A large body of research has examined both human and
nonhuman competition. Among many nonhuman species,
competition takes the form of physical aggression between
individuals or groups. This aggression is used to establish
who is granted access to what resources, including food, terri-
tory, or mates. Even among humans, competition may take a
violent and physically aggressive form. The establishment of
territory, access to coveted resources, or religious conflict
between countries and/or groups may lead to large scale con-
flict, such as war. Smaller scale interpersonal violence, such as
physical beatings or homicides, may be born out of interper-
sonal competition over a highly coveted resource, for example
two men fighting over the same woman. Across a number
of species, including humans, males tend to be the primary
perpetrators of aggressive competition. This may be due to
patriarchal social structures, evolution disproportionately ben-
efiting aggressive males, or the underlying biological mechan-
isms driving behavior. As discussed below, research suggests
that it is a combination of all three.
Unlike nonhuman animal species in which physical aggres-
sion is the primary form of competition, human competitions
do not always involve violence or physical aggression. Instead,
humans have multiple routes to obtain resources and establish
status. These alternate, nonaggressive routes include eminence
(i.e., status given as a result of high levels of achievement),
birthright, knowledge or ability, prosocial behaviors, and
social manipulation or coercion. Some of the more explicitly
competitive, but nonaggressive, means to pursue and obtain
high status include academic achievement, work promotions,
elections, or athletic competitions.
Research has shown that social competition develops at
a young age, even during prepubescence. Children learn at an
early age that there are both coercive and prosocial means for
gaining status, and they learn to utilize both effectively. Even
children who do not actively pursue high status clearly learn to
recognize the behavioral strategies being utilized by their peers.
Research shows that not only do children recognize different
strategies, but they respond to them in different ways. Even
amongst prepubescent children, coercive strategies tend to
lead to less stable, more transient forms of high status, whereas
556
Competition 557
prosocial strategies tend to lead to more long lasting and stable
forms of status. As an example, a bully may use coercion to
establish his/her dominance, but that is very tenuous. A more
friendly and prosocial strategy, alternatively, is more likely to
yield higher numbers of friends, increased popularity among
peers, and thus a stable coalition of social support.
Dominance and Competition
The drive to engage in social competition to move up the social
hierarchy characterizes much interpersonal human behavior.
'Dominance' refers to this striving for attaining and maintain-
ing high status. Within the realm of personality and social
psychology, dominance is often conceptualized as a trait-like
feature of a person's personality (i.e., it is an inherent,
enduring, and relatively static aspect of an individual's social
behavior). Dominance, as with other personality traits, varies
between individuals, with some individuals exhibiting higher
levels of dispositional dominance than others. Competition is
a common behavioral outcome of a dispositionally dominant
individual's striving for high status, and individuals high in
dominance tend to succeed in achieving the high status posi-
tions they desire.
Dominance can be expressed interpersonally in a number
of ways. Postural expansion, or carrying oneself to maximize
one's perceived size, maintaining eye contact while speaking to
another individual (whereas maintaining eye contact while
listening to another individual is perceived to be nondomi-
nant), and lower rates of smiling among women (but not men)
are some behavioral expressions of dominance. Dominant
individuals also respond to perceived status threats with anger
and aggression. High dominance has been characterized by
actions such as spontaneously and eagerly engaging in inter-
personal competition, establishing oneself as being in charge
of a group, projecting competence regardless of actual ability,
behaving assertive, and controlling the actions of others.
Dominance and Testosterone
A great deal of social endocrinology work has explored the role
of hormones in human social behaviors, and no hormone has
been given greater attention than testosterone. Testosterone is
an androgen, meaning it is a steroid hormone derived from
cholesterol that has masculinizing effects on an individual.
Testosterone is produced in both men and women by the
gonads, adrenal glands, and within the brain, though due to
the fact that the testes produce testosterone in much larger
quantities than any other source, men tend to have roughly
seven times as much testosterone as women. Despite the great
difference in testosterone levels between men and women,
relative testosterone levels within sex can have similar influ-
ences on social behaviors in both sexes. For instance, a woman
high in testosterone relative to other women will behave simi-
larly to a man high in testosterone relative to other men, even
though the high testosterone woman's testosterone levels will
be much lower than the high testosterone man's. Although all
hormone levels are subject to fluctuations caused by a number
of factors, basal (i.e., resting or baseline) steroid hormone
levels have been found to stay relatively static over extended
periods of time, leading many researchers to treat basal testos-
terone as a trait-like factor. It is worth noting that the majority
of research on testosterone and competitive behaviors has
focused on males. The findings and effects discussed below
are based primarily on research on males, but sex differences
and effects among females are discussed in section 'Sex Differ-
ences in Competition.'
In nonhuman animal species, testosterone has been
associated with rank in the social hierarchy in a variety of
species (e.g., baboons, deer, mandrills, chimpanzees, ducks,
possums, mice, rats, grouse, finches), although only when the
social hierarchy is unstable (justification for this is provided
below in our discussion of the challenge hypothesis). Testos-
terone has also been found to be positively correlated with
dominance-related behaviors, such as aggression. Just as
with nonhuman animals, research on humans has established
that there is a positive correlation between basal testosterone
levels and dominance. Research shows that testosterone corre-
lates positively with many behavioral aspects of dominance,
such as high self-regard, scores on personality questionnaires,
reactions to competitive outcomes, and social dominance.
Basal testosterone also correlates with increased sensitivity to
status-relevant information, even when that information is
presented outside of conscious awareness, and reacts by rising
or falling as a result of engaging in socially competitive
behaviors.
The link between testosterone and aggression deserves spe-
cial mention. Popularly, testosterone is believed to cause
increases in aggressive behavior, but this is not entirely tme.
Research suggests that testosterone promotes dominance beha-
viors, which can be manifested behaviorally in a number of
ways. In terms of aggression, though, dominance will only
manifest itself as physical aggression in social groups in
which physical force is an acceptable and efficient means by
which to compete for status (e.g., prison populations). Social
norms dictate that physical aggression in the pursuit of status is
a potentially costly behavioral strategy, with the possibility of
incurring punishment for violating those norms (e.g., arrests
for violent crimes). Even in those instances when testosterone
leads to aggressive behavior, research has found that the effect
tends to be moderated by neurobiological mechanisms related
to low levels of impulse control (e.g., low frontal lobe activity,
low levels of serotonin).
Although the human literature supports the idea that tes-
tosterone is linked to dominance, it is not the case that testos-
terone influences behavior in every situation. The challenge
hypothesis postulates that testosterone will influence behav-
ior only during critical periods to facilitate behavior during
moments of high competition, such as mating seasons or
during times of hierarchy instability. Although originally
developed to explain the mating behaviors of birds, the chal-
lenge hypothesis has since been expanded to explain a wide
range of testosterone-behavior relationships in a variety of
species. Researchers have confirmed that humans' physiology
and psychology interact in ways that are consistent with the
challenge hypothesis.
The critical role of the situation, according to the chal-
lenge hypothesis, is that testosterone will not have an effect
on behavior if the situation does not contain a status threat
or an opportunity to increase status. Put another way,
558 Competition
if testosterone is the underlying biological force driving
dominance-related behaviors, testosterone will not influence
behavior if there are no dominance-related gains to be had.
If the social hierarchy is firmly established, there is nothing to
be gained; then any influence testosterone might have on
behavior will disappear. It is when the hierarchy is unstable,
and thus high status positions are available, that testosterone's
influence over behavior will become apparent. This is in con-
trast to the 'mouse model' which portends a direct cause and
effect relationship between testosterone and behavior across all
situations.
In summary, research has established that basal testoster-
one strongly influences dominance behaviors in humans. Fur-
ther, both humans' and nonhuman animals' behavior will
only be influenced by testosterone in social environments in
which it is possible for an individual to increase his/her status.
This means that competition is generally only engaged in when
there is something to be gained from competing. It is also the
case that testosterone and dominance strongly interact with the
outcome of those social interactions in which status is gained
or denied. The successful attainment or denial of high status
and overt threats to one's status has been found to carry pow-
erful psychological and physiological consequences.
Testosterone and Status Mismatches
The psychological motivations that drive an individual's desire
for competition are deeply rooted in physiology, namely the
testosterone circulating through their bloodstream. Given that
this motivation is so deeply rooted, you would expect that
successful attainment or the denial of high status would
have dramatic consequences for a highly dominant individual.
Research has found that success or failure when competing
does elicit strong psychological and physiological reactions.
Researchers have found that when an individual engages in
social competition, basal testosterone predicts a number of
reactions to the successful attainment of high status or relega-
tion to low status. If an individual is high in testosterone
and experiences a victory, he/she will be content, indifferent
to status-relevant information, report lower levels of negative
emotional affect, and perform well on complex cognitive tasks.
The opposite effect has been shown for high testosterone indi-
viduals who experience a loss. These high testosterone indivi-
duals show increased levels of negative affect, cognitive
impairment, and an increased focus on status cues when rele-
gated to low status positions.
Notably, this research has found opposite patterns of effects
for those low in testosterone. Low testosterone individuals
who experience a victory become unhappy, report high levels
of distress, show cognitive impairment, and experience physi-
ological arousal, such as increased heart rate. Those low testos-
terone individuals who lose a competition, though, report
being quite content, perform very well on complex cognitive
tasks, experience a drop in heart rate, and show a lack of
interest in status-related information.
Termed the 'mismatch effect,' this illustrates the role of
basal testosterone in social competition. Individuals high in
basal testosterone have a strong desire for high status, whereas
those with low basal testosterone levels actively avoid high
status. When an individual is placed in a position of high or
low status that is incongruent with his/her desired status (i.e., a
high testosterone individual in a low status position or a low
testosterone individual in a high status position), they experi-
ence a number of negative consequences. A mismatch between
actual status and desired status causes emotional distress,
hypervigilance toward status-related information, physiologi-
cal arousal, and a decline in complex cognitive processing. This
mismatch effect also illustrates individual differences in domi-
nance by demonstrating that not everyone is striving for posi-
tions of high status, that there are some who would prefer to
avoid high status if at all possible. The adverse effects of status
mismatches can also extend from the individual to the group
level. If status mismatches exist among individuals within a
given group, those mismatches are associated with greater
pessimism among group members as to the likelihood of
the group's success on a task. Also, as the correlation between
individuals' testosterone and their status within a group
becomes more negative (i.e., mismatches increase), the group
as a whole functions more poorly.
It is clear that testosterone is intimately linked with domi-
nance, and the successful attainment or denial of high status
elicits strong reactions in dominant individuals. Winning or
losing status competitions causes psychological and physiolog-
ical reactions, and can lead to maladaptive behaviors. One
strong reaction caused by the interaction of dominance and
competitive outcomes is the short-term fluctuations in hor-
mone levels that can further influence behavior.
Dynamic Testosterone Fluctuations in Competition
Although basal testosterone levels remain static over time and
can be treated essentially as personality traits, testosterone
levels are also subject to dynamic fluctuations in reaction to
changes in the social environment. If basal testosterone is a
biomarker of trait dominance, changes in testosterone levels
are akin to a transient dominance mood state. A temporary rise
in testosterone can lead to a temporary elevation in one's social
competitive drive. Similarly, a temporary drop in testosterone
can temporarily suppress one's natural competitive motiva-
tion. The reciprocal model of social endocrinology tackles the
dynamic relationship between basal hormone levels, the envi-
ronment, and transient hormone fluctuations. Specifically,
hormones influence behavior, behavior influences the environ-
ment, changes in the environment cause temporary fluctuations
in circulating hormone levels, and these temporary fluctua-
tions in turn further influence behavior and the environment.
Hormones, and certainly testosterone, act not only as partial
causes of behavior, but also experience the effects of behavior.
Research on testosterone and competition has found that
testosterone rises in response to competitive environments.
Many athletes experience a rise in testosterone prior to
engaging in athletic competition to facilitate dominance beha-
viors in the competitive environment. Athletes' testosterone
rises significantly more before a competition played at home
than away, presumably because more status is on the line when
a team plays at home, and that the rise tends to be the strongest
when a team is competing against a serious rival. This suggests
that not only does testosterone rise in anticipation of a com-
petitive event, but that the degree of status threat moderates the
magnitude of testosterone increase.
Competition 559
A number of studies have also found that not only does
testosterone increase in reaction to a competitive environment,
but that it fluctuates in predictable ways in response to the
outcome of a competition; specifically, testosterone increases
following a victory and decreases following defeat. In fact, an
individual does not even need to actually engage in competi-
tion to elicit these effects. An individual's testosterone follows
this same pattern even when they experience a vicarious victory
or defeat, such as an imagined victory, watching their political
candidate win an election, or watching their sports team win
a game.
These temporary fluctuations are an important signaling
mechanism that prompts an appropriate behavioral response
(i.e., rising levels prompting further engagement in competi-
tion, falling levels prompting withdrawal from competition).
The temporary changes in testosterone following a loss influ-
ence how individuals choose to behaviorally react to that loss.
They are either motivated to attempt to regain high status when
testosterone increases or withdraw from the possible further
loss of status when testosterone decreases.
Even though many researchers have found that testosterone
levels change in reaction to the outcome of a competition,
some research has failed to find this effect. Instead they
found that testosterone changes following a competition were
moderated by a number of psychological factors. For instance,
in some group competitions, increases in testosterone are pos-
itively correlated with how much an individual contributes to
his/her team's victory, such that the more a given individual
was responsible for his/her team's victory, the more his/her
testosterone will increase. It has also been found that changes
in testosterone levels are negatively correlated among winners
who attributed their victory to external factors (e.g., luck), and
were positively correlated amongst losers who made the same
external attributions. This suggests that simply engaging in
competition is not enough, but that psychological factors,
such as motivation and cognitive appraisals of the competitive
situation, are what drive reactions to competitive outcomes.
Nonconscious motivational drives can also moderate the
effect of competition on testosterone fluctuations. Namely,
those who are high in implicit (i.e., nonconscious) power
motivation show the expected hormonal reactions to winning
or losing a contest. In this case, the competition-testosterone-
change link is dependent on personality factors related to an
unconscious desire for power and high status, not merely the
experience of winning or losing. Some research has also found
that some personality factors not even directly related to status
and power can also influence the relationship between compe-
tition and testosterone change, such as social anxiety.
Even though previous research has clearly linked testoster-
one to status-seeking, there have been striking inconsistencies
across studies. Although much of the research summarized
above did indeed show effects of testosterone on behavior,
other studies have found weak or null effects. One explanation
is that testosterone's effect on behavior is moderated not only
by psychological factors, but also by the hormone cortisol.
The Moderating Role of Cortisol
Cortisol, like testosterone, is a steroid hormone. Cortisol is
known primarily for its role in the physiological stress response
via activation of the hypothalamic-pituitary-adrenal axis
(HPA-axis). The HPA-axis consists of the series of communica-
tions between the hypothalamus and the pituitary and adrenal
glands. When the hypothalamus receives information that
there is a threat in the environment, it relays a signal via the
pituitary gland to the adrenal gland, which releases cortisol, as
well as other hormones such as epinephrine and norepineph-
rine. The resulting stress response causes physiological changes
which include increased respiration, heart rate, blood pressure,
and blood glucose levels. These changes are all intended to
prime the organism to physically respond to the threat by
facilitating fight or flight behaviors. Unlike the catecholamines
(i.e., epinephrine and norepinephrine), cortisol is able to pass
through the blood-brain barrier, making it the primary behav-
ioral influence during active stress responses. As such, cortisol
has been shown to have not only physiological effects, but
psychological and behavioral effects as well.
Not only is cortisol linked to the experience of stress, but it
has been linked to approach and avoidance behaviors as well.
Approach motivation is conceptualized as a broad system
motivating behavior toward desirable and rewarding out-
comes. Avoidance motivation, on the other hand, is a parallel
system that motivates an individual to behave in ways that
facilitate moving away from undesirable outcomes. Essentially,
high approach motivation is characterized by appetitive beha-
viors and sensitivity to reward, whereas avoidance motivation
is characterized by avoidant behaviors and sensitivity to pun-
ishment. High levels of cortisol have been found to correspond
to behavioral inhibition, shyness, and introverted behavior, all
avoidance motivation behaviors. Low levels of cortisol, on the
other hand, have been found to correspond to extraverted and
disinhibited behaviors, more approach-motivated behaviors.
Cortisol has been shown to play a moderating role between
testosterone and behavior. Namely, high cortisol levels can
suppress the action of testosterone. Given what we know
about the relationship between testosterone and dominance
behaviors, combined with the moderating role of status-
relevant situational cues, the impact of cortisol provides yet
another important factor underlying the manifestation of
social competition. Testosterone is the underlying force driving
social competition, but certain psychological (e.g., need for
power) and situational (e.g., status threat) conditions must
be met for testosterone to be manifested behaviorally. In addi-
tion to these conditions, testosterone's behavioral manifesta-
tion also depends on low levels of basal cortisol.
For instance, research shows that basal cortisol and basal
testosterone interact to significantly predict individual differ-
ences in overt aggression among adolescent males, such that
testosterone and aggression are related only when the individ-
ual also has low levels of cortisol. When cortisol is high,
testosterone and aggression are unrelated. Furthermore, the
effects of testosterone on competition and leadership beha-
viors are also only expressed when basal cortisol is low. For
instance, when cortisol is low, basal testosterone predicts
whether or not an individual will continue competing or will
withdraw from competition after having been defeated once.
Specifically, high basal testosterone motivates an individual to
recompete after a loss, while low basal testosterone motivates
withdrawal to avoid further losses, but this effect is only
observed if basal cortisol is low. When cortisol is high the
560 Competition
effects of testosterone following a defeat are suppressed. This
cortisol-testosterone interaction effect can even be extended to
other dominance-relevant situations, such as leadership. Once
again, basal cortisol moderates the effect of testosterone on
behavior, such that when cortisol is low, the higher an indivi-
dual's basal testosterone, the more dominant their behavior
will appear to others. This behavioral effect of testosterone
completely disappears among those leaders who have high
levels of basal cortisol.
This moderation suggests that the biological underpinnings
of social competition are more complex than a univariate
relationship between testosterone and competition. It is not
enough to have a testosterone-fueled desire to attain and main-
tain high status, but one also needs to have an approach
motivated social style to act on that desire. Or at the very
least, to not have a socially fearful, avoidance motivated behav-
ioral style.
Research on the physiological interaction of testosterone
and cortisol suggests a number of possible mechanisms for
how cortisol is able to moderate the relationship between
testosterone and behavior. Just as cortisol is tied to the HPA-
axis, testosterone is tied to the hypothalamic-pituitary-
gonadal axis (HPG-axis). The HPG-axis cues the release of
testosterone in preparation for reproductive behaviors. These
two axes have an antagonistic relationship, that is, when one
axis is activated, the other is suppressed. Thus, when the stress
response is activated, reproductive behaviors will be sup-
pressed until an organism's physiology has returned to base-
line. Cortisol, as well as other glucocorticoids, can also have an
impact on the expression of androgen receptors. High levels of
glucocorticoids downregulate the expression of androgen
receptors. Thus, high levels of cortisol might not affect testos-
terone levels, but rather affect the degree to which testosterone
could influence behavior by regulating the number of available
receptor sites for testosterone to bind to.
Sex Differences in Competition
As stated above, the majority of research on social competition
has focused disproportionately on male behavior. There is a
relative dearth of research on competitive behaviors among
women, but the existing research does suggest that there are
sex differences in how men and women engage with and react
to competition. Some evolutionary psychologists have theo-
rized that there should be clear sex differences in competitive
behavior. Namely, men are more motivated by competition,
both as individuals and as a group, whereas women are moti-
vated by social needs. For instance, some behavioral research
has found that women tend to cooperate more than men
across situations, whereas men tend to cooperate with one
another primarily as a means to compete as a group against
another group. These theories suggest that, evolutionarily,
women are more interested in protecting offspring and pro-
moting social cohesion for the purposes of maintaining social
stability and security. As is outlined below, there are mixed
findings on how women's competitive tendencies match or
differ from those of men.
A number of research studies have found that females (both
women and young girls) tend to be less competitive than their
male counterparts. There are a number of possible explana-
tions for why men and women differ in their competitive
tendencies. As mentioned above, some evolutionary theorists
have suggested that although it is beneficial for men to engage
in competitions to establish status, women benefit more from
prosocial engagement to establish social stability. Research on
stress and threat response behaviors seems to support this
view, by exploring the possibility that the traditional 'fight or
flight' response may be specific to male behavior, and that
females may respond to threat by engaging with friendly con-
specifics to establish safety and security in numbers. These
theories suggest an inherent behavioral tendency away from
competition and toward friendly social engagements among
females.
On the other hand, anthropological research suggests that
these sex differences in competitive behavior may result from
strong socializing forces. When comparing patriarchal societies
to matrilineal societies, women raised in a matrilineal society
tend to be more behaviorally competitive than the men in that
society, whereas in patriarchal societies, men tend to be the
more competitive sex. These findings are conceptually in line
with evolutionary theories in a broad sense, namely that differ-
ent behavioral orientations (i.e., competitive vs. cooperative)
yield different social benefits. The clear difference between
these views is whether these differences are innate or learned.
It appears that it is some combination of both. Although
research in anthropology has shown that socialization deter-
mines whether competitive behaviors are more prevalent
among men or women, research in social endocrinology sug-
gests that there are underlying physiological systems tied to sex
differences in competitive behaviors.
As stated previously, women's levels of circulating testoster-
one are much lower than men's, due to differences in the
sources of testosterone. The testes produce the vast majority
of testosterone in men, whereas women's testosterone is pro-
duced primarily by the adrenal glands, and to a lesser extent the
ovaries. Ultimately, women have roughly a seventh of the
amount of circulating testosterone as men. It is worth noting
that although women's testosterone levels are significantly lower
than men's, their basal testosterone is equally stable over time.
Research examining the relationship between testosterone
and dominance in women has yielded mixed results in terms
of whether testosterone's effects are the same across sexes.
Women's basal testosterone predicts aggression and domi-
nance, social status, performance in individual versus group
competitions, and cognitive impairments associated with sta-
tus mismatches. In fact, the research on the effects of status
mismatches has been conducted primarily with mixed-sex
samples, and has found similar patterns of results for both
men and women. Some research, though, do not replicate
the testosterone effects found in men among a female sample.
Research on the anticipatory changes in testosterone prior to
an athletic competition has found that some female athletes do
not experience an elevation in testosterone leading up to a
competition, but other findings support this relationship.
Other research has also found that some female athletes do
not experience the postcompetition fluctuations in testoster-
one (i.e., a win or lose dependent increase or decrease in
testosterone), but other findings have found this postcompeti-
tion fluctuation.
Competition 561
Although the replicability of testosterone-dominance find-
ings among women is mixed, preliminary research has begun
to examine the possibility of a female-specific endocrine-
dependent form of dominance. Some researchers have found
that certain hormones, for instance the estrogen estradiol, are
better predictors of dominance behaviors in women than tes-
tosterone. This research suggests that women possess a similar
endocrine-dominance relationship, simply one that is not
necessarily testosterone specific. The research to date examin-
ing dominance in women indicates that further understanding
of the potential sex differences in the physiological underpin-
nings of dominance behavior is an important and rich avenue
for future research.
Although there appear to be sex differences in competitive
behaviors, research is mixed as to what those differences are
and what is underlying those differences. Behavioral research
suggests that women are not as competitive as men, possibly
due to socialization or due to the different evolutionary goals
associated with competition versus cooperation. Social endo-
crinology research has also produced mixed results in terms of
sex differences and dominance. Some research, such as reac-
tions to status mismatches, show limited sex differences, yet
other research, such as athletic competitions, show clear differ-
ences between men and women. These mixed results may be
due to erroneously applying theories and approaches derived
from male-specific research to females.
Competition and Behavioral Economics
A rich field of research that has looked at human competitive
behaviors is the field of behavioral economics. The study of
behavioral economics relies on the basic premise that in most
instances there are limited resources available to a group, thus
creating an inherent conflict between the interests of the group
and any one individual within that group. For instance, by
focusing solely on one's own interests, the individual may
utilize more of a given resource than his or her equitable
portion. This overuse of a resource creates shortages for the
rest of the group, which in turn leads to increased and poten-
tially costly competition within the group. Behavioral econom-
ics utilizes interpersonal economic 'games' that rely on pitting
an individual's self-interest against that of a cohort. In most
instances, an individual can obtain a modest gain by cooperat-
ing with his/her partner, but could potentially obtain a larger
gain by defecting from his/her partner. High levels of disposi-
tional competitiveness have been linked to an increased pro-
pensity to betray one's partner to obtain more resources, but
a number of situational and social factors influence behavior
as well.
Research shows that the behavior of an individual's partner
exerts a powerful influence over how that individual will
behave. If a partner is viewed as acting unfairly, the other
individual will respond by rejecting any kind of cooperation,
even if that rejection knowingly causes all of the resources to be
lost for both players. This retaliatory behavior serves to rein-
force social cooperation and maintain a certain level of parity
within the dyad regarding the distribution of resources. Aggres-
sive and competitive behavior on the part of one's partner
increases the likelihood that an individual will respond with
an equally aggressive and competitive behavioral style. The
opposite effect has also been found to be true. Individuals are
more likely to behave cooperatively and noncompetitively if
they are interacting with a cooperative partner.
Two important situational factors moderate this reciprocal
effect. First, reciprocity is particularly pronounced if indivi-
duals are interacting repeatedly over a series of trials. Repeated
cooperative interactions lead to developing a sense of trust.
Because individual success is tied to the behavior of the indi-
vidual's partner, individuals recognize that punishing an
aggressive partner and trusting a cooperative partner is the
best strategy for maximizing one's gains. Second, physical dis-
tance and anonymity both decrease reciprocal behaviors. If an
individual is physically isolated from his/her partner and is
able to behave anonymously, it increases the likelihood that,
that individual will behave aggressively and competitively,
regardless of the partners' behaviors. Anonymity appears to
nullify the social motivation to reciprocate.
The threat of retaliation is also affected by the magnitude of
the resources at stake. When individuals are engaging with one
another during an economic decision making game, the higher
the stakes, the more likely individuals are to act fair and coop-
erate with one another. If there is relatively little to be gained
or lost, individuals will act more competitively, because the
potential to gain an extra amount is offset by the potential to
lose an equally small amount. On the other hand, when the
stakes are higher, not only is there a potentially larger gain to
be had, but there is also the threat of potentially larger losses.
Humans have a well established aversion to losses, so as the
stakes increase, the threat of a large loss exerts a greater influ-
ence over behavior than the potential for a large gain, meaning
that as the stakes increase, competitive behaviors decrease.
In addition to situational and social forces, individual dif-
ferences also influence whether one acts aggressively or coop-
eratively toward his/her partner. Research has shown that
dispositional dominance is a power factor in determining
whether an individual tends toward a competitive or coopera-
tive orientation. The effect of dominance is moderated by
situational forces, in some instances driving an individual to
treat their partner as a competitor and attempt to secure
as many resources for him/herself as possible, and actually
facilitate cooperative behavior between individuals in other
instances.
In some interactions, an individual high in dispositional
dominance may view cooperation as an acknowledgment that
both individuals are equal, and may opt to act aggressively to
establish his/her higher social status. Research has supported
this theory, showing that high dominance individuals view
these situations as moments of potential status instability.
If the pair is even, at the outset, high dominance individuals
will engage with their partner with a competitive motivation,
attempting to acquire more resources than the partner in order
to elevate their own status. If the pair is asymmetrical in some
way (e.g., one individual has more resources than the other),
high dominance individuals will respond to this asymmetry
as a status challenge, and respond with aggression and/or
retaliation in order to obtain or preserve their status. On the
other hand, research has also shown that in certain circum-
stances high levels of dominance may actually increase the
likelihood of cooperative behavior. Recall that there are a
562 Competition
number of strategies for obtaining high status, one of which is to
behave prosocially toward one's peers. In the context of eco-
nomic games, by acting prosocially toward one's partner, it
minimizes the likelihood that the partner will retaliate or punish
the individual, and increases the likelihood that the partner will
reciprocate the cooperative behavior. It has been suggested that
a key circumstantial factor in determining which behavioral
strategy will be adopted is whether or not an individual feels a
continued dependence on one's partner to succeed. Individuals
dependent on their partners are more likely to adopt a prosocial
strategy, thus maximizing their benefit in the long term.
Conclusions and Current Research Trends
Competition among social animals takes many forms, includ-
ing physical aggression between individuals over a specific
resource and social competition for high status positions
within a hierarchy. High status positions offer a much more
efficient and less costly means to distribute limited resources.
Just as is the case among rats, mice, chimpanzees, baboons,
and countless other social animals, it is also the case among
humans.
The primary biological fuel driving social competition is
testosterone, which manifests itself behaviorally as dominance.
Dominance behaviors are meant to facilitate one's ascension
up the social hierarchy by besting other individuals within the
same hierarchy. The effect that testosterone has on dominance
behaviors is moderated by two important factors. One is the
hormone cortisol, which acts as a behavioral inhibitor, such
that when cortisol is high, testosterone's influence on behavior
is suppressed. The other important factor is the presence/
absence of appropriate social circumstances; namely, condi-
tions under which competition will yield actual social rewards.
Recent research has begun to explore more complex neuro-
biological and genetic factors that may be influencing compet-
itive behaviors. Neuroscience research is examining the
potential role of brain structures related to emotion regulation
and impulse control on behaviors such as aggression and
economic decision making. Behavioral genetics research has
been studying how genetic variations on neurotransmitter and
neurochemical promoter genes potentially influence complex
social behaviors, such as how the underexpression of certain
neurotransmitters may reduce impulse control or blunt threat
responses. These approaches are being combined with
continued research on situational, environmental, and cultural
influences over competitive behavior to further elucidate
the dynamic and interactional relationship between the envi-
ronment and biology that gives rise to complex social beha-
viors. Ultimately, the merging of multiple disciplines has
led to a highly nuanced examination of the biological, social,
and environmental forces influencing social competitive
behaviors.
See also: Aggression; Violence.
Further Reading
Archer J (2006) Testosterone and human aggression: An evaluation of the challenge
hypothesis. Neuroscience and Behavioral Reviews 30: 319-345.
Archer J, Birring SS, and Wu FCW (1998) The association between testosterone and
aggression among young men: Empirical findings and a meta-analysis. Aggressive
Behavior 24: 411-420.
Josephs RA, Sellers JG, Newman ML, and Mehta PH (2006) The mismatch effect: When
testosterone and status are at odds. Journal of Personality and Social Psychology
90(6): 999-1013.
Mazur A and Booth A (1998) Testosterone and dominance in men. Behavioral and Brain
Sciences 21(3): 353-397.
Mehta PH, Jones AC, and Josephs RA (2008) The social endocrinology of
dominance: Basal testosterone predicts cortisol changes and behavior
following victory and defeat. Journal of Personality and Social Psychology
94(6): 1078-1093.
Mehta PH and Josephs RA (2006) Testosterone change after losing predicts the
decision to compete again. Hormones and Behavior 50: 684-692.
Mehta PH and Josephs RA (2010) Testosterone and cortisol jointly regulate dominance:
Evidence for a dual-hormone hypothesis. Hormones and Behavior 58: 898-906.
Popma A, Vermeiren R, Geluk CAML, et al. (2007) Cortisol moderates the relationship
between testosterone and aggression in delinquent male adolescents. Biological
Psychiatry 61 : 405-41 1 .
Schultheiss 0C and Rohde W (2002) Implicit power motivation predicts men’s
testosterone changes and implicit learning in a contest situation. Hormones and
Behavior At : 195-202.
Sellers JG, Mehl MR, and Josephs RA (2007) Hormones and personality: Testosterone
as a marker of individual differences. Journal of Research in Personality
41: 126-138.
Stanton SJ and Schultheiss OC (2007) Basal and dynamic relationships between
implicit power motivation and estradiol in women. Hormones and Behavior
52: 571-580.
Viau V (2002) Functional cross-talk between the hypothalamic-pituitary-gonadal
and -adrenal axes. Journal of Neuroendocrinology 14: 506-513.
Wingfield JC, Hegner RE, Dufty AM Jr., and Ball GF (1990) The ’challenge hypothesis’:
Theoretical implications for patterns of testosterone secretion, mating systems, and
breeding strategies. American Naturalist 136(6): 829-846.
Zyphur M, Narayanan J, Koh G, and Koh D (2009) Testosterone-status mismatch lowers
collective efficacy in groups: Evidence from a slope-as-predictor multilevel
structural equation model. Organizational Behavior and Human Decision Processes
110: 70-79.
Confabulation and Reality Filtering
A Schnider, Hopitaux Universitaires de Geneve, Geneva, Switzerland
© 2012 Elsevier Inc. All rights reserved.
Glossary
Confabulation The emergence of memories of events and
experiences which never took place.
Frontal-subcortical loops Anatomically defined, separate
loops connecting specific parts of the prefrontal lobes with
the striatum, globus pallidus, substantia nigra (reticular
part), and the thalamic nucleus having connections with the
loop's originating part of the prefrontal cortex.
F1 2 [15]0-PET Positron emission tomography using
radioactively labeled water to study the distribution of
metabolic activity in the brain.
Orbitofrontal cortex Area at the inferior aspect of the
frontal lobes. Its posterior medial part has limbic
connections and projects to the ventral striatum.
Substantia nigra Anatomical structure in the midbrain
with two components: (1) the reticular part participates
in the frontal-subcortical loops; (2) the pars compacta
has the dopaminergic neurons, which have multiple
functions. They are part of the brain's reward system;
their degeneration contributes to the symptoms of
Parkinson's disease.
Definition
Confabulation derives from the Latin word 'confabulari' whose
original meaning was to gossip, to chat, or to discuss some-
thing with somebody. The word 'fabula,' contained in confab-
ulation, primarily meant a chat, dialog or tattle with people,
but it also meant the saga, fairy tale, and fiction.
Confabulation made its first appearance in the medical
literature around 1900, when Carl Wernicke used the term in
his textbook on psychiatry and defined it as 'the emergence of
memories of events and experiences which never took place.'
The term replaced the previous denominator, which Sergei
Korsakoff had used in his writings on confabulation around
1890: pseudoreminiscences. Korsakoff had defined them as 'a
situation in which a patient conceived of an event that he had
not really experienced, but that had only come to his mind, as
if it had really happened to him.'
Thus, since its first appearance in the medical literature,
confabulation denoted a failure of memory, a mnestic phe-
nomenon. Later authors introduced diverse refinements to the
definition, claming that confabulations had to be uninten-
tional; that they might have fantastic features, or not; could
only occur in organic brain damage, or also in psychiatric
disease; only concerned autobiographical memory, or might
extend to semantic memory, etc. None of these limitations is
unequivocally accepted.
The term confabulation was eventually also used to describe
false statements emanating from incorrect perceptions of bodily
states or of the outer world. These confabulations may be classi-
fied as nonmnestic confabulations. For example, the term was
used to describe the false idea of patients who have half of their
body paralyzed but believe that they can move it normally, a
situation called anosognosia. Likewise, patients with psychosis
or organic brain damage may believe they are in another place
than the real one (reduplicative paramnesia) or pretend to know
unfamiliar people or not to know familiar people (Capgras
syndrome; misidentification syndromes). These nonmnestic
confabulations often occur independently of mnestic confabu-
lations and normally have different anatomical bases. Their
association with mnestic confabulations will be discussed
below. In the following, confabulation denotes mnestic confa-
bulations unless otherwise stated.
Forms of Confabulation
From the beginning of the twentieth century, many authors
distinguished different types of mnestic confabulations. The
most common distinction proposed the following dichotomy:
1 . Momentary confabulations (other terms: out-of-embarrass-
ment, classic compensatory, or plausible confabulations):
in most definitions, these confabulations were considered
a consequence of memory loss resented by the patient who
would invent answers to questions in order to avoid
embarrassment.
2. Fantastic (or productive) confabulations: these would be
produced spontaneously and resemble delirious, dreamlike
states, as seen in psychosis.
Another distinction, initially used to separate different
degrees of confabulation, is the following one:
1. Provoked confabulations: these confabulations occur in
memory tests and may also occur in healthy subjects.
They were also considered a normal response to a faulty
memory.
2. Spontaneous confabulations: these were considered a path-
ological phenomenon, possibly reflecting an incoherent
and context-free retrieval of memories. Another definition
requires patients to act in agreement with the confabula-
tions as the proof of their spontaneous generation.
The precise definitions proposed by different authors often
seemed to be influenced by, and limited to, the precise types
of brain damage that they had studied. Although the two
classifications are widely accepted, the precise meaning of
the terms varies considerably between studies. Numerous
authors used these classifications to describe different degrees,
rather than different forms, of confabulation. However, some
563
564 Confabulation and Reality Filtering
dissociations have been proven. For example, the tendency
of patients to make intrusions in memory tests, that is, to
produce words that were not originally in the test (provoked
confabulations) doubly dissociates from the tendency to act
according to spontaneously produced confabulations. Also,
the severity of confabulations does not predict an association
with behavior: patients producing the most intense fantastic
confabulations, reported at times when antipsychotic medica-
tions were not available yet, showed no tendency to act on the
basis of these confabulations.
Thus, the proposed classifications described above, which
entail a dichotomy, do not account for the known dissocia-
tions between confabulations. I suggested the following classi-
fication that accounts for the currently known forms of
confabulations:
1. Simple provoked confabulations encompassing intrusions
or occasional distortions in memory tests. These confabula-
tions may also occur in healthy subjects. Thus, when recal-
ling the words from a previously learned word list, subjects
may produce items that were not in the list.
2. Momentary confabulations, which consist of false verbal
statements produced upon some form of incitement, be
this questioning or the circumstances of a discussion. False
statements can range from a slightly distorted response to a
question - and then appear to be provoked - to elaborate,
apparently spontaneous confabulations in a discussion.
Momentary confabulations may occur in diverse brain dis-
eases, may be present initially or appear in the course after
brain damage, and are probably not a unitary disorder. For
example, a hospitalized patient may falsely believe to have
been at work in the morning (autobiographical confabula-
tion), pretend to know what he ate for lunch on January 5th
2003 (autobiographical) or invent elements when recount-
ing a fairy tale (semantic confabulation).
3 . Fantastic confabulations, which have no basis in reality and
are intrinsically nonsensical and illogical. These confabula-
tions are very rare and may occur in severe psychosis, delir-
ium or advanced dementia. For example, patients have
been described who believed that they were the richest
kings on earth, had been Adam in paradise or had already
landed on Mars.
4. Behaviorally spontaneous confabulations in which patients
recount invented recent experiences and describe plans for
the future that are incompatible with their current situation.
At least occasionally, they also act according to such false
ideas. Mostly, they enact former habits: they think they
have business meetings or that they are expected at home.
This form is always accompanied by amnesia (as measured
by free recall) and disorientation: the patients do not know
the time, place or their current role. For example, a 58-year
old mother, who had suffered rupture of an aneurysm of
the anterior communicating artery, left the examination in
the conviction that she had to feed her baby, who was 35
years old at the time.
Until now, only the dissociation between forms 1 and 4 has
been proven. Patients may have a combination of momentary
and behaviorally spontaneous confabulation, possibly also
fantastic confabulations.
Simple provoked confabulations - according to the above
definition - require focused memory testing and will then
become evident as intrusions. The other forms are normally
easily detected in a discussion, when patients talk about events
or experiences that have not really taken place. Questionnaires
containing questions that cannot normally be answered have
been proposed to quantify momentary confabulations. Fantas-
tic confabulations are often produced spontaneously by con-
fused patients but may be spurred by questions. Behaviorally
spontaneous confabulation normally requires clinical observa-
tion: while the confabulations come out in discussions, in
which patients recount experiences and actions that have not
truly taken place, the spontaneous behavior betraying the
spontaneity of the false ideas may only be revealed by behav-
ioral observation. The combination of amnesia with disorien-
tation increases the likelihood that a patient has behaviorally
spontaneous confabulation; the absence of disorientation
makes the presence of behaviorally spontaneous confabulation
unlikely.
Disorders Associated with Confabulation
Initial descriptions of confabulations emphasized their associ-
ation with anterograde amnesia and disorientation. Later
observations suggested an association with false recognition
as well as behavioral particularities, namely, anosognosia and
misidentification syndromes. The following paragraphs sum-
marise the evidence.
Simple provoked confabulation will not be particularly
considered in the following; as they also occur in healthy
subjects, these associations are obviously not mandatory for
provoked confabulations to occur. Overall, simple provoked
confabulations (intrusions) are more common in brain dam-
aged subjects, but their intensity correlates neither with the
severity of amnesia nor of executive failures. Most notably,
simple provoked confabulations are not associated with false
recognition: among comparably severe amnesic subjects, the
number of intrusions produced in a recall paradigm does not
correlate with the number of items falsely recognized in a
recognition paradigm.
In the other forms of confabulation, the associations are
more complex.
Amnesia
Relevant momentary, fantastic, and behaviorally spontaneous
confabulations are always associated with moderate-to-severe
amnesia as measured with tests of delayed free recall, that is,
when subjects have to reproduce information from memory
with no external help. By contrast, if a memory task is relatively
free of strategic demands and just requires recognition of pre-
viously presented material, amnesia is not always present. This
may be tested, for example, with a continuous recognition task,
in which subjects see a long list of items (typically pictures) and
are requested to recognize item repetitions within the series.
Under such conditions, occasional behaviorally spontaneous,
and presumably also momentary and fantastic confabulators,
may demonstrate intact learning and recognition. Thus, in
Confabulation and Reality Filtering 565
terms of simple information storage, confabulation and amne-
sia may dissociate; confabulation need not be associated with a
gap in memory.
Disorientation
Disorientation describes the inability to correctly acknowledge
the current time, place, one's role, and personal identity. These
dimensions are called orientation to time, space, situation, and
person, respectively. Personal orientation is rapidly restored
after most types of brain damage. While fantastic confabula-
tions, as described in the early literature, are often charac-
terized by absurd misinterpretation of one's own identity,
momentary and behaviorally spontaneous confabulation
were mostly described in patients with intact personal orienta-
tion. Most patients with momentary confabulations, especially
if they also have severe amnesia, are also disoriented. But the
association is not mandatory. This is different with behavior-
ally spontaneous confabulation which is characterized by a
confusion of reality that is practically always measurable as a
disorder of orientation. Although disorientation can result
from severe amnesia alone, most instances of disorientation
have the same mechanism as behaviorally spontaneous con-
fabulation, namely a failure of memory filtration and extinc-
tion capacity that will be described below.
False Recognition
False recognition describes the erroneous familiarity with items
in memory tasks; test subjects believe they have encountered a
piece of information in a specified previous context (e.g., a
word in a previously presented word list) when, in fact, the
information has not been presented. Although false recogni-
tion may occasionally be observed in memory impaired
patients, there is no confirmed association with any form of
confabulation; indeed, the two phenomena doubly dissociate.
Paramnesic Misidentification
Paramnesic misidentification describes erroneous interpreta-
tions regarding places, people, or the current situation. Some
patients with behaviorally spontaneous confabulation, possi-
bly also other confabulations, are convinced that they are not
at a hospital but rather at a familiar place, typically their home
or their work place. This false identification of one's current
whereabouts is called reduplicative paramnesia. Similarly,
patients may falsely recognize strangers as friends or collea-
gues, or deny familiarity with people they correctly perceive
(Capgras syndrome). Although these misidentifications of
places or people can sometimes be observed in behaviorally
spontaneous confabulators, the association is not stable: many
patients do not have such misidentification problems. Con-
versely, these misidentifications may occur without any other
form of confabulation and are then not accompanied by
corresponding acts. The same is true for another form of para-
mnesic misidentification: deja-vu, the strange feeling that one
has already lived the present situation in precisely the same
way. Deja-vu has no confirmed association with mnestic
confabulation.
Anosognosia
This term was originally introduced to describe the unaware-
ness of unilateral hemiplegia, but is nowadays used to describe
unawareness of disease in general. The false statements that
the patients make about their paralysed side have been called
confabulations (classified as nonmnestic confabulations in
this article). Conversely, diverse authors suggested that ano-
sognosia of memory failure was a common feature of and a
prerequisite for the occurrence of confabulation. Whereas fan-
tastic confabulators, as they have been described in the old
literature, have mostly no sense for the absurdity of their
thoughts, momentary confabulations may also be produced
by patients who are fully aware of, and suffer from amnesia.
In behaviorally spontaneous confabulation, the association is
more complex: although patients are normally unaware of the
falseness of their perceived reality, occasional patients may be
aware of having a memory problem.
Normal False Memories
Normal false memory denotes false products emanating from
normal memory processing by healthy subjects. Apart from
occasional provoked confabulations, which may also occur in
healthy subjects, the confabulations discussed in this article
occur in the context of pathological brain dysfunction or dam-
age and are, therefore, by definition distinct from normal
false memory. Nonetheless, a brief review may be warranted,
all the more so because some mechanisms hypothesized to
explain confabulation were meant to cover normal false mem-
ories, too.
Memory storage and retrieval is thought to rely on associa-
tive processes that do not necessarily produce a precise record
of an experience. Even personally very engaging memories -
so-called flash-bulb memories - may undergo significant dis-
tortion over time, even if they are normally relatively resistant
to alterations. The fact that the healthy brain processes the
meaning, rather than just perceptual features of information
renders memory vulnerable to associative errors. This weak-
ness, resulting from the associative power of memory, is
exploited in the DRM-paradigm (Deese-Roediger-McDermott-
paradigm) in which subjects learn a list of words (e.g., door,
glass, pane, shade, ledge, sill, house, open, curtain, frame, view,
breeze, sash, scream, shutter) which together associatively
encircle a target word not present in the list (e.g., window).
When healthy subjects learn such a list and later recall it, they
are likely to produce the target word, which is also falsely
recognized as having been present in the list. Amnesia seems
to have a protective effect against this type of susceptibility. In
the standard DRM-paradigm, confabulating amnesics did not
differ from nonconfabulating amnesics as regards susceptibility
to the DRM-effect.
Errors of normal memory have had the most dramatic
impact in eyewitness testimony. Indeed, innumerable studies
have shown that the precision of memory storage and retrieval
may be heavily influenced by cognitive and mood states at the
time of an event (e.g., ethnical biases), information received
after an event (e.g., confirmatory feedback) and the phrasing
of questions. Manipulative procedures are most efficacious if
566 Confabulation and Reality Filtering
the memory of an event is relatively weak and manipulative
information or questions are plausible. Under certain circum-
stances, it is possible to plant completely invented memories
that may be held with similar conviction as true memories.
The relationship between normal false memories and con-
fabulation has never been formally explored. However, it
appears that normal false memories are created during the
normal processing (retrieval, reconsolidation) of memories
and may be augmented or induced by manipulation at storage
or just before retrieval of information. There is no evidence that
pathological confabulations have to be planted first or require
suggestive questioning. They are very often composed of frag-
ments of memories of real events experienced before the occur-
rence of brain damage. It is therefore likely that normal false
memories and pathological confabulations have independent
mechanisms effective at different stages of memory processing.
Etiologies and Anatomy of Confabulations
Not all confabulations require brain damage. Provoked con-
fabulations can also be induced in healthy subjects, although
their frequency markedly increases after brain damage. In the
following, the causes and the anatomy of behaviorally more
relevant confabulations will be reviewed: momentary, fantas-
tic, and behaviorally spontaneous confabulation.
The distribution of brain diseases provoking confabula-
tions has evolved over the last century. The sequels of malnu-
trition and alcohol can nowadays be limited with vitamin B1
supplementation; syphilis of the nervous system can be treated
with antibiotics. Conversely, previously deadly diseases such
as herpes encephalitis and aneurysm bleeding are nowadays
more often survived; unfortunately, some survivors are left
with severe cognitive deficits including confabulation.
Since the early descriptions, a multitude of causes of con-
fabulations were recognized: paralytic dementia (syphilis),
alcoholism and malnutrition, infections, hypoxia, traumatic
brain injury, diverse types of poisoning, etc. With many of
these causes, confabulations occurred during an acute confu-
sional state (delirium), that is, a state of incoherent thinking,
hallucinations, and fluctuating attention.
Momentary Confabulations
Momentary confabulations have been recognized in many
brain diseases. In the early literature, malnutrition, especially
in the context of chronic alcohol abuse, was the most fre-
quently reported cause of confabulation, nowadays known to
be caused by a deficit of thiamine (vitamin B 1 ) . The Wernicke-
Korsakoff syndrome - according to the original descriptions -
encompasses amnesia, a strong tendency to confabulation,
disorientation, plus physical impairments: oculomotor distur-
bances and ataxia. The precise cause of the amnesia itself is
still a matter of debate: destruction of the mamillary bodies or
the dorsomedial thalamic nucleus, or frontal and medial tem-
poral lobe dysfunction have been invoked. In what way con-
fabulatory amnesia in this syndrome anatomically differs from
nonconfabulatory amnesia, is unknown.
Nowadays, the most quoted cause of relevant momentary
confabulation is rupture of an anterior communicating artery
aneurysm with destruction of the basal forebrain and orbito-
frontal cortex. However, confabulatory amnesia is very rare
with modern neurosurgical and neuroradiological treatment
and may concern a maximum of 3-5% of affected people.
Many other causes of momentary confabulation are known,
ranging from traumatic brain injury and tumors to ischemic
stroke, encephalitis, or hydrocephalus. Many of these diseases
do not induce circumscribed brain lesions and do, therefore,
not contribute to the understanding of the anatomical basis of
momentary confabulation. In those diseases that cause circum-
scribed brain lesion (tumors, stroke, etc.), confabulations
appear to be more common when a lesion involves the ante-
rior ventral part of the brain including the orbitofrontal cortex,
basal forebrain, and hypothalamus. But the association is not
strict: occasional patients with posterior lesions involving the
medial temporal lobe or even the parietal lobe, who confabu-
lated significantly, have been described.
Fantastic Confabulations
The most striking examples of fantastic confabulation,
completely nonsensical and illogical, were described in the
early literature in patients having advanced syphilis of the
nervous system (paralytic dementia) or schizophrenia. Both
diseases are nowadays causally or symptomatically treatable.
Neither has a circumscribed anatomical basis. In the more
recent literature, fantastic confabulations, typically less absurd
than those following the above causes, were reported in the
early phase after acute brain damage of diverse etiologies,
including, for example, aneurysm rupture, encephalitis, or
multiple sclerosis. In addition, they have been described in
advanced dementia, which differs from a confusional state by
more stable basic attention.
Behaviorally Spontaneous Confabulation
Behaviorally spontaneous confabulations share many causes
with momentary confabulations: rupture of an anterior com-
municating artery aneurysm, traumatic brain injury, herpes
encephalitis, etc. Indeed, behaviorally spontaneous confabula-
tors sometimes go on to produce only momentary confab-
ulations: they no longer act according to their confabulations
and orientation improves. The anatomical basis of behavior-
ally spontaneous confabulation has become clearer over the
last years. Figure 1 shows an overlap of focal brain lesions of
patients presenting behaviorally spontaneous confabulation
(with disorientation) in comparison with nonconfabulating
patients having similarly severe amnesia, as determined with
measures of free recall. As Figure 1 shows, the confabulating
patients had a strong lesion overlap in the area of the posterior
medial orbitofrontal cortex. Additional lesion sites have been
described: a lesion of the knee of the right internal capsule;
stroke of bilateral paramedian thalamus; amygdala damage on
the right side combined with perirhinal damage on the left
side; or lesions (sarcoidosis, tumor) of the anterior hypothala-
mus. In summary then, all hitherto reported cases presenting
with behavioral spontaneous confabulation, which persisted
beyond a confusional state (delirium), had lesions that
involved the posterior medial orbitofrontal cortex, in particu-
lar area 13, or brain areas having a direct connection with it.
Confabulation and Reality Filtering 567
Nonconfabulating amnesics
Behaviorally spontaneous confabulators
(b)
Figure 1 Anatomy of behaviorally spontaneous confabulation.
(a) Lesion overlap of nonconfabulating amnesics, (b) Lesion overlap
of behaviorally spontaneous confabulators. Shaded areas indicate
lesions of the paramedian cortex (left template) or inferior surface
(right template). Dashed lines are used to indicate lesion of the
lateral convexity (left template). A, amygdala; F, basal forebrain; H,
medial temporal lobe including hippocampus; HT, hypothalamus.
Reproduced from Schnider A (2008) The Confabulating Mind. How the
Brain Creates Reality. Oxford: Oxford University Press, with permission
from Oxford University Press.
Area 13 was also activated - as measured with H 2 [15]0 posi-
tron emission tomography - in healthy subjects, who suc-
cessfully performed a task of memory selection on which
behaviorally spontaneous confabulators had failed.
A continuing mystery is that only a small minority of
patients having a type of brain damage that normally accounts
for significant momentary or behaviorally spontaneous con-
fabulation will actually confabulate beyond the confusional
state. This is true for ruptured aneurysms, diencephalic tumors,
traumatic brain injury, and alcoholic Korsakoff syndrome.
Amazingly, only a few percent of these patients subsequently
present with behaviorally spontaneous confabulation. Also,
most patients with this syndrome eventually stop confabulat-
ing and regain a correct sense for reality. It seems that the
structures essential for reality adaptation in thinking, whose
failure accounts for behaviorally spontaneous confabulation,
are redundantly organized.
Clinical Course
There are very few studies describing the clinical course of con-
fabulation, and most of these are case reports. In degenerative
disorders, confabulatory tendency may increase over time,
although no specific pattern has been described. If confabula-
tion is due to acute brain damage, it normally decreases over
time and transitions between different forms of confabulation
may occur. Patients who produce fantastic confabulations in
an acute confusional state (delirium) may subsequently cease
to confabulate, produce momentary confabulations, or suffer
the reality confusion characterizing behaviorally spontaneous
confabulation. The latter transition has only been described in
patients having damage to the posterior medial orbitofrontal
area or structures directly connected with it. Patients having
behaviorally spontaneous confabulation may simply stop to
confuse reality and to confabulate; others may stop to produce
inappropriate acts but still produce momentary confabula-
tions in response to questions. Momentary confabulations
may occur in the course of a memory disorder with no transi-
tion to other forms of confabulation. These observations indi-
cate that the different forms of confabulation are at least partly
independent disorders rather than different degrees of a com-
mon disorder. Confabulatory behavior of any kind often
ceases despite continuing amnesia.
The course of behavioral spontaneous confabulation has
been studied more specifically. Virtually all patients eventually
stop to confuse reality. They stop confabulating and regain
correct orientation about time, place, and their current role.
The duration of the confabulatory phase seems to depend on
the lesion site. If a patient has behaviorally spontaneous con-
fabulation beyond termination of the initial confusional state
and has a lesion that involves the posterior medial orbitofron-
tal area, the confabulatory phase may last many months. In a
notable case, the transition from the state of reality confusion
to appropriate behavior occurred within one week after
18 months. If the lesion does not directly involve the posterior
medial orbitofrontal area, there is a high chance that the con-
fabulatory phase lasts less than 3-6 months. Very rare cases
with permanent reality confusion over several years have been
described. Clinical improvement is not reliably associated with
the recovery of amnesia or of executive dysfunction (although
these functions, in groups of patients, tend to improve). The
improvement is, however, associated with recovery of the
capacity to distinguish between memories that pertain to ongo-
ing reality and memories that do not. The evaluation of this
capacity will be described below.
There is no specific treatment with proven efficacy for
behaviorally spontaneous confabulation. Single patients were
trained to use diaries and notebooks with the goal of teaching
them to respect a schedule. Another goal was to have them take
notes whose belated study would provide evidence to the
patients that their thoughts were false. In severe reality confu-
sion, this approach - while worth a trial - is not efficacious.
Some patients may profit from neuroleptic treatment, whose
effect can be estimated within a few days. In our experience,
patients should not permanently be confronted with the fact
that their ideas are false. It is more useful to provide an envi-
ronment of protected freedom where patients can express their
(false) ideas about reality and freely move around without
putting themselves in danger, for example, by stepping into
the street while they still have attentional deficits. Hospitaliza-
tion in a specialized unit with the possibility to prevent the
patients from leaving may be necessary.
568 Confabulation and Reality Filtering
Mechanisms of Confabulation
Since the early descriptions of confabulation, at the end of the
nineteenth century, investigators have speculated about its
mechanisms. Common themes have been that confabulators
fill gaps in memory (gap-filling hypothesis), may do so to
obtain satisfaction (motivational hypothesis), have the combi-
nation of amnesia with a mental control disorder (executive
hypothesis), have a problem in the association of elements
constituting a memory (reconstructive hypothesis), fail to verify
whether a memory refers to a real event in the past or not
(monitoring hypotheses), or confuse information from distinct
periods (temporal order hypothesis). Most hypotheses were
derived from observation alone; controlled evidence has re-
mained extremely scarce. In addition, many theoreticians made
no difference between the diverse forms of confabulation and
suggested that their theory was valid not only for all forms of
mnestic confabulations, but also for nonmnestic confabulations
and false recognition. The following is a brief summary of the
prevalent hypotheses and their empirical bases.
Gap-Filling
One of the earliest interpretations of confabulation, especially
momentary confabulations, was that the patients fill gaps in
memory to avoid embarrassment. Arnold Pick proposed a
special version of the gap-filling account in which he inter-
preted gap-filling as a physiological process similar to the
filling of the blind spot in vision. This process would serve
the coherence of memories. Gap-filling is still proposed in
present-day textbooks.
As far as empirical data exist, they do not support the gap-
filling account. Diverse studies have shown that provoked
confabulations, momentary confabulations, or behaviorally
spontaneous confabulations do not correlate with the ten-
dency to respond to questions about items that do not exist
or which the patients cannot reasonably answer. Their ten-
dency to respond to such questions does not differ from non-
confabulating amnesics. Thus, confabulators do not have an
indiscriminate tendency to fill gaps in memory. In addition, as
described above, even severe confabulators may not necessarily
have a 'gap in memory' if their encoding and retrieval capacity
is tested with a conceptually simple recognition task. Thus,
while gap-filling may be a reasonable explanation for certain
instances of confabulation, it is definitely not a generally valid
account of any type of confabulation.
Motivational Hypothesis
An early variation of the gap-filling idea held that elaborate
confabulations (intense momentary or fantastic confabula-
tions) may provide satisfaction to the patient. The idea has
also been applied to nonmnestic confabulations, in particular
falsely favorable accounts of one's hemiplegia (anosognosia) or
patients' beliefs that they were near their home rather than at
hospital (reduplicative paramnesia) . These nonmnestic confabu-
lations were interpreted as embellishment of the true situation,
motivated by the wish to be in a better state than the real one.
The motivational hypothesis has recently been revived.
Indeed, patients with apparently elaborate momentary
confabulations after lesions in the orbitofrontal area tended
to produce more positive accounts of their past and present
than healthy subjects asked to invent responses to similar
questions. The idea was again that the wish to embellish reality
might be an inducer or modifier of confabulations. It is unclear
whether the comparison with intentionally invented ideas by
healthy subjects is valid. Also, the account does not explain
why false memories come up in the first place. Finally, the
positive emotional bias of confabulations may also depend
on sociocultural circumstances: Korsakoff noted that confabu-
lations in his patients mostly referred to deaths and funerals.
Executive Hypothesis
A very early observation was that confabulation was not
explained by memory failure alone but required additional
mental impairments, nowadays called executive dysfunctions.
Single cases have been described in whom fantastic or momen-
tary confabulations decreased with improvement of executive
functions. Such findings are, of course, not specific. Diverse
group studies showed that brain-damaged subjects who con-
fabulated had overall worse executive functions than patients
who did not confabulate. In the absence of an appropriate
matching criterion, this observation just shows that more
severe brain damage is likely to produce more severe deficits,
without any specific conclusion about confabulation. Also,
executive dysfunction is such a broad class of failure that the
specificity of the account is unclear. Importantly, diverse stud-
ies showed that behaviorally spontaneous confabulators do
not differ from nonconfabulating patients with similarly
severe amnesia on a select series of executive functions. Thus,
momentary and behaviorally spontaneous confabulations are
not explained by general executive dysfunction. Fantastic con-
fabulations have only been described in disease states charac-
terized by severe and multifaceted executive dysfunction
(delirium, severe dementia, severe schizophrenia), but the
association has never been formally explored.
Provoked confabulations appear to require some form of
cognitive flexibility: among similarly severe amnesics, the
number of intrusions in memory tests was positively correlated
with measures of mental drive and relatively better retrieval
success in memory testing. It appears that provoked confabula-
tions at least partly constitute the pay-off for a particularly
broad search in an imperfect memory.
Monitoring Hypotheses
Korsakoff suspected that confabulations were based on some
defect in the association of memories, a defect that would
nowadays be called a disorder of reconstructive processes.
Where precisely the defect lies, has never been specified.
Many authors, including early ones, thought that confabula-
tions resulted from an inability to verify whether an upcoming
memory relates to a real event (reality monitoring) or to
verify the memories' provenance (source memory). Although
this interpretation is phenomenologically hardly contestable,
empirical evidence is lacking. While confabulating patients
who failed in reality and source monitoring have been
described, nonconfabulating amnesics often appear to have
the same difficulty. Thus, the knowledge about when and
Confabulation and Reality Filtering 569
from what source a memory was acquired in the past is
not critical for the occurrence of confabulation. Hitherto, no
experimental paradigm able to separate between patients hav-
ing any form of confabulation and an appropriately matched
nonconfabulating brain-damaged control group has been
demonstrated. Nonetheless, as will be shown further down,
behaviorally spontaneous confabulation can indeed be inter-
preted as a defect of preconscious filtering of upcoming mem-
ories, a defect that might be conceived as a failure of a
preconscious, real time or on-line monitoring process.
A particularly elaborate version of a monitoring hypothesis
is the strategic retrieval hypothesis by Moscovitch and cow-
orkers. It describes a chain of processes, subserved by specific
medial temporal and frontal structures, that leads up to correct,
or incorrect, memory retrieval. A critical structure is the ventro-
medial prefrontal cortex, which together with the frontal pole
would convey the feeling for the rightness of memories. The
hypothesis was tested by comparing groups of confabulating
and nonconfabulating patients who were asked to recall per-
sonal events upon a cue or to recount in detail fairy tales or
bible parts. Confabulating patients indeed produced more
erroneous recollections. Although the authors suggested that
these findings explained behaviorally spontaneous confabula-
tion, the limitation to a singular pathology (rupture of an
anterior communicating artery aneurysm in one study) leaves
open the possibility that the findings simply reflect the severity
of cognitive failures in general. In an earlier study, confabulat-
ing patients had indeed had more severe executive dysfunction.
More importantly, semantic memory need not be affected in
behaviorally spontaneous confabulation. Thus, this hypothesis
pertains particularly to momentary confabulation. It suggests
that - at least in the explored etiologies - the severity of
momentary confabulation is also measurable by the number
of confabulations in semantic memory. However, it does not
clarify the precise mechanism or stage of processing whose
defect is responsible for confabulation. Of note, other authors
used similar experimental evidence (errors in fairy tale recall)
to support the temporal order hypothesis of confabulation,
which will be discussed next.
Temporal Order Hypotheses
An observation already made by early clinicians was that con-
fabulations very often consist of bits and pieces from real
events in the patients' past, but assembled with incorrect tem-
poral order. This element, by the way, relates to most instances
of all forms of confabulation, possibly with the exception of
intense fantastic confabulations that defy any concept of logic.
An early suggestion was, therefore, that elaborate confabula-
tions reflected a loss of the temporal order in memory. Single
case observations supported this view. However, an inability
to know in what temporal sequence information was previ-
ously seen also occurs in nonconfabulating amnesic patients
and in patients having prefrontal lesion with otherwise intact
memory. Thus, the simple knowledge about when in the past
an event happened is not relevant for the occurrence of
confabulation.
A more recent refined version of the temporal order
hypothesis holds that confabulating patients have lost the
temporal frame in consciousness - they have a pathological
awareness of personal temporality. Similar to the monitoring
hypotheses, the temporal order hypothesis formulated this
way can hardly be contested, but the authors did not limit
the theory to any specific form of confabulation and provided
no experimental paradigm to separate confabulating from
nonconfabulating amnesics or to explore the mechanisms
behind the shift in personal temporality.
Nonetheless, in conceptual agreement with monitoring
and temporal order hypotheses, experimental evidence on
behaviorally spontaneous confabulation has indeed shown
an inability to refer memories to ongoing reality, and in that
sense a defect of monitoring or loss of the temporal frame in
thinking. The mechanism, which will be discussed in the next
paragraph, is specific for behaviorally spontaneous confabula-
tion; no evidence exists that it applies to other forms of
confabulation.
Confabulation and Reality
Confabulations are - first of all - verbal descriptions of false
memories about events and experiences which never took
place. In most instances, these false verbal statements relate
to past events and do not necessarily influence the way in
which patients perceive ongoing reality. This is different with
behaviorally spontaneous confabulations: the patients not
only recount recent experiences which have not taken place
in reality, but also falsely interpret ongoing reality. This confu-
sion of reality is manifested by measurable disorientation
about time, place, and current role, and in the fact that the
patients act according to their falsely perceived reality. In most
instances, they enact previous habits, such as, going to work.
Thus, behaviorally spontaneous confabulation constitutes more
than a false verbal account of recent experiences: it is a disorder
of reality perception. Its study may help us understand how the
brain recognizes that a memory refers to ongoing reality, or not.
Given its anatomical specificity, it also represents a natural
human lesion model for the orbitofrontal influence on reality
perception.
Confabulation and Memory Suppression (Filtration)
We found that behaviorally spontaneous confabulators fail to
suppress the interference of memories that do not pertain
to the 'now.' We arrived at this conclusion using the following
experimental task: patients performed repeated runs of a con-
tinuous recognition task in which they saw a long series of
pictures among which they had to recognize repetitions. Most
importantly, all runs were composed of exactly the same pic-
ture series. In this way, the first run demands learning and
recognition of new information. From the second run on,
subjects are familiar with the pictures; now, the task demands
the ability to realize that a picture - although it appears
familiar - comes up for the first or a repeated time within
the ongoing run.
We found that the performance of healthy subjects and
nonconfabulating amnesics remained constant over the differ-
ent runs. Behaviorally spontaneous confabulators, in contrast,
had a specific increase of false positive responses, that is, they
indicated more and more frequently that they thought they
had just seen a picture in the ongoing run, when indeed they
570 Confabulation and Reality Filtering
had seen it in a previous run, up to 45 min earlier. Thus, they
failed to suppress the interference of their currently irrelevant
memory of an item's appearance in a previous run. This failure
also very highly correlated with the degree of orientation,
measured with 20 questions about orientation to time, place,
current situation, and personal information. In addition, this
failure individually paralleled the clinical course of the reality
perception of the patients, irrespective of the persistence of
amnesia. Translated to normal thinking, the results meant
that the ability to realize what day it is, where we are, and
what our current role is, depends much more on the ability to
fdter out ideas that do not pertain to ongoing reality than on
the ability to learn new information.
Normal Reality Filtration
The same experimental paradigm was used in adapted form
to study reality filtration in healthy subjects. In agreement
with the anatomical data in patients, who had lesions of
the posterior medial orbitofrontal cortex or directly connected
structures, we found that healthy subjects performing a sim-
ilar task while their brain activation was monitored using
H 2 [15]0-PET, activated the medial temporal lobe as they
learned the new pictures in the first run. In subsequent runs,
however, the posterior medial orbitofrontal cortex, in particu-
lar area 13, started to activate, reflecting the need to select,
rather than acquire, currently relevant memories. A subsequent
study using an even more difficult task showed that this selec-
tion process not only activated orbitofrontal area 13, but also
subcortical structures involving the head of the caudate
nucleus, the substantia nigra, and the medial thalamus. These
structures participate in frontal-subcortical loops; some of
them are part of the dopaminergic reward system. A study
using evoked potentials showed that this selection or filtering
process occurs at an early stage of processing, between 200 and
300 ms. Refined analysis suggested that filtering of memories
that do not pertain to ongoing reality are possibly conveyed by
transient inhibition of synchronization in temporo-parietal
association cortex between 200 and 300 ms. In contrast to this
early response, the distinction between new and repeated infor-
mation in the first run, reflecting learning and recognition, was
expressed in amplitude modulation at about 400-600 ms.
These findings indicated that the filtering of memories accord-
ing to their relation with ongoing reality (200-300 ms) occurs
before the recognition, and presumably the re-encoding, of
information, which occurs at 400-600 ms. In other words,
before we even know what we think, the brain (orbitofrontal
cortex) has already decided whether an upcoming memory
relates to ongoing reality or constitutes a fantasy.
Reality and Extinction Capacity
What might be the mechanism behind this filtering capacity? The
hypotheses on confabulation described in the last paragraph
would probably suggest prefrontal high-level monitoring pro-
cesses. This appeared unlikely to this author. Indeed, the 'inven-
tion' of memory in evolution would appear to be a dangerous
step if an organism were not endowed at the same time with a
mechanism allowing it to immediately sense whether a memory
relates to ongoing reality, and might therefore guide behavior, or
whether it constitutes a mental association having no current
behavioral relevance, a fantasy. We, therefore, suspected that
the filtering of currently relevant memories depends on a phylo-
genetically ancient faculty. One striking characteristic of behav-
iorally spontaneous confabulation is the incorrigibility of
inappropriate action plans. Although patients may be constantly
at odds with the hospital environment which should normally
convince them that they are not at a business meeting or at
home, they continue to act according to their false reality. Thus,
the patients continue to act on the basis of ideas (anticipations)
which may have been valid in the past but which do not refer to
current reality. This behavior is reminiscent of the behavior of
animals that continue to act upon a previously reward-predicting
cue although it has stopped to be followed by reward. In techni-
cal terms, we suspected that behaviorally spontaneous confabu-
lation reflects deficient extinction capacity. Primate experiments
showed that the only prefrontal area whose damage induces a
specific deficit of extinction is orbitofrontal area 13. Single cell
recordings showed that this area, which is constantly damaged
or disconnected in our patients, contains a particularly high
density of neurons that specifically fire in response to the nonoc-
currence of expected reward.
In agreement with this interpretation, we recently found
that disorientation and behaviorally spontaneous confabula-
tion are very strongly associated with a deficit of extinction
capacity, as measured by a simple, explicit anticipation task.
A similar task performed by healthy subjects induced strong
activation of the orbitofrontal cortex (pole and area 13) and
induced specific electrocortical potential variations at 200-
300 ms. These findings suggest that, rather than invoking
high level monitoring processes, the brain uses an ancient
biological faculty - extinction - to keep thinking and behavior
in phase with reality.
Conclusion
Confabulation is not a unitary phenomenon. Both functional
and anatomical dissociations can be recognized. At the current
stage of knowledge, four forms of memory-related confabula-
tions with distinct mechanisms can be distinguished:
Provoked confabulations, manifested as intmsions in memory
tests, appear to reflect a normal response to a faulty mem-
ory. They are more frequent in brain-damaged subjects but
also occur in healthy subjects. They have no specific anato-
mical basis.
Momentary confabulations evoked by the context of a discussion
or by questions are the most common form of confabula-
tion and may relate to autobiographical or semantic mem-
ory. No specific mechanism or anatomical basis has been
described, although they seem to be more frequent after
ventromedial prefrontal lesions, similar to behaviorally
spontaneous confabulation. Momentary confabulations
are probably a group of different disorders with different
mechanisms and different clinical courses, but these sub-
types have yet to be determined.
Fantastic confabulations are a rare phenomenon occurring in
a delirium (confusional state), advanced dementia or psy-
chosis. Their mechanism or mechanisms have never been
specifically explored.
Confabulation and Reality Filtering 571
Behaviorally spontaneous confabulation constitutes a confusion of
reality as reflected in inappropriate acts according to the
confabulation and in disorientation. They seem to emanate
from the defect of an orbitofrontal filtering mechanism,
which probably relies on extinction capacity, that filters
upcoming memories according to their relation with ongo-
ing reality at an early stage of memory evocation, before the
precise content is recognized.
Future studies should specify the type of confabulation
under investigation and substantiate claims about mechanisms
by controlled, empirical data.
See also: Agnosia (including Prosopagnosia and Anosognosia);
Amnesia and the Brain; Memory.
Further Reading
Dalla Barba GF (2002) Memory, Consciousness and Temporality. Boston, MA: Kluwer.
Johnson MK, Hashtroudi S, and Lindsay DS (1993) Source monitoring. Psychological
Bulletin 114: 3-28.
Korsakoff SS (1891) Erinnerungstauschungen (Pseudoreminiscenzen) bei
polyneuritischer Psychose. Aiigemeine Zeitschrilt fur Psychiatrie und psychisch-
gerichtliche Medicin 47: 390-410.
Loftus EF (1979/1996) Eyewitness Testimony. Cambridge: Harvard University Press.
Moscovitch M and Melo B (1997) Strategic retrieval and the frontal lobes: Evidence
from confabulation and amnesia. Neuropsychotogia 35: 1017-1034.
Nahum L, Ptak R, Leemann B, and Schnider A (2009) Disorientation, confabulation, and
extinction capacity. Clues on how the brain creates reality. Biological Psychiatry
65: 966-972.
Schnider A (2003) Spontaneous confabulation and the adaptation of thought to ongoing
reality. Nature Reviews Neuroscience 4: 662-671.
Schnider A (2008) The Confabulating Mind. How the Brain Creates Reality. Oxford:
Oxford University Press.
Weinstein EA and Kahn RL (1955) Denial of Illness: Symbolic and Physiological
Aspects. Springfield, IL: Charles C. Thomas.
Conflict Communication
D D Cahn, State University of New York at New Paitz, New Paltz, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Aggressive conflict communication Conflict
communication consisting of name calling, swearing, put
downs, verbal threats, and menacing/controlling/painful
physical behaviors.
Communication process Senders and receivers arrive at
similar meanings through the exchange of messages
(symbols and signs).
Conflict A difference or incompatibility between people,
which becomes significant when it has adverse effects on
their relationship and/or escalates to verbal abuse and
physical violence.
Conflict communication behavior Conflict management;
observable interaction that occurs when people are in a
conflict situation.
Mediation Action in which a neutral third party intervenes
in a dispute to facilitate communication between the
conflicting parties.
Personal relationships Friends, colleagues, neighbors,
romantic partners, spouses, and family members.
Strategy A cognition; a mental or psychological state; a
disposition; a preference for a way of responding to other
people during a conflict.
Conflict may be defined as a difference of opinion or incom-
patibility between people. However, it becomes significant
when it has adverse effects on their relationship and/or esca-
lates into aggressive conflict communication.
Conflict communication may be viewed as a process that
includes antecedents and consequences. Antecedents occur at
an early stage of potential conflict communication wherever a
difference of opinion or incompatibility exists. Conflict com-
munication becomes an actuality when one or more parties are
aware of the existence of this difference or incompatibility or
believe that the incompatibility exists. Consequences occur at
the final stage of development when the conflict threatens the
relationship because each of the partners perceives that signifi-
cant, undesired sacrifices must be made to satisfy the other or
the conflict communication escalates into verbal abuse and
physical violence.
According to researchers, there are at least three ways in
which conflicts may escalate and do harm to a relationship.
First, the more excited and heated the conflict (in terms of
physiological arousal, especially for men), the more likely the
partners are to disengage from their relationship during the next
few years. Second, some patterns of conflict are more disastrous
to the relationship in the long run even if they appear more
desirable at the beginning. Third, certain nonverbal behaviors
during conflict (e.g., woman's disgust, man's miserable smile,
etc.) predict relationship breakups later. The fact that certain
communication behaviors and ways of dealing with conflict
are associated with relationship dissatisfaction and breakups
necessitate a better understanding of conflict communication.
A better understanding of the subject is achieved by placing
relevant theory and research within a particular research per-
spective or tradition. This is no easy task for two reasons. First,
when attempting to put together empirical findings on the
subject of conflict communication, one discovers that the
extensive research on conflict in personal relationships appears
disorganized. Numerous empirical studies in several related
disciplines, such as psychology, sociology, communication,
and family, and published research reports appear to reflect
different orientations to doing research, use different types of
measures, and reflect the jargon of their particular disciplines.
The first step then is to identify the major perspectives on
gathering data and to organize the relevant studies along the
lines of these major research approaches.
A second problem is that more than one research perspec-
tive or approach appears to dominate the study of conflict in
personal relationships. Scholars view the nature of conflict
communication differently, depending on their particular
research approach. Therefore, this article describes conflict
communication and its role within the context of four different
research approaches.
The System-Behavior-Interactionist Approach to the
Study of Conflict Communication Behavior
According to the system-behavior-interactionist research
approach, poorly managed conflicts increase in degree and
complexity, and may go beyond differences regarding a specific
problem, issue, or argument because of the emotional nature
of many close relationships. In its most negative form, conflict
is expressed as an emotionally charged communication behav-
ior pattern that typically involves anger and hostility, escalates,
and harms the personal relationship.
On the Nature of Conflict Communication
Conflict communication from this perspective may be viewed as
verbal and nonverbal interactions between persons expressing
opposing interests, concerns, or opinions. Partners' conflict com-
munication is categorized as constructive or destructive accord-
ing to whether it results in mutually satisfying outcomes, or
escalates and harms the partners' emotional ties.
Research Question Asked in This Approach
Researchers who work within the system-behavior-interactionist
paradigm have invested considerable time and effort, attempting
572
Conflict Communication 573
to answer the following question: How do partners resolve
their differences without doing harm to their personal
relationship?
Typical Case for Data Collection
Usually, researchers who adopt this approach to the study
of conflict in personal relationships focus on dyadic interaction
and typically recruit romantic partners, spouses, family mem-
bers, or friends, create a conflict situation, audiotape or video-
tape their interaction, and compare the conflict communication
patterns of those who are satisfied with their relationship with
the patterns of those who are not. Recording the interactions
between partners in conflict appears to be useful for subsequent
examination of the interactions and for the identification of
constructive and destructive dyadic communication patterns.
Relevant Theories and Key Concepts
Relevant theories view communication and conflict as stochastic
processes in which the data for research take the form of a
series of events that occur in a sequential pattern, preceding
events constraining the probabilities of subsequent ones. Part-
ners' conflict communication may be compared to a game in
which the move of one partner constrains the alternatives
available to the other, thereby reducing uncertainty in predict-
ing subsequent moves or plays. Because of the interdependent
nature of behaviors during conflict communication (where
antecedent behaviors influence the probabilities of subsequent
behaviors), the system-behavior-interactionist approach is
particularly useful for examining communication behaviors
that are linked to one another in an escalating process.
Sources of Dyadic Conflict
A partner's behavior is said to be problematic when it
produces conflict communication. From the system-behavior-
interactionist perspective, problematic behaviors are of interest
because they fit the assumption that subsequent behaviors
(i.e., conflict communication) are dependent on preceding
behaviors (sources of conflict). Because one way to deal with
conflict is to eliminate the behavior that leads to it, it is neces-
sary to first identify the problematic behavior.
Efforts to identify problematic behaviors in personal relation-
ships have produced lengthy lists. Moreover, the same behaviors
are not always perceived as problems by both men and women,
and different behaviors appear to be perceived as problems at
different stages in the development of a relationship.
While some researchers describe a number of problematic
behaviors that lead to conflict communication in personal
relationships, others focus more on the escalation of conflict.
Once the conflict is under way, behavioral interaction mea-
sures are useful for observing and recording the interrelated
behaviors that occur during conflict and lead to escalation.
Measuring Instruments for Collecting Data in Dyads
A number of dyadic measures are available for observing the
stochastic nature of conflict communication behavior. For
example, the Couples' Interaction Scoring System (CISS)
developed by Gottman consists of eight verbal behaviors or
content codes (disagreement, agreement, mind reading, com-
munication talk, solution proposal, summarizing other, sum-
marizing self, feeling, or problem information) and three
nonverbal behaviors or affect codes (positive, negative, or
neutral). Interestingly, with the exception of agreement, verbal
behaviors (content codes) did not discriminate between dissat-
isfied and satisfied couples, although nonverbal behaviors
(affect codes) did generally discriminate between the two
types of couples.
By listing specific, concrete verbal, nonverbal, affect, domi-
nance, and avoidance behaviors, measuring instruments like
these provide a clear idea of what system-behavior-interactionists
mean by conflict communication behavior and enable them
to objectively observe the escalation of conflict in personal
relationships.
Research Findings
Negative conflict communication is characterized by negative
affect, coercive/controlling, escalating, rigid messages, and
conflict avoidance.
Negative affect
When engaged in conflict, dissatisfied partners rely more on a
type of communication called negative affect (or punishing,
aversive) messages. The measuring instruments described ear-
lier have revealed the following negative characteristics of con-
flict communication:
• Dissatisfied partners display higher rates of aversive beha-
viors, express more criticism/blame/accusations, are less
responsive, and display fewer problem-solving behaviors
and reinforcers than satisfied partners.
• Dissatisfied partners are more likely than satisfied partners
to use cross-complaints and counter-complaints, and less
likely to engage in validation sequences.
• Compared to satisfied partners, dissatisfied partners are
more likely to begin a conflict by directly attacking each
other with criticism and negatively loaded statements, fol-
lowed by attempts by each to justify himself/herself and
blame the other.
Coercive/controlling beha viors
Researchers have revealed that coercive and controlling behav-
ior during conflict interaction (as well as at other times) is
associated with dissatisfying marital relationships. However,
there appear to be sex differences in the use of controlling
messages. In conflict interactions that were behaviorally
coded, men assumed a more coercive stance, while women
took an affiliative or accommodating position.
Escalation: negative reciprocity
Once hostility is expressed by either partner, it is likely to
escalate in frequency over the course of the interaction. In
both dissatisfied and satisfied partners, negative communica-
tion behavior is more likely to be reciprocated than positive.
Of course, dissatisfied partners demonstrate stronger negative
reciprocity than happy ones.
574 Conflict Communication
Early research findings indicate that reciprocity of negative
affect is symmetrical. However, researchers detail different
types of negative affect messages and reveal that most of the
husband's negative affect consists of anger and contempt,
while only a little of the wife's negative affect is anger and
contempt. Most of her negative affect is whining, sadness,
and fear. While a husband may reciprocate his wife's anger,
she does not reciprocate his. Instead, the wife responds with
fear to her husband's anger, suggesting that their relationship is
not symmetrical or fairly balanced.
According to Gottman, dissatisfied partners begin a conflict
verbally with complaints and criticism follow nonverbally with
hostile behavior and end without agreement on a solution.
Rigidity
Research shows that interactions between dissatisfied partners
show a higher degree of structure and more predictability of
one spouse's behaviors from those of the other than is found in
the interactions between satisfied partners.
Conflict avoidance
One way of managing a conflict is the use of indirect methods
such as teasing and humor for expressing unacceptable emo-
tions on threatening or embarrassing topics. Because these
methods serve to remind partners of their bonding, they can
promote solidarity, reestablish intimacy, and excuse a slight.
Not all jokes are equal, however. Benign humor consists of
jokes about the self, about the relationship, or about the part-
ner in a gentle manner. Hostile humor includes jokes about the
partner in a negative way, particularly with sarcasm. Research-
ers report that more satisfied couples use humor not to resolve
sources of conflict, but to bring a verbal disagreement to an end
on a playful note.
Other avoidance behaviors include denial of conflict,
change of topic, contradictory statements about the presence
of conflict, statements that direct the focus of conversation
away from the conflict issues, and abstract, noncommittal,
and indirect statements. Researchers have observed a strong
reciprocity tendency in partners' interaction in that avoidance
behaviors were typically followed by other avoidance beha-
viors. They have also found that less satisfied couples engaged
in fewer avoidance behaviors than more satisfied partners. At
least one researcher reports that conflict avoidance in marriage
is a complex topic in that different types of married couples use
conflict avoidance to different degrees and in different ways,
ranging from hostile to friendly.
There may be a difference in the way men and women
engage in conflict avoidance behavior. According to Gottman,
men show a larger autonomic nervous system (ANS) response
to stress, respond more readily, and recover more slowly than
do women. If extreme ANS response is viewed as harmful,
unpleasant, and undesirable, then men might be more
inclined than women to avoid situations that would be asso-
ciated with repeated ANS activation. Thus, men may become
more conciliatory and are likely to avoid conflict or terminate
it by withdrawing.
Other researchers have observed that both husbands and
wives are more likely to be demanding when arguing over a
change one wants in the other, and as likely to be withdrawing
when arguing over a change their partner wants them to make.
Although neither sex indicated that it would be more demand-
ing than the other sex, data revealed that, overall, men with-
draw more than women.
Conflict Training: Helping Couples Resolve Issues and
Improve Their Intimate Relationships
Some behaviors lead to more positive outcomes than
others, that is, some behaviors de-escalate emotional out-
bursts. For example, discussion that focuses on resolving a
problem is less emotionally upsetting than name calling and
partner blaming.
How can one learn or be taught de-escalation behavior in a
conflict? Behavior modification may be used to reinforce posi-
tive behaviors but not negative behaviors. In addition to
behavioral modification, training in conflict communication
skills is directed toward enhancing emotional ties within the
personal relationship. By observing their own concrete verbal
and nonverbal behaviors on videotape, partners learn how
they convey interest or disinterest, and warmth or coldness,
as well as attraction or repulsion.
The system-behavior-interactionist approach accounts for
many empirical research studies of conflict in personal rela-
tionships. In addition to its productiveness as a research para-
digm, it has also generated many useful techniques for helping
partners improve their personal relationship by encouraging
positive, constructive, and integrative conflict communication.
The Rules-lnterventionist Approach to Mediated
Conflict Communication Rules
The notion of mediation as conflict communication may at
first appear novel. Although opportunities for interaction seem
limited by the presence of an outsider (a mediator), disputing
partners may exchange information, influence one another,
and develop mutual agreements. However, mediation may
not be appropriate for every dispute. Where effective, media-
tion facilitates communication between partners, is future ori-
ented, and often includes renegotiation of the personal
relationship. Thus, mediation is a unique form of conflict
communication in which partners may participate in a more
formal conflict resolution process than they normally encoun-
ter in their everyday lives.
On the Nature of Conflict Communication
A dispute is defined as conflict communication when peo-
ple are unable to resolve the issue by themselves, and
normal relations are unlikely until the dispute is resolved.
The rules-interventionist approach views conflict communica-
tion as mediation in which a neutral third party creates and
enforces certain rules that enable partners to resolve their
dispute. Mediators strive for the development of reasonable
outcomes for all concerned. Mediation is normally seen as
an alternative to adjudication, in which a judge imposes a
decision upon the disputing parties.
Conflict Communication 575
Research Question Asked by Rules— Interventionists
The rules-interventionist approach to the study of mediated
conflict has generated studies designed to answer the question:
What role does a third party play in helping partners learn how
to resolve issues? This research approach is useful for incorpor-
ating the actions of an impartial third party or mediator along
with those of the conflicting partners to determine the effects of
the role of the mediator and the partners on the outcome of the
mediation process.
Typical Case for Data Collection
Understanding helps personal relationships and interaction over
time. For one or both partners, however, understanding may no
longer hold, resulting in a breakdown in communication
between them. Researchers who take a rules-interventionist
approach to the study of conflict communication call attention
to the role of a third party in helping partners communicate,
resolve disputes, and restore or redefine their relationship as the
typical case involving partners in personal relationships. For
example, mediators may help divorcing spouses legally dissolve
the marriage, restructure the family, and create new relationships
in which former spouses can cooperate with each other (on
matters such as custody and visitation) long after the divorce.
Relevant Theories and Key Concepts
Because mediated conflict presents a context or type of rela-
tionship entirely different from that of dyadic conflict commu-
nication as discussed earlier, the study of mediation appears to
need a different research approach. The rules-interventionist
approach is particularly useful because it is designed for three
interacting roles - two partners and a mediator - and the effect
on the outcome of each person's role.
Unlike the system-behavior-interactionist approach, in
which stochastic probability governs the outcome of interac-
tion, in the rules-interventionist approach, the mediator inter-
venes to place interaction under the control of rules designed to
limit emotional conflict, reduce uncertainty and surprises, and
encourage open communication, cooperative problem solving,
and equitable treatment.
Successful mediators begin mediation by announcing com-
munication rules. These rules go beyond explanation of the pro-
cess and legality of mediation, and place limits on the agenda and
the tone of the discussion. Ruled out are common negative verbal
tactics such as name calling, verbal abuse, and introduction of
irrelevant issues and nonverbal tactics such as screaming, threat-
ening gestures, standing up, or refusing to look at each other.
Mediation theorists view personal relationships as complex,
private systems and as delicate entities that must be respected. They
assume that partners are in the best position to decide mutually
acceptable outcomes with minimal intervention by others. How-
ever, while partners determine issues and make decisions for
themselves, the mediator enforces the communication rules.
Sources of Mediated Conflict
There is a difference between the sources of conflict that give
rise to particular behavioral dyadic interaction patterns and
those that give rise to mediation. Compared to private conflicts
that tend to occur in the privacy of the home or involve only
the partners themselves, mediated conflict is seen more as a
social or public event because of the presence of the third party
as mediator. In the case of divorce mediation, the sources of
conflict may be reduced to five types.
• Control over resources : money, property, space, children, and
so forth that may be viewed as not sharable.
• Preferences and nuisances', activities of one person impinge
on another's sensibilities, preferences, or sensitivities.
• Values: concerning what should be, for example, religious
and ideological issues.
• Beliefs: concerning what is, for example, fact and reality.
• The nature of the relationship: repair and redefinition of
personal relationships.
Mediation may be rated and evaluated by using triadic
interaction measures to observe and record conflict communi-
cation. One such measure was created by Donohue.
Measuring Instrumenfs for Collecting Data in Triads
Donohue and his colleagues developed a measure for coding
partners' and mediators' verbal acts. They distinguished between
two types of verbal acts: cues to subsequent utterances and
responses to prior utterances. In addition, the partners' verbal
acts were identified according to the following categories:
• Attacking: personal accusations; negative evaluations.
• Defending: presenting proposals, providing information,
requesting information or clarification, or providing a
rationale for one's position.
• Integrating: any utterance that agreed with the prior utter-
ance, demonstrated support for a proposal or position.
Finally, Donohue and his colleagues added the following
categories to code the mediator's interventions.
• Structuring: identifying and enforcing interaction rules.
• Reframing: helping to restructure disputants' own propo-
sals; pointing out areas of agreement.
• Expanding/requesting: requesting proposals; requesting clar-
ification of proposals; requesting reaction to proposals.
By applying both the coding schemes of Donohue to
recorded mediation, one might better understand the role of
the mediator and how her or his actions help to settle disputes.
Research Findings
Researchers have applied different interventionist measures
to essentially overlapping data derived from 80 audiotapes of
divorce mediation. Using these tapes for coding triadic interac-
tion, researchers found that successful mediators were more
likely than unsuccessful mediators to use more intense struc-
turing and reframing interventions in response to attacks. The
more successful mediators were also more likely to rephrase
negative comments and turn them into more positive ones.
Researchers have shown that successful mediators were more
in control of the mediation, used more interventions to involve
the divorcing partners in finding the information necessary for
agreement, and distributed more of these interventions fairly
and consistently between the disputants.
576 Conflict Communication
Other researchers report that the most important difference
between cases that were settled successfully and those that were
not was the amount of time disputants expected mediators to
spend discussing the final settlement. In the successful cases,
mediators not only felt obligated to spend more time discuss-
ing the terms of the final agreement, but were also expected to
spend more time discussing possible solutions in general. They
were not expected to spend much time explaining the media-
tion process to the spouses, requesting disclosure of feelings,
and making statements about the parties' attitudes.
In all, researchers discovered that the following patterns of
actions implied that rules were operating:
• Speaker time was fairly evenly distributed among the two
spouses and the mediator.
• Mediators tended to address both spouses, whereas husbands
and wives usually directed their remarks to the mediator.
• Mediators tended to ask more questions than did either of
the parties.
• Mediators expressed statements on procedural issues three
times more than did the parties.
• Mediators conveyed more information on what mediation
is and made more statements regarding the process itself
than did the parties.
• Mediators made the most statements summarizing the
spouses' comments. Spouses made more emotionally
toned statements than did the mediator.
• One-third of the statements made by both spouses were self-
disclosures. A number of statements were about the attitudes,
motives, and actions of others, usually the other spouse.
• Perhaps in an effort to establish rapport and encourage the
spouses to share their feelings, mediators made statements
that showed more empathy than did those uttered by the
spouses.
• Mediators attempted to balance proposals by specifying
how both parties could be involved, while each spouse
tended to specify what he or she could do.
• Very few statements were classified as interruptions (where
the thought was not resumed).
Divorce mediation sessions that ended in agreement
appeared to progress from differentiation to integration
through three phases: information exchange, problem solving,
and finally resolution actions.
Structured Mediation: Helping Partners Settle Disputes
To help both parties emphasize cooperation and gain some-
thing from mediation, third-party interventionists need a thor-
ough understanding of mediator goals and tactics that include
structuring the process of mediation, reframing the disputants'
positions, and expanding the information resource.
The Cognitive Approach to the Study of Conflict
Communication Strategies
When thinking about the topic of conflict communication in
personal relationships, one probably envisions partners
engaged in a verbal disagreement. Cognitive researchers view
conflict communication in an entirely different perspective - as
mental (cognitive) strategies that range from preferences for
direct confrontation to the desire to avoid conflict. As these
strategies are mental or psychological in nature, the process of
conflict communication is not observable. From the cognitive
perspective, to view conflict communication as a process
means that one must ask partners to give their own retrospec-
tive reports on the selection of conflict communication
strategies.
On the Nature of Conflict Communication
The cognitive approach views conflict communication as stra-
tegies that are internal responses to perceived sources of con-
flict. Because some conflict communication strategies involve
intention to avoid open confrontation, conflict communica-
tion may exist even when partners are not overtly disagreeing.
Research Questions Asked in This Approach
Questions of interest asked by cognitive researchers include the
following: What antecedent conditions influence one's choice
of a conflict communication strategy? What alternative strate-
gies are best for dealing with different sources of conflict? What
beliefs underlie one's predisposition to engage in aggressive
conflict communication? This research approach is useful for
examining the role played by predispositions in the form of
conflict communication strategies in a broad, general model of
relationship development and the effects of these strategies on
the outcome of the conflict.
Typical Case for Data Collection
Cognitive research focuses on developing relationships
(including romantic partners, friends, coworkers, and room-
mates and also offspring) as the typical case for data collection.
Relevant Theories and Concepts
Some researchers take the position that a great deal of conflict
can be explained in terms of attribution processes and efficacy
expectations. Attribution refers to the process of assigning rea-
sons or motivations for particular behaviors, while an efficacy
expectation refers to the belief that a problem can be solved by
confronting it. These researchers argue that attributions of
blame or intent give rise to hostile behavior, such as retaliation.
However, they go on to argue that regardless of one's attribu-
tions of cause, responsibility, and blame, given low efficacy
expectations, a partner is unlikely to want to resolve the con-
flict and will prefer to avoid it or withdraw from it. Given high
efficacy expectations, however, the person is likely to under-
take such efforts and confront the problem. In such cases,
resolution attempts are directed toward that which is perceived
as most easily changed and likely to produce satisfactory
results.
Sources of Conflict In Developing Relationships
Perceived imbalance in the resources of exchange, perceived
inequity, and perceived unequal distribution of power in
a relationship likely lead to feelings of relationship
Conflict Communication 577
dissatisfaction which itself may be a source of conflict. While
studying relationship dissatisfaction in general as a source of
conflict, researchers use self-report techniques to assess the
degree of dissatisfaction in the relationship. In this approach,
cognitive variables like relationship dissatisfaction are believed
to be capable of producing other effects. Thus, self-reported
relationship dissatisfaction may be seen as a cognition that
influences the preference for particular confrontation and con-
flict avoidance strategies.
Measuring Instruments for Collecting Data in Developing
Relationships
One of the most well-known lines of research on alternative
strategies for managing conflict in intimate relationships actually
grew out of business administration/management research.
Rahim proposed a typology for the management of organizational
conflict, consisting of the following five strategies: dominating
(forcing, competing), integrating (confronting, collaborating),
obliging (smoothing, accommodating), avoiding, and compro-
mising (give and take) styles that reflected either a high or
low degree of concern for oneself and concern for the other
(e.g., partner).
Other researchers view aggressive conflict communication
as a self-reported potential for verbal abuse. A few researchers
have determined the propensity toward aggressive conflict
communication on the basis of scores achieved on the Dys-
functional Beliefs Assessment and on the Child Abuse Inven-
tory. Scores on these measures operationally defined potential
verbally abusive parents.
Research Findings
There appear to be important male-female differences that affect
partners' perceptions of rewards and costs and their role in
conflict. Men appear to differ from women in the way they
perceive conflict that in turn affects their relationship satisfac-
tion. At least one researcher has asked dating partners to keep
daily observation records of conflicts over a 2-week period.
Such research reveals that the perception of many unresolved
conflicts is most salient to women's perceived relationship
satisfaction, whereas the perception of a large number and the
stability of conflicts (i.e., the same issues coming up again and
again) are most salient to men's relationship satisfaction.
Other male-female differences in perceptual processes have
been found to have an impact on conflict in intimate relation-
ships by distorting the intent of the message sent. For instance,
the behavior of dissatisfied spouses compared to that of satis-
fied spouses is perceived more negatively by partners than
intended. This confirms an earlier finding that dissatisfied hus-
bands are more inclined to attribute 'negative connotations' to
their wives' attempts to communicate affection, happiness, and
playfulness than satisfied husbands.
There is evidence that partners may tend to prefer the strat-
egy of avoidance - a tendency for partners in personal relation-
ships like marriage to avoid dealing with conflicts. Yet, a
number of research findings support open confrontation as a
constructive conflict communication strategy. For example, at
least one study reports that black couples and white couples
who believe in avoiding conflicts reported lower marital
happiness in the first year of marriage and again 2 years later
than those couples who believe in confronting conflicts.
A number of factors that contribute to preferences for
open confrontation or conflict avoidance strategies have been
observed. People are more likely to prefer to confront when
they are committed, equal in power, prepared for confrontation,
and the issue is important. Men are more likely than women to
blame their mates for problems, avoid emotional involvement,
and prefer coercive conflict communication as their way of
dealing with conflict. Some personality types are more prone
to open confrontation and others to avoidance. Culture and
immediate social context are also influencing factors. For exam-
ple, there is a tendency among male and female managers to
prefer more competitive strategies of conflict resolution when
dealing with workers and more accommodating sUategies with
their partners at home.
More recently, researchers found patterns of attributions
about a child's behavior that result in dysfunctional child-rearing
beliefs, automatically producing in mothers verbally abusive
responses to their child's (mis)behavior. Later, researchers related
mothers' child abuse potential with the mother and child's
play behavior and found that children of high-risk mothers
(for aggressive conflict communication behavior) displayed
higher levels of involvement combined with lower levels of
cooperation compared to children of lower-risk mothers.
Educating Partners on How to Manage Conflicts
Some of the sources of conflict result from faulty perceptions
or inferences. Training in cognitive restructuring helps partners
deal with the sources of conflicts and improve their relation-
ship. Essentially, cognitive restructuring is a means of increas-
ing relationship satisfaction by relabeling faulty cognition, that
is, misperception and unwarranted inference. Of particular
interest are perceptions of undesirable behavior and inferences
made from them. Cognitive restructuring includes helping
partners see the connection between their behavior and a
partner's feelings, identifying myths and accepting reality,
emphasizing the positive aspects of their relationship, and
training in empathy.
The Qualitative Research Approach to the Study of
Conflict Communication Perceptual and Evaluative
Processes
The qualitative approach can help researchers better under-
stand the sense-making processes parties use to legitimize
their conflict communication behavior. To do this, the
qualitative research approach relies more on interpretive
means of collecting data. By using methods such as partici-
pant observation and unstructured interviews, the qualitative
approach applies interpretative techniques to narratives (per-
sonal stories) and open-ended interviews of people who are
in conflict situations. In their search for multiple interpreta-
tions, qualitative researchers are known to consider 'alternative
voices,' especially the victims of domestic abuse. To increase
real-world relevance, qualitative researchers seek to make the
research experience a part of the subjects' natural habitat, such
as the home environment. It is argued that the understanding
578 Conflict Communication
of conflicting parties can assist practitioners, social workers,
family, and psychotherapists working with conflicting parties,
especially in violence-prone premarital, marital, and parent-
child relationships.
On the Nature of Conflict Communication
Conflict communication from this perspective is redefined as
each party's subjective perception or assessment of the quality
(or the impression) of their own and the other's conflict com-
munication behavior. Implications of this approach are (1)
What might seem appropriate verbal and nonverbal behavior
to one person may not seem so to the other, and (2) What
might be appropriate verbal and nonverbal behavior in one
situation may not be appropriate in another.
Research Question Asked in This Approach
Researchers who work within the qualitative research para-
digm attempt to answer the following question: How do
conflicting parties perceive their own and the other's commu-
nication behavior in a conflict situation? This includes the
attempt to understand how they legitimize their own behavior
as well as how they evaluate their own and the other's conflict
communication behavior.
Typical Case for Data Collection
Researchers who adopt this approach can either observe as
participants in the conflict situation or interview one or both
parties in conflict. In cases of potentially aggressive conflict
communication, parties may be interviewed separately; in less
threatening situations, however, the parties may be interviewed
together. In some studies, people may watch videotapes of
their conflicts and provide insight into why they did what they
did, how they felt, and what they were hoping to accomplish.
The parties involved may be in a premarital, marital, or parent-
child (including elderly parent- adult and parent- adolescent)
relationship. Recording the interactions of partners in conflict
appears to be useful for interpretation later.
Relevant Theories and Key Concepts
According to this approach, conflict communication is the
product of people sharing and creating meaning in a conflict
situation. Social constmction theory attempts to gain access to
the phenomenology of people's lives. The ways in which people
think and use categories to structure their experience and analyze
the world are constituted in their communication. Because peo-
ple leam from the society and times in which they live, conflict
communication behavior is said to be socially prescribed.
Ethnography provides descriptions of behavior in cultural
groups: behavior that conforms to rules, roles, rites, and rituals.
This type of study might take the form of a field study or a case
report. As participant observers, ethnographers usually spend
time in another society or cultural subgroup, living with the
local people and learning about their way of life to better under-
stand their social behavior and thinking processes. They may
also apply their methods to families, groups, or couples. Ethno-
graphers focus on relationships and social structure, verbal and
nonverbal language codes, practices, rites, and rituals. Ethnogra-
phy is useful in identifying rules that govern conflict communi-
cation in specific communities, groups, or couples.
Grounded theory is inductive in nature, suggesting that the-
ory emerges from the data. While quantitative research takes a
deductive approach (one begins with a theory and then tests or
examines it by collecting data), grounded theorists begin by
collecting data (observe video taped conflicts) . Next they identify
a series of codes from the text. Using what they call the constant
comparison method, they group these codes into similar cate-
gories that may serve as a basis for the creation of a theory.
Story telling and narratives have developed into a theory
and study of narrative and narrative stmcture, and the ways
they affect people's perception of events. By collecting individ-
ual stories from particular couples/families and other groups,
organizations, and cultures; by listening to and comparing
different accounts; by examining how narratives are con-
structed around specific events; by learning which events in a
culture, subculture, or group's history generate stories and
which ones do not, quantitative researchers gain a deeper
understanding of the subjects' realities, which are closely
linked to their social experiences. By doing so, they discover
the group's political, social, and cultural dynamics.
Sources of Dyadic Conflict
Dialectical tensions are said to lie at the heart of conflict
between partners in a personal relationship. The relationship
is supported or threatened by the way people manage the
dialectical tensions of autonomy versus connection (cohesion)
and stability versus change (adaptability) in their everyday
lives. People who experience difficulty managing their conflicts
are viewed as less balanced in their management of the dialec-
tical tensions than other people, who have more resources,
such as communication skills. Those less skilled are also
more likely to repeat old patterns rather than learn new ones.
Measuring Instruments for Collecting Data in Dyads
Purely qualitative researchers do not use measures beyond
nominal categories, such as those classifications used in con-
tent analysis. For example, they may interview couples (living
together) and victims in shelters for battered women. In this
case, the narratives of people in dysfunctional relationships may
be compared with those in more 'normal' relationships. How-
ever, other qualitative researchers use measurement at the ordi-
nal, ratio, and interval levels when they incorporate various
existing measures of relationship satisfaction, commitment, or
trust to identify target groups, whose members they are inter-
ested in interviewing. While they may or may not use higher
level measurement tools for the independent variables, both
groups of qualitative researchers are not inclined to use measures
and statistical methods for analyzing the dependent variables.
Research Findings
A researcher reported that an aggressive communication
encounter is a process of 'escalating antagonism' between per-
son A and person B: A's action upsets B; B orders A to cease the
action; A fails to comply; B and A verbally abuse each other,
Conflict Communication 579
making way for physical violence. An implication of this obser-
vation is that the conflict communication may increase in
intensity until aggressive conflict communication and even
physical violence appear justified to at least one of the parties.
In a study of ten couples with a history of abuse and ten
couples without, cohesive and adaptability dialectical tensions
consisted more specifically of vague versus precise language,
opposition versus collaboration, relational versus content talk,
despair versus optimism, interfering versus facilitating interde-
pendence, complaints versus compliments, and effective versus
ineffective change. The greater the tension on these seven
dimensions, the less cohesive and less able they were to adapt
to the other.
Other researchers interviewed 15 abusive male spouses to
better understand how they perceive themselves and their
spouses, and to account for their abusive behavior. The men
provided accounts for their abuse in the form of justifications,
excuses, denials, or efforts to minimize the significance of their
acts, and blamed their wives for their abusive behavior. Mean-
while, another researcher revealed that abused women also
provide accounts of their mates' behavior, but they dissociate
the 'real him' from his abusive acts - that is, he was not being
himself. In both cases, accounts provided by the partners
revealed that abusers and their victims perceive the abusive
acts differently but in ways that attempted to excuse the abuser.
By interviewing 3 1 people, another researcher attempted to
better understand the sense-making process couples use to
legitimize use of aggression. She found that many of the parti-
cipants assessed their use of verbal and physical aggression
according to social customs, finding it inappropriate and inef-
fective generally. However, many noted their aggression was
appropriate and effective, given the circumstances of their
relationship. These findings suggest a unique relational culture
where verbal and physical aggression are perceived as appro-
priate, even though such acts deviated from the social norm.
Researchers report that interpersonal violence is more likely
when there is verbally aggressive conflict communication. One
study found that the first episode of adolescent-to-parent abuse
experienced by parents consisted of verbally aggressive conflict
communication. Later, these adolescents progressed to more
severe types of emotional or physical abuse.
Family researchers found that mothers' patterns of attribu-
tions about their child's behavior resulted in dysfunctional
child-rearing beliefs and aggressive conflict communication.
This cognitive approach has identified factors such as beliefs
and attributions, recognized abusers' violence-prone predispo-
sitions, and described the mental beliefs and attributions
needed to enable one to avoid aggressive communication
behavior in conflict situations.
Educating Partners to Better Manage Conflicts
Findings from qualitative research help to resolve existing
social problems. They may or may not contribute to the devel-
opment of theory. Qualitative research lends some support to
the effectiveness of treatment programs for those who abuse
others with aggressive conflict communication. Results from
these studies provide some evidence that a change in how an
abuser perceives his or her responsibility and thinks about the
other may persist into the future, if reinforced by his or her
relationships.
Although the perspective view conflict in personal relation-
ships, each perspective has its own definition of conflict,
theories, and research methods:
• system-behavior-interactionists define conflict as a stochas-
tic process relying on behavioral interaction theory using
dyadic behavioral research methods;
• rules-interventionists define conflict as rule-governed activ-
ity relying on rules theory using recorded triadic interaction
research methods;
• cognitive theorists define conflict as a predisposition to
behave relying on social exchange, network, and perception
theories using subjective self-report research methods;
• qualitative researchers define conflict as a subjective
perception/assessment of the quality of one's own and
other's conflict communication behavior relying on social
construction theory, ethnography, and grounded theory
using subjective interviews and narratives as data.
Although researchers tend to adopt one of these views of
conflict, they should realize that each perspective adds to a
better understanding of the antecedents and consequences of
conflict communication behavior.
See also: The Clinical and Cognitive Psychology of Conflict; Divorce;
Equity Theory; Family Systems; Marital Dysfunction; Parent-Offspring
Conflict; Sex Differences; Social Exchange.
Further Reading
Bevan JL, Tidgewell KD, Bagley KC, et al. (2007) Serial argumentation goals and their
relationship to perceived resolvability and choice of conflict tactics. Communication
Quarterly 55(1): 61-77.
Cahn D (ed.) (1994) Conflict in Personal Relationships. Hillsdale, NJ: Erlbaum.
Cahn DD (2009) Friendship, conflict and dissolution. In: Encyclopedia of Human
Relationships. Thousand Oaks, CA: SAGE, http://www.sage-ereference.com/
human relationships/Article_n226.html.
Cahn D (ed.) (2009) Family Violence: Communication Processes. Albany, NY: SUNY.
Cahn D and Abigail RA (2007) Managing Conflict Through Communication.
Boston, MA: Allyn & Bacon.
Donohue WA (1991) Communication, Marital Dispute, and Divorce Mediation.
Hillsdale, NJ: Erlbaum.
Koerner AF and Fitzpatrick MA (2002) You never leave your family in a fight: The impact
of family origin on conflict-behavior in romantic relationships. Communication
Studies 53(3): 234-251 .
Lee S (2008) Mediation technigues of an informal Intermediary in intercultural-
Interpersonal conflict. Human Communication 11(4): 461-482.
Olson LN (2002) “As ugly and painful as it was, it was effective:” Individuals’ unique
assessment of communication competence during aggressive conflict episodes.
Communication Studies 53(2): 171-188.
Rogan RG and La France BH (2003) An examination of the relationship between verbal
aggressiveness, conflict management strategies, and conflict interaction goals.
Communication Quarterly 51(4): 458-469.
Wilson SR, Morgan WM, Hayes J, Bylund C, and Herman A (2004) Mothers’ child
abuse potential as a predictor of maternal and child behaviors during play-time
interactions. Communication Monographs 71(4): 395-421.
Conformity and Obedience
D J Packer, Lehigh University, Bethlehem, PA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Conformity Typically defined as occurring when perceived
or actual social pressure from others causes a person to alter
their opinions, feelings, or behavior. These changes may be
public, private, or both.
Descriptive norms Modal patterns in how people
(usually in specific groups or contexts) actually think,
feel, and behave.
Dissent Nonconformity expressed or exhibited with the
intention of changing a norm.
Informational influence A psychological mechanism
underlying conformity in which people rely on others for
information about the world. Is motivated by a desire for
accuracy and efficacy.
Injunctive norms Widely held conceptions regarding how
people (again, usually within specific groups or contexts)
should think, feel, and behave.
Normative conflict A state of disagreement with group
norms, descriptive or injunctive.
Normative influence A psychological mechanism
underlying conformity that results from a desire to meet the
expectations of others. Is motivated primarily by the need
for approval.
Obedience Occurs when a person acquiesces to the request
or instruction of someone in authority, changing their
opinions, feelings, or behavior.
Social identification A psychological mechanism
underlying conformity that involves adopting a
normative group prototype as a guide for thought,
feeling and action. Is motivated by a desire to express a
collective identity, and by solidarity and certainty goals.
Under certain circumstances, social identification can also
motivate dissent.
Investigations of conformity and obedience are foundational to
social psychology. Indeed, early studies examining these phe-
nomena conducted by Solomon Asch (conformity) and Stanley
Milgram (obedience) rank among the most famous psychology
experiments ever conducted. Both sets of studies are still taught
to thousands of college students each year, and policy makers
and media commentators routinely draw ostensible lessons
from each. This article reviews the literature on conformity and
obedience, with particular attention to their underlying psycho-
logical mechanisms. Whereas a great deal of research attention
has been paid to conformity and there is vast post- Asch empirical
literature, the same is not the case for obedience. For a variety of
reasons (including the impossibility of ethically replicating his
work, at least exactly), our understanding of obedience has not
advanced a great deal since Milgram's day.
Conformity
Nineteenth century social scientists were highly attuned to the
influence that social pressures could exert on opinions, beliefs,
and actions. In his book, Extraordinary Popular Delusions, and
the Madness of Crowds, journalist Charles Mackay described
a series of movements in which popular opinion widely, if
sometimes only briefly, swayed in strange and foolish direc-
tions. In a variety of crazes, people showed sudden and mass
passions for, among other things, investing in tulips, for
alchemy, and for peculiar shapes of beard. Beliefs regarding
sensible economic decisions, natural and supernatural phe-
nomena, and masculine ideals appeared to be highly flexible
and contagious, spreading in waves among a malleable popu-
lace. Thus, writing in 1903, Tarde stated that 'society is imita-
tion' and 'social man is a veritable somnambulist.'
The advent of modern psychology in the twentieth century
saw the application of scientific methodology to the investiga-
tion of the psychological processes that underlie conformity.
Muzar Sherif conducted a series of studies in the 1930s that
took advantage of a visual phenomenon known as the auto-
kinetic effect. When people are seated in a darkened room and
a stationary point of light is directed at the opposite wall, the
light appears to move about in arbitrary directions. Employing
this ambiguous stimulus, Sherif asked participants to estimate
how far the light had moved and found that their distance
estimates were influenced by the estimates of others. Experi-
mental stooges giving prespecified answers were able to shift
participants' estimates by several inches. There were limits,
however, to how far estimates could be pushed around, sug-
gesting that social influence was subject to reality constraints,
and that participants' responses reflected a combination of
both their own impressions and the ostensible observations
of others. Subsequent research showed that the responses of
experimental confederates established a type of cultural norm
that persisted after all original group members had gradually
been replaced by new participants. However, because partici-
pants in each generation combined their own impressions with
the impressions of others, over successive generations the
influence of the norm was steadily attenuated. Still, the experi-
mentally manipulated norm did exert significant power over
responses, and this research gave credence to notions of
human relations as inherently conformist.
The history of psychological research on conformity can
itself be described in terms of normative trends, punctuated
by occasional reactions against them. Solomon Asch led the
first such reaction. Asch recognized that groups and societies
could not expect to function healthily if driven only by con-
formist processes, and that nonconformity was essential to
580
Conformity and Obedience 581
creativity, innovation, and social change. Nonconformity by
this formulation is a group-oriented and cooperative act, and
given its utility, is a behavior that ought to occur with some
frequency in human interaction. Asch noted that the autoki-
netic and related investigations of conformity employed
ambiguous stimuli, which put individuals in the awkward
position of having no firm ground to stand on when disagree-
ing with the responses of others. In a classic series of studies,
Asch gave participants a less ambiguous task to perform, which
he posited would reveal significantly higher rates of disagree-
ment. Naive participants were placed in a group with a number
of confederates and were asked to make a series of visual
judgments, matching the length of a line to one of three
possibilities. These judgments were exceedingly simple, such
that control participants not privy to the responses of others
responded with over 99% accuracy. Experimental participants,
however, were exposed to others' decisions that were manipu-
lated to be uniformly incorrect on critical trials prior to making
their own response. In the face of unanimity from the rest of
the group, participants made significantly more errors, indicat-
ing a degree of conformity to an incorrect social norm. Criti-
cally, however, rates of conformity on critical trials in these
and subsequent studies hovered around 30-40% and, consis-
tent with Asch's contention, the modal response was clearly
nonconformity.
Given Asch's intentions, it is an historical irony that his
work is now most often used to make the opposite point and
invoked as an illustration of the extraordinary power of social
influence on decision making. Whereas what Asch and his
immediate contemporaries found most compelling was strong
evidence for disagreement in the face of social pressure, the
majority of contemporary references to these experiments
focus on conformity as the important finding. As such, the
bulk of subsequent research has focused on (a) identifying
domains in which conformity accounts for important patterns
of behavior and (b) understanding the psychological mechan-
isms underlying conformity. Both enterprises have been highly
successful. In the former case, conformity has been identified
as an influential force in almost every conceivable realm of
human behavior, including eating, consumer choice, sexual
behavior, prejudice, humor and fashion. A variety of contribut-
ing mechanisms, which will be reviewed below, have also been
identified. In contrast, less attention has been paid to processes
underlying nonconformity (although we do know quite a lot
about its effects).
Mechanisms Underlying Conformity
Normative Influence
Several classes of mechanism have been proposed to account
for conformity to social norms. Normative influence refers to
conformity that results from a desire to meet the expectations
of others. The motivation underlying conformity in this case is
a need for approval. There is a substantial evidence for norma-
tive influence processes, such that conformity increases in
situations where others are likely to notice and disapprove
of deviant behavior, and decreases in the absence of these
factors. When participants are able give anonymous responses
in the Asch paradigm, for instance, rates of conformity show a
significant decline, although conformity does not disappear
entirely. Similarly, individuals are more likely to conform to
collective norms when they value the group in which they are
situated or when they believe that the group values them.
The influence of Asch's work is nowhere more apparent than
in the fact that his conformity paradigm has been replicated in
countries and cultures around the globe. A meta-analysis of
these studies examined national rates of conformity as a func-
tion of each country's level of individualism/collectivism.
Whereas in individualist cultures, it is considered socially desir-
able for decisions to be driven by personal opinions and goals,
in collectivist cultures it is socially preferable to place collective
opinions and goals ahead of one's own. Consistent with nor-
mative influence, participants in countries with higher levels of
collectivism were found to display greater conformity than
participants in individualist nations.
Punishment of Deviance
The type of norms that underlie normative influence have been
termed injunctive norms, which refer to widely held concep-
tions of how people should think, feel, and behave. These are
collectively held standards regarding what is appropriate and
inappropriate. Normative influence derives part of its force
from the fact that people have a striking willingness to enforce
these sorts of norms. There is a great deal of evidence, both
anecdotal and experimental, that groups react negatively to
deviants, punishing norm violations by imposing social costs
(e.g., ridicule, rejection), as well as physical and economic
hardships. Real-world examples of punishment for devi-
ance abound, particularly in religious and political contexts.
In these domains, dissention from official doctrine earns the
dissenter a variety of pejorative labels, including heretic and
traitor. These are dangerous names to bear, and those who
incur them have a distressing tendency to come to sticky
ends: condemned in show trials, burned at the stake, assassi-
nated, etc.
Early psychological research found that individuals who
expressed divergent opinions in small groups were liked less
and received greater attention (e.g., were the targets of more
persuasive attempts) than individuals whose opinions matched
the group norm, and this was particularly the case in cohesive
and valued groups. Generally speaking, the more strongly iden-
tified members are with a group, the more likely they are to
punish deviance. The propensity to impose heavy costs on
deviance in political and religious contexts may reflect the fact
that these tend to be particularly cohesive and valued groups.
However, it is also likely due to the fact that these sorts of groups
are, at least in part, defined by the possession of common
beliefs. In groups that are defined by sharing a particular set of
beliefs (e.g., regarding the role of the state, abortion or the
meaning of the sacraments), anyone who challenges a core
belief poses a threat not only to that particular norm, but also
to the meaning and viability of the group as a whole.
A related phenomenon known as the black sheep effect
illustrates the group-based nature of punishment for deviance.
This line of research finds that identical negative acts are
regarded more negatively and result in greater rejection when
they are committed by an ingroup member as opposed to an
outgroup member. As such, it is not the behavior per se that
582 Conformity and Obedience
people are responding to, but rather behavior by members of
their own group. There are two potential reasons for this. First,
although certain types of injunctive norm (expectations for
appropriate behavior) may be applied to everyone regardless
of group membership, other norms are regarded as group
specific, and punishment is confined to those to whom the
norm is perceived to apply. Second, the bad behavior of
ingroup members reflects poorly on the group, and other
members are motivated to punish and prevent these sorts of
acts in order to maintain a positive group image and reputa-
tion. As in the research described above, strongly identified
group members exhibit the black sheep effect more strongly.
The mechanisms by which cooperation among members of
a species evolves has been a persistent puzzle in evolutionary
psychology. Although cooperation often bestows collective
benefits (e.g., we all benefit if everyone pays their taxes and
we can afford safe roads and fire departments), it is, all else
being equal, in any individual's interest to avoid cooperating
(i.e., to defect). If I do not pay my taxes but everyone else does,
I keep a higher proportion of my income and also benefit from
safe roads and fire service. Critically, however, if it is in every-
one's individual interest to avoid cooperation in this fashion,
rates of cooperation will rapidly decay. Anyone who does
cooperate will be exploited by others and will suffer reduced
fitness; thus, orientations toward selfish behavior are more
likely to be selected for than orientations toward cooperation.
Recent research has shown, however, that cooperation can be
sustained by sanctions for noncooperation, and particularly by
third-party punishment in which people who are personally
unaffected by noncooperation nevertheless punish defectors.
If others reliably punish individuals who attempt to cheat the
system, it can change the cost-benefit ratio (and fitness levels)
in favor of cooperation. Individuals appear to be willing to
punish noncooperators even at some cost to themselves
and when there are no costs to not punishing. This behavior
has been observed across a wide variety of societies and cul-
tures, although rates of punishment vary quite substantially.
It appears that a propensity to punish noncooperation may be
partially a genetic predisposition that is itself shaped by local
norms and can thus serve as at least a partial solution to the
cooperation problem. All things considered, punishment of
those perceived to act against the collective interest seems to
be a deeply ingrained reaction.
There are, however, boundary conditions to the punishment
of deviance. Recognizing that some degree of divergence and
dissent is often functional for groups (as will be discussed in
further detail below), researchers have recently suggested that
groups may not be as allergic to deviance as is often assumed.
The intergroup sensitivity model focuses specifically on group
criticism and suggests that criticism of an ingroup is likely to be
well received and even persuasive to the extent that it is perceived
to be constructive. Critically, whether or not criticism is per-
ceived to be constructive is dependent on the person who
expresses it; in particular, their standing in the group. Members
who are believed to have the interests of the group at heart
because, for instance, they have been part of the group for a
long time, are given the benefit of the doubt and are responded
to positively when they express concern regarding the group and
its norms. Similarly, high-status group members have greater
latitude in the extent to which they can deviate from group
norms before incurring disapproval and sanction. In contrast,
relatively recent or low status group members, as well as the
members of outgroups, elicit negative reactions when they direct
criticism toward a group.
Informational Influence
Another mechanism underlying conformity is known as
informational influence, and reflects a reliance on others
for information about the world. Many situations are more
ambiguous than the Asch paradigm, and it often makes sense
to compare one's own impressions or opinions with those of
others in order to better understand the nature of reality.
The central motive underlying informational influence is a
desire for accuracy. Under many circumstances it is reasonable
to assume that although individuals are subject to errors
and biases, aggregating information from across individuals
is likely to yield a more accurate representation or understand-
ing of what is correct. The type of norms that often underlie
informational influence have been termed descriptive norms,
and these reflect modal patterns in how people actually think,
feel and behave. The notion is, quite simply, that individuals
use information from others in combination with their own
impressions, preferences or beliefs when rendering judgments
and making decisions. The fact that conformity does not dis-
appear (although it is reduced) when participants in Asch-style
paradigms can respond entirely anonymously provides evi-
dence consistent with the informational influence mechanism.
It has further been shown that as the difficulty or ambiguity of
a task increases, conformity also increases, presumably because
individuals are less confident in their own perspectives and
turn to others for information. Importantly, this effect is stron-
ger when the motivation for accuracy has been increased by
providing participants with incentives for correct responding.
Informational influence processes have been incorporated
into the economic literature, where theorists have modeled infor-
mational cascades. An informational cascade occurs in sequential
decision-making situations when individuals possess informa-
tion about others' prior decisions (e.g., whether they saw movie
X or movie Y last night), but not necessarily the reasons for those
decisions. A typical cascade model might assume that people
weight their own opinions equally with the opinions implied
by the actions of others. In this model, when there is disagree-
ment between the two (e.g., I am leaning toward seeing movie X,
but I know that you went to see movie Y) the rational response is
to engage in a mental coin toss, weighting each option equally.
If, as a result, I end up seeing movie Y, our mutual friend Bob can
infer with a 0.75 probability that both you and I were of the
opinion that movie Y is the superior choice. In this situation, even
if Bob privately believes the opposite, it is rational for him to take
his date to see movie Y the next night. Although it is unlikely that
people engage in precisely these rational computations, the
notion of informational cascades highlights the important point
that conformity is not necessarily an irrational response; there are
many situations in which it may be rational to weight others'
information as heavily or perhaps even more heavily than one's
own. Put differently, thinking rationally does not necessarily
represent an antidote to conformity. These models show that
rational computations can result in conformity and thus
that conformity is not limited to the somnambulist.
Conformity and Obedience 583
Models of informational cascades further illustrate how
once a pattern of behavior is established, it can become self-
perpetuating, and they can thus account for why some things
(e.g., movies, products, fashions, ideas) become wildly popu-
lar whereas others do not. Cascades can occur when there is
little a priori reason to prefer one thing over another, or even
when there are objective reasons to prefer the choice that does
not become popular. Critically, these models can also account
for the fragility and fluidity of popular trends. As cascades
lengthen, the behavior of others carries less information
about their personal beliefs and preferences. If Bob made his
decision in the manner described above, Cindy can infer noth-
ing about his personal preference from which movie he
selected. As a result, the presence of even a single dissenter
may be enough to break the cascade because the deviant's
choice does provide information about their opinion; in fact,
by diverging from the cascade, the dissenter may signal a
particularly strong and well-formed point of view. If this hap-
pens to coincide with Cindy's own view, she is likely to con-
tinue the dissent. In this situation, even a long unbroken string
of unanimous choices can be attributed to others 'just follow-
ing the crowd.' Breaks in cascades and shifts to new ones may
be particularly likely in contexts where there is a bias toward
recent information or an assumption that newer is better.
Social Identification
An additional mechanism underlying conformity arises from
social identification with groups. This mechanism highlights
the critical importance of group memberships for conformity
processes and stems from the social identity approach to
group behavior. Social identity theory and its descendent,
self-categorization theory, posit that when individuals enter a
group context, they adopt the group membership as a compo-
nent of their self-concept (i.e., as a social identity). Members
will differ in the extent to which they self-categorize in terms of
a particular group membership, but to the extent that they
value a particular group, their sense of self shifts from an
individual to a collective level. The theory posits that group
members undergo a process of depersonalization in which
idiosyncratic characteristics (e.g., traits and personal values)
become less active and exert less influence on behavior, while
shared characteristics representative of the group's prototype
(e.g., norms and collective values) become more active and
more influential. In this formulation, group norms provide
information about what prototypic group members do,
think, and feel, and to the extent an individual is motivated
to be a good group member, he or she is likely to conform to
these norms.
The key motive underlying the social identification mecha-
nism thus does not reflect a simple desire to be approved of by
others, but rather a desire to express an important aspect of
one's identity - to instantiate the characteristics that define and
uphold the image of the group in question. Other motives are
also involved: identified group members may be interested in
maintaining collective solidarity and cohesion because cohe-
sive groups are capable of more efficacious action. Further,
identified group members are interested in maintaining a
clear set of group norms because consensus implies correct-
ness and reduces uncertainty. Thus, the social identification
mechanism can account for the findings, outlined above, that
individuals who value and identify with groups are more likely
to conform to normative patterns of group behavior. This
approach can also explain why identified members are more
willing to derogate and punish other ingroup (but not out-
group) members who violate group norms.
Cognitive Principles
Whether one is talking about normative influence, informa-
tional influence or social identification, norms are knowledge
structures that operate within the cognitive system. In the case
of the Asch paradigm, the normative response on each trial
is presumably computed online as fellow group members ver-
balize their decisions, and the extent to which the norm is
correctly interpreted and the degree to which it influences
behavior is a function of attention. If the norm is not accurately
perceived or is unnoticed, it is unlikely to drive behavior. Many
norms, however, are more stable than this and reflect stored
knowledge about descriptive or injunctive patterns of behavior
(e.g., drivers in Paris often honk their horns, one should not
wear white after labor day). The influence of this stored knowl-
edge on behavior follows standard cognitive principles, such
that it must be activated/accessible to exert effects. Accessibility
maybe relatively chronic or may vary as a function of situa-
tional salience. A series of clever field studies conducted by
Robert Cialdini and colleagues has shown that the behavior of
another individual can direct attention to and make accessible
both descriptive and injunctive norms. When an experimental
confederate was observed to litter (vs. not) in a heavily littered
environment, participants littered more frequently. However,
participants who observed another person litter in a clean
environment (and particularly in an environment where litter
had been neatly swept up) were less likely to litter than if they
had not observed anyone else litter. These findings indicate
that individuals were not simply imitating the behavior of the
other; rather others' actions made salient wider norms regard-
ing littering as implied by the cleanliness of the overall envi-
ronment. The importance of salience and accessibility are
further highlighted by the social identification mechanism,
in which different norms are active and influential in different
group contexts.
Minority Influence
As noted above, the field of social influence itself can be
characterized by long normative trends, interrupted by occa-
sional notes of dissent. Despite Asch's original intent, his work
inspired a great deal of interest in the processes and effects
associated with conformity. This work proved enormously
fruitful; however, by the 1970s it seemed to Serge Moscovici
that the field was too focused on the influence exerted
by majorities. Certainly majorities have powerful effects, but
Moscovici believed that minorities also play an important role
in collective life and can, under the right circumstances, influ-
ence others' opinions and decisions. Moscovici proposed that
people expect to achieve consensus on most if not all issues,
and believe that there is generally one correct perspective to
hold. As such, when someone encounters a perspective that
584 Conformity and Obedience
differs from their own, it generates conflict, which the individ-
ual is motivated to resolve. The manner in which this conflict
is resolved differs depending on whether the discrepant per-
spective comes from a majority or a minority. If it comes from
a majority, a normative influence mechanism similar to that
described above kicks in, such that individuals conform in
order to maintain cohesion and acceptance. In contrast, if
the discrepant perspective comes from a minority, an informa-
tional influence type mechanism operates, such that indivi-
duals are motivated to critically evaluate and perhaps validate
(if there is enough supporting evidence) the minority's
position.
Critically, majority versus minority influence results in dif-
ferent types of opinion change. Because majorities exert influ-
ence via normative pressure, majority influence is more likely
to affect public than privately held opinions. This response is
termed compliance, and reflects the fact that people may pub-
licly express agreement with majority perspectives while main-
taining private doubts. Interestingly, however, to the extent
that they exert influence, minorities are more likely to have
the opposite effects: exerting less influence on public opinions
because individuals are reluctant to associate themselves with a
minority, but greater influence on private opinions because
minorities trigger more rigorous consideration of available
information. A change in private but not public responses is
termed conversion. Generally speaking, there is good evidence
for these hypotheses. Across studies, it is clear that minorities
can exert influence and that they are particularly likely to exert
indirect effects on opinion. For example, a minority opinion
regarding a specific issue may not affect others' perspectives
on exactly that issue, but it is likely to affect others' opinions on
related issues. Further, minorities exert influence to the extent
that they are perceived to be consistent; if one is going against
the grain, it pays not to be wishy-washy.
Related lines of research have examined the effects of mino-
rities and dissenters on group functioning more generally.
Irving Jams' famous analysis of groupthink during the Cuban
Missile Crisis and other events made salient the fact that extreme
cohesion and unanimity in groups can lead to poor, even disas-
trous decisions. Although the groupthink construct itself
remains controversial, what is now uncontroversial is the fact
that diversity and divergence in group settings often leads to
improved collective decisions and group outcomes. Researchers
have found, for instance, that the presence of dissent and/or
diversity increases creativity and innovation within groups, and
reduces the extent to which groups become polarized - moving
them away from rather than toward extreme positions. Interest-
ingly, improvement in these sorts of outcomes does not depend
on the dissenter or minority being correct. Consistent with
Moscovici's model, minorities encourage others to think more
divergently and to consider alternate perspectives; it is the con-
sideration given to multiple possibilities that ultimately
improves decisions. Although the expression of minority views
is largely functional for collectives, it is important to note that
there can be costs in terms of reduced cohesion. Recent research
has shown that when a minority successfully converts a group to
their position (and becomes a majority), the group suffers an
overall loss of cohesion as former majority members reduce
their level of identification with the group, which is not replaced
by an equivalent increase in identification among the new
majority. Interestingly, these ill effects are not observed if the
minority achieves its goals by increasing tolerance for their
position but does not seek to convert other group members.
Mechanisms Underlying Nonconformity
Although the minority influence literature has illuminated the
effects that deviance can have on groups, relatively little research
has investigated the mechanisms or motives that might cause
individuals to express minority opinions and make nonnorma-
tive decisions. Researchers and theorists have generally assumed
(at least tacitly) that nonconformity can be accounted for by the
absence of the mechanisms that underlie conformity. As such,
nonconformity is predicted when individuals are not motivated
to seek the approval of others, do not fear social sanctions, do
not need or trust others' information and/or are weakly identi-
fied with relevant social groups. In each case, a motive that
might trigger conformity is replaced by an orientation toward
independence and autonomy from others. Although indepen-
dence and autonomy are undoubtedly the cause of some non-
conformity, they seem to miss a whole class of nonconformist
behavior in which people act not out of a desire to separate or
distinguish themselves from others but rather act on principle or
out of concern for others - in particular for important groups.
Interestingly, this was how Asch himself conceptualized non-
conformity. Intuiting the importance of diversity and divergence
for collective functioning (intuitions that, as we have seen,
turned out to be correct), he did not believe that nonconformity
was an individualistic act, but rather a cooperative, group-
oriented one. In this conception, nonconformity is often moti-
vated by a desire to see one's groups succeed, and individuals
are likely to express dissent when they think that doing so can
improve collective outcomes.
It is, however, difficult to account for this type of nonconfor-
mity simply in terms of the absence of mechanisms that underlie
conformity, each of which implies social separation. This diffi-
culty is further compounded by the fact that nonconformity
often poses individuals with a social dilemma - specifically, a
public goods dilemma. Public goods dilemmas of all types pit
collective and personal interests against one another, such that
by acting to benefit a group, individuals incur personal-level
costs. We have already encountered an example of one such
dilemma: paying taxes to afford roads and fire departments.
We have further seen how social sanctions can be used to ensure
cooperation; that is, to ensure that people resolve these dilem-
mas in favor of the collective interest. Somewhat ironically,
however, the ubiquity with which sanctions are applied to
deviance may give rise to an entirely new public goods
dilemma when it comes to dissent. Dissent is often associated
with positive collective outcomes - it is an act that, all things
considered, is often beneficial to the collective. At the same
time, however, dissent is often associated with significant
social costs to the individual - individual dissenters risk their
good name, social standing, sometimes even their livelihood
and lives by deviating from widely held norms. As such, would-
be dissenters must often be willing to sacrifice personal-level
interests for potential collective benefits, which seems an
unlikely contingency among autonomous individuals stand-
ing apart from others. For these reasons, conformity and
Conformity and Obedience 585
nonconformity are unlikely to be reciprocal, fully explainable
by a common set of processes working in reverse.
Asch's notions regarding nonconformity have recently been
revived, and researchers have started to attend more closely to
mechanisms that underlie nonconformity. One such approach is
the normative conflict model, which predicts how group mem-
bers are likely to respond to norms under a variety of conditions.
This model represents a refinement and extension of the social
identification approach to conformity outlined above, and pre-
dicts the likelihood of several conforming and nonconforming
responses as a function of three psychological factors:
(a) social identification,
(b) normative conflict (i.e., disagreement with group norms), and
(c) estimates regarding the likely costs and benefits of confor-
mity versus nonconformity.
A great deal of research in the social identity tradition has
shown that when people adopt a group membership as an
important component of their self-concept (i.e., identify with
the group), it alters decision-making processes. When indivi-
duals make decisions outside of group contexts or in groups
with which they are weakly identified, actions and outcomes
are evaluated primarily in terms of their individualistic implica-
tions: how will this affect me? In contrast, when individuals
make decisions in groups with which they are strongly identified,
actions and outcomes are evaluated primarily in terms of their
collective implications: How will this affect us? As a result,
identified group members are frequently observed to resolve
social dilemmas of all kinds by acting in the collective interest,
even in the absence of sanctions.
The normative conflict model posits that identified group
members may respond in the same way to the dilemma posed
by dissent. The key assumption is that all group members are
likely, at one time or another, to experience disagreement or
conflict with group norms. For weakly identified members, this
conflict may stem from a tension between personal and collective
interests (e.g., what is good for me vs. good for us). In contrast, for
strongly identified members, conflicts are more likely to arise
when group norms are perceived to be harmful to the collective
itself. This type of normative conflict can come from discrepan-
cies between descriptive and injunctive norms, such that how
the group is actually behaving is believed to be at odds with the
group's own ideals or standards for behavior. However, norma-
tive conflict might also arise from the application of personally
held standards (e.g., moral beliefs) to collective behavior or even
the application of norms incorporated from other group mem-
berships. Importantly, it is predicted that strongly identified
members are often willing to express dissent if they perceive it
to be in the collective interest, even if they are likely to incur
personal costs for doing so. In contrast, weakly identified group
members are not expected to dissent on behalf of the group, and
are actually predicted to be more conformist than strong identi-
fiers when dissent carries personal costs. A number of recent
studies have provided evidence consistent with these hypotheses.
Types of Conformity and Nonconformity
Although it is possible to talk about conformity versus non-
conformity in general terms, theorists and researchers have
often noted that there are various forms of both. We have
already encountered the distinction between compliance and
conversion, two forms of conformity arising from different
psychological processes and with distinct behavioral conse-
quences. The distinction between private and public beliefs
forms the basis for a number of taxonomies of social response.
The normative conflict model also provides a way of distin-
guishing between different forms of conformity and noncon-
formity; specifically, as a function of social identification (low
vs. high), normative conflict (low vs. high), and estimates of
the pros versus cons of conformity/nonconformity among
individuals in each of these four cells (see Figure 1).
• Weakly identified group members experiencing low levels
of normative conflict (i.e., little disagreement with group
norms) exhibit either indifference or strategic conformity
depending on what they believe to be in their personal self-
interest. Generally speaking, these members are not moti-
vated to conform to group norms (are indifferent), but they
will conform if they expect to be rewarded for doing so or
punished for not.
• Strongly identified members experiencing low normative
conflict exhibit either loyal conformity or strategic noncon-
formity. Generally speaking, these members are happy to
conform to group norms with which they agree, although
under certain circumstances they may see the collective
value of deviance even though they do not disagree with
collective norms (e.g., playing devil's advocate).
• Weakly identified group members experiencing high levels
of normative conflict (i.e., sharp disagreement with group
norms) display either disengagement or personally oriented
dissent. Personally oriented dissent reflects an attempt to
change group norms in ways that would serve one's individ-
ual interests; this is a selfish type of dissent and may be
relatively rare due to the fact that others are highly likely to
punish dissent that is not perceived to be collectively con-
structive. Disengagement reflects a class of response in which
individuals try to distance themselves from a group that they
disagree with; rather than attempting to change the group,
these behaviors set one apart from the collective.
• Finally, strongly identified members experiencing high nor-
mative conflict display either collectively oriented dissent
or uneasy conformity. Collectively oriented dissent reflects
attempts to change group norms in ways that are believed
to benefit the collective. These group members are, how-
ever, not entirely blind to the personal costs associated with
dissent, which at times may result in uneasy conformity to
group norms - one would like to do something to change
them but holds back, deterred by possible consequences.
General Relational Norms Versus Context-Specific
Norms
It is important to distinguish between two additional types
of norms: norms that apply generally within classes of
relationships, and norms that develop as patterns of and
prescriptions for behavior within specific relationships. It is,
for instance, generally normative to conform in collective
contexts - when one joins a group it is with the understanding
that one will learn and adopt normative patterns of action
586 Conformity and Obedience
Low normative conflict
Low
Primarily concerned with
identification
personal costs/gains
High
Primarily concerned with
identification
collective costs/gains
Indifference
Strategic conformity
Loyal conformity
Strategic nonconformity
High normative conflict
Low
Primarily concerned with
identification
personal costs/gains
High
Primarily concerned with
identification
collective costs/gains
Disengagement
Personally oriented dissent
Collectively oriented dissent
Uneasy conformity
Figure 1 Types of conformity and nonconformity predicted by the normative conflict model.
(through socialization). Interestingly, however, this norm of
conformity can itself be revised by group norms, such that in
some groups it becomes normative not to conform. Recent
research has shown evidence for this paradoxical state of
affairs: there are groups in which it is normative to be distinct
and where nonconformity represents conformity. Other types
of relationships also have general norms associated with them.
A particularly powerful general norm is known as the norm of
reciprocity. This norm is most powerful in equality matching
relationships, where the understanding is that all favors done
by one party for another must be reciprocated, sometimes
exactly, at a later point in time. The deeply embedded nature
of this norm is often exploited by marketers and salespeople,
who offer naive victims a small gift, which triggers in them an
urge to reciprocate - by listening to a sales pitch or even
shelling out hard cash. The manifestation of this general
norm can also be shaped by local norms. For example, the
range of others to whom one is obligated to reciprocate is
normatively determined: the small gift trick seems to work
well in North American society, but may be less effective in
cultures where haggling with salespeople is normative and
reciprocity is reserved for friends and family.
Obedience
Obedience to authority also often arises from a general rela-
tional norm. It is typically assumed that people should attend
Conformity and Obedience 587
and acquiesce to the requests or instructions of individuals in
positions of authority. Obedience is, like conformity, ubiqui-
tous in human life. When your dentist tells you to 'open wide,'
you do it even though you know that what is about to ensue is
likely to be unpleasant. Different types of authorities get their
power to influence from different sources. Some obedience
stems from the ability of authorities to affect others' outcomes.
On the negative side, an authority may coerce obedience with
explicit or implicit threats of punishment; on the positive side,
they may reinforce obedience with rewards. However, author-
ity can also arise from other factors, including the legitimacy of
a role (e.g., kings, presidents, police officers), charisma and
attraction (e.g., celebrities, some politicians), and expertise
(e.g., dentists, doctors, professors).
It was at one point widely agreed upon by psychologists
that obedience was a good thing - a virtue to be instilled in
children from an early age. It was also often assumed that
obedience was not a spontaneous behavior, but that it had to
be learned and reinforced. The events of the twentieth century,
however, changed these conceptions. Obedience to authority
became widely recognized as a problem after the Second World
War when social scientists tried to understand how vast num-
bers of seemingly ordinary people became involved in atroci-
ties, culminating in the Holocaust. It became clear that whether
obedience is a virtue or a vice depends on the morality or
malevolence of the relevant authority figures, and authorities
did not come off at all well throughout the last century.
Stanley Milgram initiated a program of research to better
understand the psychological processes involved in destructive
obedience. In a series of studies, participants were instructed by
a lab-coated experimenter (the authority figure) to administer
(ostensible) shocks in 15 V increments to a 'learner' whenever
he made a mistake; these shocks started off mild (15 V) and
rose steadily (to 450 V). As the voltage of ostensible shocks
increased, the learner emitted a series of prescripted vocaliza-
tions, which included increasingly intense expressions of phys-
ical pain (beginning at 75 V), as well as verbal requests to be
released (beginning at 150 V). By the middle, the learner was
screaming in pain, and by the end, he had fallen silent.
Remarkably, in many versions of the study, the majority of
participants was fully obedient to the experimenter and
continued the procedure to the very end, despite considerable
personal discomfort. The original studies were conducted in
the 1950s; however, recent partial replications of the paradigm
reveal comparable rates of obedience even today.
The Milgram paradigm is complex, and the psychological
mechanisms underlying the obedience he observed are multi-
faceted. A variety of mechanisms have been proposed, and
although direct empirical evidence is often lacking, many are
plausible contributors. It has been suggested, for instance, that
obedience was facilitated by the sequential nature of decision
making and the gradual manner in which the shocks increased.
Each decision led to the next, and a foot-in-the-door type of
process may have made it difficult to justify refusing to admin-
ister a shock having agreed to give the previous. Similarly, the
gradual increase in severity may have desensitized participants
to the learner's expressions of pain. Cognitive dissonance
between participants' values and their behavior may have altered
conceptions of the harm they were causing (e.g., 'it can't be as
bad as all that'). Participants may also have engaged in what
Milgram called an agentic shift, giving up their sense of personal
responsibility and allowing the experimenter to take charge. It is
also possible that participants looked to the experimenter to
define the reality of what was occurring. The experimenter was,
after all, supposed to be an expert: if he did not see a problem
with what was happening, perhaps there was no problem.
As with nonconformity, less attention has been paid to the
opposite response: disobedience. In part, this is because rates
of disobedience were relatively low in Milgram's studies, with
only a few noncompliant participants to examine in any given
study. However, a recent meta-analysis combined data across a
series of comparable studies to examine disobedience among a
larger sample. What emerged was a previously undocumented
systematicity in disobedience, such that disobedience was most
likely to occur at the point (150 V) when the learner made his
first request to be released. A much higher proportion dis-
obeyed at this point than at any other across studies, and this
was the single most frequent point of disobedience within each
study. Interestingly, there was no linear relationship between
level of shock and obedience as would be expected if increas-
ingly intense expressions of pain were reliably motivating non-
compliance. These findings strongly suggest that disobedience
to authority in this paradigm was not driven by an empathic
response to the suffering of another, but rather may have been
due to recognizing the learner's first request for release as a
choice point. It appears that those participants who disobeyed
the experimenter at this point may have had a different under-
standing of what was and was not a legitimate use of authority
in the experimental situation. When push came to shove, these
participants gave the learner authority over his own fate, and
were obedient to his rather than the experimenter's request.
However, despite these conjectures, much about Milgram's
effects and about processes of obedience and disobedience
more generally remains, for the moment, unclear.
See also: Attitude Change; Crowd Psychology; Group Dynamics;
Individualism; Persuasion.
Further Reading
Ashe SE (1955) Opinions and social pressure. Scientific American 193: 31-35.
Bikhchandani S, Hirshleifer D, and Welch I (1992) A theory of fads, fashion, custom,
and cultural change as informational cascades. Journal of Political Economy
100: 992-1026.
Bond R and Smith PB (1996) Culture and conformity: A meta-analysis of studies using
Asch’s (1952b, 1956) line judgment task. Psychological Bulletin 119: 111-137.
Burger JM (2009) Replicating Milgram: Would people still obey today? American
Psychologist 64: 1-11.
Cialdini RB, Reno RR, and Kallgren CA (1990) A focus theory of normative conduct:
Recycling the concept of norms to reduce littering in public places. Journal of
Personality and Social Psychology 58: 1 01 5-1 026.
Cialdini RB and Trost MR (1998) Social influence: Social norms, conformity, and
compliance. In: Gilbert D, Fiske S, and Lindsey G (eds.) The Handbook of Social
Psychology , 4th edn., vol. 2, pp. 151-192. New York: McGraw-Hill.
Deutsch M and Gerard HB (1955) A study of normative and informational social
influences upon individual judgment. Journal of Abnormal and Social Psychology
51: 629-636.
Henrich J, McElreath R, Barr A, et al. (2006) Costly punishment across human societies.
Science 312: 1767-1770.
Hodges BH and Geyer AL (2006) A nonconformist account of the Asch experiments:
Values, pragmatics, and moral dilemmas. Personality and Social Psychology
Review V): 2-19.
588 Conformity and Obedience
Hornsey MJ (2005) Why being right is not enough: Predicting defensiveness in the face
of group criticism. European Review of Social Psychology 16: 301-334.
Jacobs RC and Campbell DT (1961) The perpetuation of an arbitrary tradition through
several generations of a laboratory microculture. Journal of Abnormal and Social
Psychology 62: 649-658.
Jetten J and Hornsey H (201 1 ) Rebels in Groups: Dissent, Deviance, Difference, and
Defiance. New York: John Wiley & Sons.
Kruger Jl and Massay AL (2009) A rational reconstruction of misbehavior. Social
Cognition 27: 786-812.
Levine JM (1999) Solomon Asch’s legacy for group research. Personality and Social
Psychology Review 3: 358-364.
Marques JM and Yzerbyt VY (1988) The black sheep effect: Judgmental extremity
towards ingroup members in inter- and intra-group situations. European Journal of
Social Psychology 18: 287-292.
McAuliffe BJ, Jetten J, Hornsey MJ, and Hogg MA (2003) Individualist and collectivist
norms: When it’s ok to go your own way. European Journal of Social Psychology
33: 57-70.
Milgram S (1974) Obedience to Authority: An Experimental View. New York: Harper & Row.
Moscovici S (1980) Toward a theory of conversion behavior. In: Berkowitz L (ed.)
Advances in Experimental Social Psychology, vol. 13, pp. 209-239. New York:
Academic Press.
Nail PR, MacDonald G, and Levy DA (2000) Proposal of a four-dimensional model of
social response. Psychological Bulletin 126: 454-470.
Nemeth C (1995) Dissent as driving cognitive, attitudes, and judgments. Social
Cognition 13: 273-291 .
Packer DJ (2008a) On being both with us and against us: A normative conflict
model of dissent in social groups. Personality and Social Psychology Review
12: 50-73.
Packer DJ (2008b) Identifying systematic disobedience in Milgram’s obedience
experiments: A meta-analytic review. Perspectives on Psychological Science
3: 301-304.
Prislin R and Christensen PN (2005) Social change in the aftermath of
successful minority influence. European Review of Social Psychology
16: 75-111.
Sunstein CR (2006) Infotopia: How Many Minds Produce Knowledge. New York:
Oxford University Press.
Terry DJ and Hogg MA (1996) Group norms and the attitude-behavior relationship:
A role for group identification. Personality and Social Psychology Bulletin
22: 776-795.
Wood W, Lundgren S, Ouellette JA, Busceme S, and Blackstone T (1994) Minority
influence: A meta-analytic review of social influence processes. Psychological
Bulletin 115: 323-345.
Consumer Psychology
S S Posavac, Vanderbilt University, Nashville, TN, USA
F R Kardes, University of Cincinnati, Cincinnati, OH, USA
S Hoeffler Vanderbilt University, Nashville, TN, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Advertising Paid communication in which a sponsor
provides information about products, services, and/or ideas.
Behavioral decision theory The body of research that
investigates how people make decisions.
Brand management Making strategic decisions regarding
the marketing of products and services.
Consumer psychology The study of how individuals form
judgments and make decisions related to the consumption
of products and services, and how marketplace phenomena
affect attitudes, choices, and personal and group identities.
Inference The act of forming beliefs about unknown
attributes about a target based on attributes that
are known.
Information processing Consumers' acquisition and use of
information to help them form judgments of, and make
choices regarding, products and services.
Persuasion The act of trying to influence others' beliefs
and attitudes.
Sales promotion Providing an incentive aimed at
stimulating purchase of products and services either by
intermediaries (e.g., retailers) or the end consumer.
At its most general level, consumer psychology is the study of
how individuals navigate myriad judgments and decisions
related to the consumption of products and services. Research-
ers interested in consumer psychology study phenomena as
diverse as the determinants of what attracts consumers' atten-
tion, how they process incoming information, and how they
integrate it with existing knowledge. Moreover, consumer
researchers consider what is remembered from consumption
experiences, and how these memories guide future judgments
and choices. Similar to social psychologists, constructs such as
attitudes and motivation are extremely important for consumer
psychologists. One key defining feature of consumer psychol-
ogy is that it is both a basic and an applied science, and is
typically taught and practiced in business schools with the
ultimate goal of informing and improving marketing manage-
ment decision-making, and hence organizational outcomes.
Moreover, by understanding the psychological principles that
govern consumption, researchers also hope to contribute to
consumer welfare.
Although consumer psychologists study a diverse set of
phenomena, one contribution of researchers in the field is
that of a consistent picture of the human decision-maker. In
contrast to the assumptions of economic standard value theory,
individuals do not act as dispassionate, strictly rational, calcu-
lators of utility when they make consumer decisions. Instead,
although consumers certainly tend to prefer more utility to less,
they do not use all of the information available to them when
they form judgments and make decisions, they are heavily
influenced by normatively irrelevant aspects of the context in
which consumption takes place, and they routinely respond
in accordance with social pressure, whether actively imposed
by others or imagined. Although consumers' choices are not
often strictly rational as per economists' assumptions, consu-
mers display a variety of consistent and predictable tendencies
in how they acquire, aggregate, and use information in their
decisions that facilitate adaptive navigation of a complicated
world. Consumer psychologists endeavor to understand these
tendencies and processes that influence individuals' judgments
and ultimately guide their choices.
An overview of core research areas and key findings that are
central to consumer psychology follows. After the presentation
of basic research, a summary of important implications of
consumer psychology for managerial decisions is offered.
Core Principles of Consumer Psychology
Attention and Memory
Consumer researchers expend a great deal of effort researching
how to best structure persuasive messages such as advertise-
ments. A number of different models of communication have
been proposed, but the first step in any model is capturing
consumers' attention. This is not an easy task because consu-
mers are exposed to hundreds, if not thousands, of persuasive
messages every day. Some determinants of consumer attention
include personal interest in, and knowledge about, a marketing
stimulus (consumers attend when they have preexisting inter-
est and are highly knowledgeable). Aspects of marketing sti-
muli themselves also determine if they will garner attention.
Marketing stimuli that are novel, vivid, concrete (i.e., easy to
visualize), and close versus distant in sensory, temporal, and
spatial proximity are typically readily attended.
Of course, just because an individual has attended to a
stimulus does not imply that it will be processed, transferred
to memory, and brought to bear in future choice making. The
study of memory is fundamental to all of human behavior
and is of particular importance for consumer psychologists.
Researchers have investigated what and how information is
encoded, stored, and retrieved, the processes and likelihood
of retrieval failure, and how remembering and forgetting influ-
ence behavior. Consumer psychologists have used, and con-
tributed to, this literature in myriad ways. One important
attribute of memory is that it is active and constructive, and
often what consumers recall is a story that plausibly integrates
589
590 Consumer Psychology
bits of retrieved information versus an exact and invariant
recalling of experiences akin to watching and rewatching a DVD.
One of the most influential notions to arise from memory
research is that memory is associative. That is, memories are
created and stored in networks characterized by representa-
tions of concepts, or nodes, and links between nodes. The
strength of association between nodes is determined by
the frequency and recency of associations, as well as by the
depth of processing. When one node is activated, either
because it has been attended to in context or retrieved from
memory, related nodes may also be activated through a process
of spreading activation in which related nodes are activated to
the extent that they are strongly linked to previously activated
nodes. This basic property of memory has been used by con-
sumer psychologists in understanding diverse phenomena,
such as attitude formation, structure, change, and the likeli-
hood of attitudes driving behavior, predicting which brands
are given consideration prior to choice, how the priming of
concepts affects subsequent judgment and decision-making,
and how firms' investments in sponsorships may color brand
perceptions.
Consumer Information Processing
Consumer psychologists devote a great deal of attention to
understanding the processes that underlie the use of informa-
tion to make judgments and decisions (attention, comprehen-
sion, encoding, retrieval, integration), the extent of processing
(deep vs. shallow), and the direction of processing (one-sided
or belief-consistent vs. two-sided or evenhanded). The nature
of how consumers process information depends on their moti-
vation and ability to deliberate. When the motivation and the
ability to deliberate are high, the extent of processing increases
(consumers are likely to consider more alternatives and more
attributes), and the likelihood of two-sided processing, in
which consumers consider information that is inconsistent as
well as consistent with their beliefs, increases. When motiva-
tion or ability is low, fewer alternatives, fewer attributes, and
simple heuristic cues that enable consumers to reach judg-
ments and decisions quickly and easily are likely to be used.
Furthermore, consumers are likely to focus on information
that is readily accessible from memory or that is consistent
with their beliefs. Accessibility depends on the recency, fre-
quency, and intensity of information processing.
Consumers can attend to and think about only seven pieces
of information at a time. More than about seven pieces over-
loads consumers and forces them to use heuristics or shortcuts
to simplify information processing. Consequently, consumers
rarely consider more than seven brands when they need to
make a purchase decision, and frequently consider fewer
brands. Pioneering brands (i.e., new brands that are the first
entrant in a new category) and other salient or attention draw-
ing brands are likely to be considered. Being considered early
in a consumer's decision process lends tremendous competi-
tive advantage for a given brand, because brands that receive
selective consideration often benefit from a brand positivity
effect, in which a focal brand comes to be perceived as supe-
rior. As motivation or knowledge increases, the number
of brands considered increases. Generally, as the number of
brands considered increases, consumers make better decisions.
Information that is easy to process is weighed heavily in
judgment, even if other more relevant but more difficult to
process information is available. Information is easy to pro-
cess if it is easy to pronounce, easy to read, attention drawing,
encountered frequently, accessible from memory, or if it
has straightforward implications for judgment and decision-
making. Information that is easy to process seems more
familiar, more valid, more relevant, and more esthetically
pleasing, and such information leads to more confidently
held judgments.
Availability refers to whether information is stored in mem-
ory, and accessibility refers to the 'activation potential' of the
available information. Accessibility depends on the strength of
the association between stored knowledge and situational
cues, the recency with which information has been acquired
or last activated, the frequency of prior activation, and the
intensity with which information has been processed. Rather
than using all relevant information stored in memory, consu-
mers frequently use the subset that has been primed by the
environment. A wide variety of marketing communications
have been shown to serve as priming stimuli (e.g., ads, internet
messages, salesperson interactions, retail environments, con-
sumer magazines), and these priming stimuli have been shown
to influence a wide variety of judgments (e.g., expensiveness,
evaluative, and likelihood judgments) and mindsets (e.g.,
comparing brands in one product category can increase pur-
chase intentions in a different product category). Consumers
often overestimate the likelihood of occupations (e.g., doctors,
lawyers), objects (e.g., luxury cars, swimming pools), and
behaviors (e.g., having wine with dinner, incidence of crime)
shown frequently on television. This is particularly the case
for heavy viewers, despite the fact that most consumers do
not believe that television reflects reality. Priming can result
in either assimilation (a shift in judgment of a target toward the
priming stimulus) or contrast (a shift in judgment of a target
away from the priming stimulus), depending on the degree of
overlap between the priming stimulus and the target. Assimila-
tion occurs when overlap is high, and contrast occurs when
overlap is low.
Attitudes
The concept of attitudes, or evaluations of objects, is tremen-
dously important for consumer psychologists. Consumers can
hold attitudes toward a variety of very different targets, in-
cluding products, product categories, firms, advertisements,
spokespeople, and so on. To the extent that consumers hold
favorable attitudes toward a target, they are likely to engage in
approach behaviors such as attending to advertisements for
favored products and purchasing them, and remaining loyal
to liked brands over time. Negative attitudes, in contrast, drive
avoidance behaviors. Measuring attitudes allows consumer
psychologists to predict consumers' behavior, and behavior
can be influenced if attitudes are changed (e.g., when a con-
sumer is exposed to a persuasive message, or has a good or bad
experience after trying a product).
Consumers' attitudes may be formed based on three funda-
mental processes. First, attitudes can be formed as a result of
classical conditioning, in which originally neutral targets come
to be evaluated in accordance with other stimuli to which
Consumer Psychology 591
repeated associations are made. Attitudes can also arise from
processes of operant conditioning. The rewards and punish-
ments that drive consumer attitudes are diverse, and include
inputs such as direct experience with a product, and the social
consequences, or anticipated consequences, of making certain
consumer choices. To the extent that consumers feel rewarded
after engaging in a given behavior, the behavior is likely to recur,
whereas behaviors that evoke punishments, or fail to provide
rewards, are unlikely to be repeated. A third, more cognitively
intensive, attitude formation process involves breaking a target
object down into its attributes, considering how valuable each
attribute is, and the likelihood of each outcome occurring. Then,
the products of the values and likelihoods are summed to create
an overall expected value of the object, which is then compared
against other objects. Calculating attitudes in this fashion is
central to the Theory of Planned Behavior, a very influential
and enduring model of behavior that proposes that the expected
value of options, which considered along with social norms,
drives behavior.
A critical concern for consumer psychologists is the extent
to which attitudes and choices are strongly associated. To the
extent that attitudes are strong, they are likely to guide behav-
ioral decision-making, whereas weak attitudes are less likely to
play a role in behavior. Measures of attitude strength include
extremity (how positive or negative the evaluation is), accessi-
bility (how easily an evaluation of an object is recalled from
memory), and certainty (confidence that an attitude held is
correct). Attitudes characterized by ambivalence, that is, both
positive and negative evaluations, tend to be less likely to guide
behavior. Overall, though, as long as the correct attitude is
measured, attitudes are of enormous utility in predicting con-
sumers' behavior.
Consumer psychologists are devoting increasing attention
to understanding the implications of the distinction between
explicit and implicit attitudes. Explicit attitudes are delibera-
tive, and are based on consumers introspecting on, and subse-
quently self-reporting, their evaluations. Implicit attitudes, in
contrast, are much more automatic, and are based on associa-
tions in memory that do not necessarily reside in conscious
awareness. Explicit and implicit attitudes toward different tar-
gets vary in terms of how strongly related they are, and can be
manipulated independent of each other. Moreover, explicit
and implicit attitudes differ in terms of how they are influenced
and how they guide behavioral decision-making.
Persuasion and Attitude Change
The importance of attitudes in determining consumers' choices
has driven a long-standing interest in understanding how, and
under what circumstances, consumers' attitudes change. Of
particular focus is persuasion, the attempt to change another's
attitude. The earliest paradigm of attitude change research has
been termed the Message Learning approach, and has revealed
insights into how attributes of the source of a communication,
and the message itself, affect whether a persuasive attempt is
successful or not. More persuasion from a message occurs
when the source is credible (e.g., trustworthy, has high exper-
tise) and attractive (both socially and physically). Persuasion is
also facilitated when the message is constructed to match ele-
ments of the argument and the intended audience. Specific
considerations include when a one versus two-sided argument
should be used, where to position the strongest evidence in a
message, the optimal number of message exposures the audi-
ence should receive, whether conclusions should be specified,
and how to best use fear as a motivator.
A next paradigm includes dual-process models of persua-
sion, which have been enormously useful both in under-
standing persuasion and in stimulating research. Both the
Elaboration Likelihood Model and the Heuristic-Systematic
Model posit that attitude change may occur along one of the
two paths. When consumers have both the motivation and
the ability (i.e., there are no time constraints and no distrac-
ters) to carefully consider a persuasive communication, the
extent of persuasion is determined by their judgments of
the quality of the arguments in the message. In contrast,
when either motivation or ability is absent, persuasion is
driven by peripheral cues in the communication context (i.e.,
the number of arguments offered to support the communica-
tion, aspects of the source of the communication such as
expertise, attractiveness, and similarity). Although persuasion
may result from either path, when consumers evaluate the
merits of the message, more persuasion results, it is more
enduring over time, and it is less likely to decay than if persua-
sion is driven by peripheral cues.
In addition to research on the processes of persuasion,
researchers have identified six fundamental influences that
contribute to persuasion. First, to the extent that one likes
someone, one is likely to be persuaded by that person. Liking
often arises when consumers perceive that another is similar
to them, when the other offers compliments, and is attractive.
A second influence is reciprocity, or the social norm that favors
should be repaid. When someone feels indebted, even from an
unsolicited favor, they are likely to try to repay the favor. Third,
consumers like to be seen by others as well as by themselves as
being consistent. Once consumers are committed to an attitude
or a course of action, they tend to stick to their commitments.
A fourth influence is social proof, or the tendency to follow the
lead of similar others. Authority refers to consumers' proclivity
to defer to experts. Communications that appeal to expertise
often succeed because consumers infer that experts' opinions
are valid. A sixth influence is scarcity; to the extent that some-
thing is in demand, it seems more valuable. Marketers use
these persuasive influences in creative ways in the attempt to
convince consumers of the merit of their offerings.
Inference
A challenge that consumers often face is that they do not have
complete information about products and services that they
are considering purchasing. Consumers are likely to go beyond
the given information and draw inferences about missing
information when they are sensitive to the omission of impor-
tant information, and when they have accessible implicit the-
ories or assumptions that link evidence to conclusions in a
subjectively logical fashion. For example, most consumers
have an implicit theory that links price and quality; as price
increases, perceived quality increases. Typically, the relation-
ship between price and quality in consumers' implicit theories
is much stronger than the actual relationship between price
and quality in the marketplace. Consumers also have implicit
592 Consumer Psychology
theories about products made in different countries. In the
United States, many consumers believe that Japanese electron-
ics, German cars, and French wine are of high quality. Other
countries, such as China, Korea, and Mexico, are often asso-
ciated with low-quality products. Consumers also have im-
plicit theories about product popularity and quality. Highly
popular products are often perceived as higher in quality than
less popular products. Consequently, consumers like bestsel-
ling books, popular restaurants, and products that everyone
else seems to use or to like. Consumers have many different
implicit theories relevant to persuasion. For example, experts
are trustworthy, length implies strength (long messages are
more persuasive than short messages), and consensus cues
(millions of satisfied customers cannot be wrong) can have a
profound impact on persuasion. Unfortunately for marketers,
consumers also have an implicit theory that marketers are
untrustworthy and that one should not believe everything
one sees or hears in marketing communications, which pro-
duces resistance to persuasion.
In addition to drawing inferences about global quality,
persuasiveness, and trustworthiness, consumers also attempt
to draw inferences about specific missing attributes when they
realize that information is missing. However, consumers are
frequently insensitive to omissions and this can lead to strong
judgments (extreme and confidently held judgments) based on
weak evidence. High levels of prior knowledge or involvement
can increase sensitivity to omissions and lead to more moder-
ate judgments. Comparing products described in terms of
different attribute dimensions can also increase sensitivity to
omissions. Reading a relatively large amount of information
about one product can increase sensitivity to omissions
concerning another product described by a relatively small
amount of information, even when the two products belong
to completely different categories.
Categorical knowledge also helps consumers to fill in miss-
ing details. Once a product has been categorized, consumers
assume that the product has all of the properties and character-
istics that are typically associated with the category. Even if a
consumer has never tasted panna cotta, he or she is likely to
assume that panna cotta is sweet when they learn that it is a
dessert. Prior knowledge and experience is organized into
knowledge structures called schemata that are also useful for
fdling in gaps in knowledge. For example, the linear ordering
schema suggests that if A is better than B and B is better than C,
then A must be better than C. Similarly, the syllogism schema
suggests that if A implies B and B implies C, then A implies C.
Behavioral Decision Theory
Behavioral Decision Theory refers to the research paradigm
that explores consumers' decision tendencies. For example,
research on the framing effect shows that the manner in
which choice alternatives are framed or described has a pro-
found effect on choice, even though consumers typically
believe that their preferences are stable. This occurs because
consumers think differently about outcomes described in
terms of gains (above a reference point) versus losses (below
a reference point). Loss aversion, or the tendency to weigh
losses more heavily than equivalent gains, can explain the
framing effect, the compromise effect, the endowment effect,
the status quo bias, and the sunk cost fallacy. For example,
when outcomes are framed in terms of gains, consumers are
risk-averse. When outcomes are framed in terms of losses,
consumers are risk seeking. Consequently, when asked to
choose between a program that would save 200 out of 600
lives for sure and a program that would save 600 with p = 0.33
or 0 with p = 0.67, most people prefer the former safe option.
When asked to choose between a program in which 400 out of
600 would die for sure and a program in which 0 would die
with p= 0.33 or 600 would die with p = 0.67, most people
prefer the latter risky option. Of course, the two safe options
are identical and the two risky options are identical too.
Consumers often prefer a compromise brand that is average
on two equally important attributes over brands that are excel-
lent on one attribute but poor on the other. Poor performance
on an attribute is often interpreted as a big loss and loss
aversion leads consumers to avoid extreme brands.
Research on the endowment effect shows that owning an
object leads consumers to value that object more. This occurs
because buying a new object is interpreted as a gain and selling
an object one currently owns is interpreted as a loss. Loss
aversion makes consumers reluctant to sell objects they cur-
rently own. Consequently, selling prices are often much higher
than buying prices.
People often prefer the status quo to change (the status quo
bias). This occurs because change often involves gains on some
dimensions and losses on others, and loss aversion can make
people reluctant to accept change. Consequently, when con-
sumers are told that a pharmaceutical product is currently on
the market, they recommend keeping it on the market. When
they are told that the same product is currently not on the
market, they recommend keeping it off the market. People
also prefer to keep their current jobs, automobile color, finan-
cial investments, and medical insurance policies.
As the amount of time, money, or effort invested in a
project increases, the reluctance to abandon the project
increases even if people would be better off doing so (the
sunk cost fallacy). After spending millions of dollars on a
dam, building, or construction project, people are reluctant
to abandon the project even when it would be less expensive
to do so. After spending millions of dollars on old mixing
technologies, many steel companies went out of business
because they refused to invest in new superior technologies.
Loss aversion makes people reluctant to give up what they have
already invested, even when the investment is unlikely to
ever pay off.
Motivation
Motivation refers to tension that results from the discrepancy
between a current state and needs or wants that consumers are
driven to ameliorate. Consumers feel motivation to address both
biological (e.g., thirst, hunger, safety) and learned (e.g., prestige)
needs, and needs may be utilitarian and practical, or hedonic
and emotional. Many classic psychological concepts and the-
ories have been adopted by marketers, including Drive Theory
and the motive to maintain homeostasis, motivational conflicts
(i.e., approach-approach, approach-avoidance, avoidance-
avoidance, and double approach-avoidance), and Maslow's
Hierarchy of Needs.
Consumer Psychology 593
One of the most influential motivational theories in con-
sumer psychology is regulatory focus. Research has explored
how individuals' motivation to seek positive outcomes while
avoiding negative outcomes affects a host of behaviors. Two
self-regulation systems help individuals pursue these motives.
The promotion system seeks advancement and growth, and
approach strategies are employed in the pursuit of positive
outcomes. The prevention system, in contrast, derives from
security needs such as safety and protection, and employs
avoidance strategies to preclude negative outcomes. Although
these systems are independent, one's overall regulatory focus at
a given time may vary, and is determined both by individual
differences in socialization and by contextual factors such as
the framing of behavioral decision problems. Regulatory focus
has been shown to play an important role in numerous con-
sumer behavior phenomena, including the evaluation of pro-
ducts, response to advertising, perceptions of the importance
of different product attributes, and the propensity to adopt
new products.
Although regulatory focus considers general motivation,
consumers often behave to address more concrete goals.
Goals are representations of desired and undesired states, and
proximally activate behavior. In some cases, goals may conflict,
in which case the goal associated with more importance (i.e.,
stronger activation as a function of internal associations and
external context) is likely to dominate. Goals may be actively
set by consumers and are hierarchical in nature, that is, they
vary according to how specific they are. Goal pursuit refers to
the actions consumers take to achieve their goals. The goal
concept is important for consumer psychologists conceptually
because understanding goal-directed activity helps to unravel
the complexities of consumer behavior. Marketing practi-
tioners make use of goals in myriad ways; if consumers' goals
are known, firms can better address consumers' needs in
marketing communications, product development, pricing,
and so on.
Individual Differences in Consumer Psychology
A variety of traits that vary across individuals relate to what
information they consider, and how they process it, as well as
what values and preferences are brought to bear in the con-
sumer choice context. Consumer psychologists care about indi-
vidual differences for two reasons. When one observes a given
consumer behavior, that behavior is influenced both by the
context in which the behavior occurs and by variables internal
to the consumer. Thus, studying individual differences helps us
to better understand the nature of the consumer as a decision-
maker. Moreover, studying how consumers who vary on some
theoretically relevant dimension respond differentially in a
given consumption context can help us to understand the
processes that underlie the behavior.
Researchers have considered a wide range of individual
differences. Although exhaustive cataloging is beyond the
scope of this article, several individual differences that have
proved to be of extraordinary utility are discussed. The Need
for Cognition, for example, refers to the extent to which con-
sumers like to engage in careful, effortful thinking, and to
understand and make sense of the world. Consumers who
are high versus low in Need for Cognition engage in more
exhaustive information search and more effortful processing
of persuasive communications, are more open-minded, have
more persistent attitudes, consider a wider range of prices
when making a purchase, and prefer verbal versus visual
information.
Conceptually related yet distinct from Need for Cognition
is the Need for Closure, which refers to the desire to come to a
decision, any decision, quickly versus engage in exhaustive
deliberation before making a choice. Other individual differ-
ences relevant for consumer psychologists include the Need for
Uniqueness, which refers to the pursuit of differentiating one-
self from others through the acquisition and use of consumer
goods, Romanticism (inspirational, imaginative, intuitive, and
creative people for whom feelings come first) versus Classicism
(unemotional, straightforward, economical, facts come first
types), and, as noted earlier, regulatory focus considered as
a trait.
Consumer Neuropsychology
A next generation of consumer decision research involves
understanding how brain function guides consumer behavior,
often by localizing parts of the brain responsible for different
types of information processing. Researchers have used meth-
odologies such as functional magnetic resonance imaging
(fMRI) scans, which enable researchers to determine which
brain areas are most active when consumers engage in particu-
lar tasks, and lesion studies, in which the behavioral decision-
making of patients with injuries to specific neural structures is
compared to individuals with injuries in other regions, as well
as to those without any injury. This is an emerging science, but
it has already produced a number of provocative findings
regarding the role of particular brain structures, as well as
the functions (e.g., emotion) governed by those structures, in
decision-making.
Consumer Welfare
Consumer psychologists study a number of behaviors with the
goal of enhancing consumer welfare. For example, researchers
have explored the harm done by the advertising of unhealthy
products such as sugar-laden food and smoking materials.
Other research discusses how marketing can propagate racial
stereotypes or unrealistic standards of female attractiveness.
Social marketing research investigates how to persuade consu-
mers to adopt healthy or prosocial behaviors. Finally, research-
ers have considered the effects of marketing on disadvantaged
and at-risk populations. Overall, although the primary focus in
consumer psychology is on understanding consumer behavior
to facilitate organizational performance, a great deal of atten-
tion is also given to facilitating consumers' well-being.
Managerial Applications of Consumer Psychology
Research in consumer psychology is often applied in industrial
settings, most typically in marketing. By understanding how
consumers form judgments and make decisions regarding pro-
ducts, services, and marketing communications, marketers
hope to be better able to offer products that match consumers'
594 Consumer Psychology
preferences, price and promote products optimally, and offer
products to consumers via distribution channels (e.g., retail,
online) that best facilitate sales. Consumer psychology has
generated an enormous body of knowledge that can enhance
organizational performance. Some examples of how consumer
psychology has helped marketers to be more effective follow.
'Brand management,' or 'branding,' refers to a set of key
decisions that marketers make in the aim of creating a product
that meets consumers' needs, including both tangible needs
(e.g., food relieves hunger) and intangible needs (e.g., the
prestige of owning a luxury car), and that is associated with
favorable attitudes and is likely to be purchased. Understand-
ing consumers' motives as well as the social influences on them
is essential. Some examples of marketers' considerations when
trying to build brands include creating marketing communica-
tions and product packaging that conveys an attractive, clear,
consistent, meaning that facilitates awareness of the product,
and signals what the product does, and why it is different, and
better, than competitors. The 'differentiators' can be on a num-
ber of dimensions, including quality, reliability, price, and/or
image. Consumers choose products that best meet their needs
on these dimensions. The relationships consumers have with
brands are dynamic, and in many ways approximate human-
to-human relationships.
Consumer psychologists know a great deal about why con-
sumers choose particular products, and this knowledge is very
useful in making a variety of brand management decisions.
One such set of decisions regards creating and managing
brand elements, such as brand name, logos, symbols, slogans,
and packaging. Marketers' decisions are made in light of the
constructs discussed earlier; brand elements should be atten-
tion grabbing, memorable, and contribute to brand salience so
that the brand receives consideration. Elements should be
likable and correspondingly facilitate likeability of the brand,
so that strong, favorable, attitudes that guide choice are devel-
oped and maintained. Moreover, brand elements should be
meaningful, so that brand associations in memory are forged,
facilitating consumers' esteem for, and choice likelihood of,
the product.
Marketers also have to set strategies regarding marketing
communications that facilitate high purchase likelihood of
their brands. Ideally, tools, such as advertising, sales promo-
tion, public relations, publicity, and new media (i.e., social
networking Web sites), should be integrated so that consumers
can easily attend to and process information about the brand,
perceive the product in a way that the marketer wants, and
hold consistent, strong attitudes toward the brand. An example
of a common persuasion tool studied extensively by consumer
psychologists is comparative advertising, in which a compari-
son is made between the advertised brand and one or more
competitors. Researchers have shown that such ads may be
very effective in persuading consumers because they are very
good at garnering attention, but that in some circumstances
they may be ineffective or even harmful to the firm that uses
them. In general, when making a comparison in an advertise-
ment, it is more persuasive to make a positive comparison
(e.g., their brand is good, our brand is better) than a negative
comparison (e.g., their brand is poor, our brand is good)
because consumers often infer that a negative comparison is
biased and unfair. The exception to this rule is that firms that
are well known and trusted can make negative comparisons
without provoking an inference of bias from consumers.
Inference research has been invaluable in helping marketers
weigh the potential benefits and risks of sales promotion, a
commonly used set of tools that offers some extra value to
consumers if they make a purchase. Common sales promo-
tions include coupons, rebates, and free gifts if consumers
purchase a given product. Although it is unsurprising that
offering such sales inducements often lead to increased sales,
consumer psychologists have helped marketers understand
how promoting a product can erode consumers' evaluations
of it. Specifically, because consumers perceive a strong relation-
ship between price and quality, when a product is offered with
a price discount, consumers may infer that it is a bad product.
Research has shown when such an inferential process is a
threat and how to ameliorate the risk (e.g., one strategy is to
advertise when one offers a price promotion to bolster brand
attitudes).
A ubiquitous problem in managing brands is how and
when to offer brand extensions, that is, launch a new product
using an existing brand name (e.g., the parent brand). Busi-
nesses often favor this strategy because consumers are often
more likely to try a new product if it is associated with a brand
name that they know and trust. Moreover, consumers perceive
the new product in light of the associations that they have with
the existing brand in memory, and thus businesses often can
spend less on marketing support for a brand extension versus a
new product with a new brand name. Consumer psychologists
have devoted extensive study to determining influences on two
key criteria for the success of a brand extension: whether con-
sumers will purchase the new product in sufficient frequency
for the manufacturing firm to turn a profit, and, importantly,
what implications the launch of the new product will have for
the existing brand. In general, brand extensions are likely to be
successful when there is a strong match between the associa-
tions that consumers have in memory regarding the parent
brand and the associations that will lead to favorable attitudes
for products in the category of the proposed extension, and
when associations about the parent brand are strong enough to
carry over to the brand extension. Launching a brand extension
can have harmful consequences for the parent brand when
new associations with the parent brand are developed that
undermine consumers' attitudes toward the parent brand.
Such a process is likely to occur, for example, when a low-
quality brand extension shares the name of a high-quality
parent brand, even if objective features of the parent brand
do not change.
Although brand extensions are a useful strategy for driving
profits, new products are firms' lifeblood and the future.
Indeed, one of the most vexing problems faced by business
mangers today is the successful creation and introduction of
new products. Despite the vast resources expended by practi-
tioners and an impressive amount of research on new product
introductions, the failure rate of new products is still extremely
high; it is estimated that between 60% and 90% of all new
products fail. Recently, the focus has shifted from refining the
process of new product development from the companies'
perspective toward understanding the role of consumer psy-
chology with respect to the precursors and motivations to
adopt new products. At a global level, consumer researchers
Consumer Psychology 595
investigate how core psychological theories (i.e., approach/
avoidance, regulatory focus, goals) impact how consumers
learn about, evaluate, and ultimately, adopt new products.
Some of the topics that are currently being investigated include
understanding how consumers evaluate the importance of new
benefits (and perceptions of learning costs that are associated
with obtaining these new benefits), how consumers integrate
multiple sources of information to both learn about and eval-
uate new products, and how the adoption and use of new
products act as core elements of consumers' identities.
Many psychological factors come into play when consu-
mers predict how useful new products will be. For example,
there are common biases in the estimation of the benefits
provided by new products, whereby consumers overweigh the
importance of existing benefits that they know well while
underappreciating the value of novel benefits with which
they are unfamiliar. Consumers' perceptions of the amount
of learning required to extract novel benefits have a huge
impact on the adoption of new products. Another important
question is associated with understanding the role of multiple
sources of information (i.e., company sponsored versus con-
sumer recommendations) on adoption intentions for new
products. For example, the emergence of the social media has
increased the ability of potential consumers to read reviews
and engage in conversations with fellow consumers who are
early adopters of new products. How these interactions will
shape the adoption pattern for new products has yet to be
determined. Finally, the role of new product adoption in pre-
senting oneself to others is an important factor in adoption
decisions that is not particularly well understood. Although
understanding the connections between consumers and
brands has been extremely helpful in the branding domain,
defining the connections between new products and multiple
levels of the self (i.e., actual, ideal) could be a fruitful avenue
for incorporating the core principles of consumer psychology
toward furthering one's understanding of the adoption of new
products.
See also: Associative Learning; Attention; Decision Making
(Individuals); Media Influence on Behavior; Memory; Motivation;
Neurotechnologies; Organizational Behavior; Persuasion; Social
Cognition.
Further Reading
Albarracin D, Johnson BT, and Zanna MP (2005) Handbook of Attitudes. Mahwah, NJ:
Erlbaum.
Ariely D (2008) Predictably irrational: The Hidden Forces that Shape Our Decisions.
New York: Harper Collins.
Bagozzi R, Gurhan-Canli Z, and Priester J (2002) The Social Psychology of Consumer
Behavior. London, UK: Open University Press.
Belch G and Belch M (2008) Advertising and Promotion: An Integrated Marketing
Communications Perspective , 8th edn. New York: McGraw-Hill/Irwin.
Chaiken S and T rope Y (1 999) Dual Process Theories in Social Psychology. New York:
Guilford Press.
Cialdini RB (2008) Influence: Science and Practice, 5th edn. New York: Allyn & Bacon.
Hastie R and Dawes RM (2001 ) Rational Choice in an Uncertain World: The Psychology
of Judgment and Decision-Making. Thousand Oaks, CA: Sage.
Haugtvedt CP, Herr PM, and Kardes FR (2008) Handbook of Consumer Psychology.
New York: Taylor & Francis.
Kahneman D and Tversky A (2000) Choices, Values, and Frames. Cambridge, UK:
Cambridge University Press.
Kardes FR, Cronley ML, and Cline TW (2011) Consumer Behavior. Cincinnati, OH:
South-Western College.
Kardes FR, Posavac SS, and Cronley ML (2004) Consumer inference: A review of
processes, bases, and judgment contexts. Journal of Consumer Psychology
14: 230-256.
Keller KL (2008) Strategic Brand Management: Building, Measuring, and Managing
Brand Equity, 3rd edn. Upper Saddle River, NJ: Pearson Education.
Payne JW, Bettman JR, and Johnson EJ (1993) The Adaptive Decision-Maker.
Cambridge, UK: Cambridge University Press.
Weller JA, Levin I, Shiv B, and Bechara A (2007) Neural basis of adaptive
decision-making in risky gains and losses. Psychological Science 18: 958-964.
Relevant Websites
http://www.marketingpower.com/ - American Marketing Association.
http://www.acrwebsite.org/ - Association for Consumer Research.
http://www.myscp.org/ - Society for Consumer Psychology.
Coping
S Blum, M Brow, and R C Silver, University of California, Irvine, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Coping Behavioral, cognitive, and emotional attempts to
mitigate or manage stressful or threatening circumstances;
the continual process of interpreting and responding to life's
demands.
Emotion-focused coping Coping efforts aimed at
managing negative feelings related to a stressor with the goal
of reducing the impact of those feelings.
Primary appraisal Initial recognition of a potential threat or
source of stress and then consideration of its potential
negative impact.
Problem-focused coping Coping efforts aimed directly at
changing or managing the stressor itself.
Secondary appraisal Evaluation of coping resources and
options available to remove the stressor or mitigate its
impact.
Stress/stressor Any change (internal or external) to life
patterns and habits typically construed as a threat to life or
comfort; also, the physical and psychological manifestations
resulting from exposure to these threats.
Introduction
All life seeks a sense of constancy or stability, yet the greatest
constant in life is change. As changes occur in the human body,
the autonomic nervous system works to restore everything to
a state of balance, called homeostasis. This process requires
no conscious thought or action by the individual to respond to
that change. The external world, that is, daily life, also is filled
with new situations and changing circumstances that challenge
stability, from everyday irritants such as a bad night's sleep or
heavy traffic to acute conditions like flu or a disagreement with
a loved one, and ongoing strains such as chronic disability or
the death of a spouse. Because these changes are a part of the
normal course of life, people must fairly consistently monitor
and adjust - themselves, the change factor (stressor), or the
environment - in order to regain that sense of stability. Thus,
the healthy human body may take care of itself internally.
What about changes that require adaptation? Do they 'stress
us out?'
Endocrinologist Hans Selye introduced the terms stress and
stressor while studying the biological responses of rats to exter-
nal stimuli. He defined stress as any change to an organism that
requires or elicits adjustment from that organism in order to
return to homeostasis. Selye did not characterize the factor
(stressor) triggering the need to adjust as positive or negative,
only that it exists and can result from both pleasant and
unpleasant events. So, while studies have shown strong asso-
ciations between negative life events and symptoms of psycho-
logical distress, psychiatric disorders, and even mortality, the
stressor itself is not necessarily the culprit. Stress is not inher-
ently bad. Rather, it can provide the impetus for people to learn
and grow (see eustress [eu = good] also coined by Selye in
the mid-1970s). Because it opposes simple stagnation in life,
a certain degree of stress is healthy for humans. Selye's
biological framework was seminal to the field of human cop-
ing because he was the first scholar to conclude that a key
determinant in the stress-health relationship is neither the
type (good/bad) nor level (a little/a lot) of the stressor itself,
but how one reacts to, that is, copes with, that stressor.
Coping is the ongoing behavioral, cognitive, and emotional
processes people use to manage those life circumstances that
threaten feelings of stability. Coping can also occur proactively
in order to limit the threat to future homeostasis. Its goal is to
avoid, minimize, master, or manage life's situations in such a
way as to decrease any potential negative impact or outcome. In
the social sciences, the study of stress and coping encompasses
many factors that relate to life changes and how people manage
those changes, so scholarly discourse on these topics is both
immense and diverse. The goal of this article is to provide
a general overview of coping in psychosocial research. First,
two historically predominant theoretical frames - coping as a
personality disposition and coping as a person-situation pro-
cess - are reviewed and the current direction of these perspec-
tives is summarized. Next, major coping strategies and coping
resources are listed, and finally, a few additional factors that
influence the coping process are offered. The material herein
relates to adult coping and does not include issues specific to
children and chronic psychopathology. It is our hope that this
brief discussion will provide readers with general information
about the coping literature (i.e., research), as well as help them
gain an understanding of their own coping processes.
Theories of Coping
Numerous terms and expressions have been used to convey
scholars' beliefs about the origin, or basis, of coping. Nearly
all of these conceptualizations can be viewed within a theoreti-
cal framework of coping as either part of one's personality, an
active process, or most recently, the combination of these two.
The following are some main points from each of these theoret-
ical perspectives.
The Personality Framework
The foundation for coping as a part of personality began with
the work of Sigmund Freud. According to early psychody-
namic/psychoanalytic theory, conflicts between the id and
superego create anxiety, and the ego responds by employing
596
Coping 597
defense mechanisms to repress these negative feelings. Though
additions have been made to this list over time, the original
Freudian defense mechanisms are denial (refuse to believe
despite evidence to the contrary), displacement (act out on
someone else rather than the actual target), intellectualization
(pay little attention to emotions), projection (place own nega-
tive thoughts or feelings on another), rationalization (replace
the real reason with a logical one), reaction formation (believe/
feel the opposite of the real, unwanted belief/feeling), regres-
sion (return to a child-like state/behavior), repression (keep
the unwanted in the unconscious), sublimation (act out unac-
ceptable impulses in a socially acceptable way), and suppres-
sion (try to put the unwanted into the unconscious).
Though it remained popular for many years, Freud's view of
coping was sharply criticized for defining all defense mechan-
isms as existing completely within the unconscious and as
pathological in nature. Personality was considered to be inflex-
ible and driven solely by sexual or aggressive urges of the id.
Other scholars defined defense mechanisms in a less negative
fashion and held that personality formed the basis for coping
in both the conscious and the unconscious. Instead of being
pathological in nature, coping and the desire for control are
seen as normal, and indeed, healthy, functioning. As part of the
individual's conscious world, this frame for coping relates
personality to the likelihood of exposure to certain types or
numbers of stressful events (e.g., choosing to become a fire-
fighter or emergency room physician), perception and inter-
pretation of the stressor (e.g., an unwanted threat or exciting
challenge), responses to the stressor (e.g., learn more about the
situation, pray, make a plan), and the availability of resources
(e.g., a strong group of friends, sufficient income, health). This
definition of coping as being more than an unconscious course
of action undeniably advanced our knowledge of coping and
personality. Early research on personality, however, was hin-
dered by a lack of consensus on descriptors and other research
procedures, which made it difficult for scholars to share results,
duplicate one another's work, and establish verifiable results.
To address this issue, researchers systematically created a
list of five specific personality traits. Called the Big Five and
sometimes known by the acronym OCEAN, those traits are
openness to experience, conscientiousness, extraversion, agree-
ableness, and neuroticism (alternately called emotional stabil-
ity for its opposing trait on the continuum). Popular among
researchers, the Big Five framework provides the solid, com-
mon foundation necessary to properly explore relationships
between personality and coping. Studies indicate that people
are relatively consistent in their choice of coping responses
across different situations, and traits are related to the types
of coping strategies an individual is likely to use, as well as the
order in which the person will try given strategies. The most
frequently studied trait has been neuroticism. People high in
neuroticism tend to interpret common situations as threaten-
ing, emphasize the negative features of a situation (i.e., nega-
tive processing bias), and are more likely to recall stimuli
associated with failure. Neuroticism also is related to the less
effective, passive, or possibly harmful ways of coping: denial,
wishful thinking, self-blame, avoidance, hostility, distanc-
ing, and distraction. In contrast, extraversion, conscientious-
ness, agreeableness, and openness to experience are related to
more positive appraisals of potentially threatening situations,
interpreting negative events as a challenge, and emphasizing
the positive aspects of a situation (i.e., positive processing
bias). Extraversion predicts low use of avoidance, denial, or
disengagement from goals. Conscientiousness is related to
planning and seeking help. Conscientiousness and agreeable-
ness are related to lower substance abuse, and openness to
experience is related to planful problem-solving.
Although the Big Five personality traits have received a great
deal of scholarly attention, much of the research has examined
the associations between a single trait, a particular outcome
(e.g., substance abuse, depression), and coping responses.
However, individuals possess a combination of these five traits,
which may interact to influence the types of experiences they
select for themselves and how they respond to the potential
stress associated with these experiences. For example, a person
who is high in extraversion and neuroticism but low in consci-
entiousness may be impulsive, pleasure seeking, and have low
control over his/her emotions, which may lead to quitting his/
her job impulsively. This decision may lead to great stress if
another job cannot be found. In addition, these same traits
that led to the stress of job loss may also influence subsequent
coping strategies selected. In this case, this combination of
traits may lead the individual to seek out an entirely new
field of employment.
Coping as a Process
The post-World War II social and academic Zeitgeist of the
1950s and 1960s saw a shift in focus away from behaviorism,
views that reduce human motivation to expectations about
rewards and punishments, and toward Richard Lazarus' trans-
actional theory of coping with stress. Lazarus's process model
of coping emphasizes an active interplay between the individ-
ual and the environment, and accentuates the role of the
person's cognitive appraisal of a given stressor and how best
to act upon that stressor.
Lazarus' transactional theory is a cyclical process of appraisal
and feedback, and the way a person copes with a stressor is
largely determined by the meaning that the person attributes to
the event in his/her life. Primary appraisal occurs when the
individual senses a change in the environment (stress), then
determines whether this change is good, has no impact, or
poses some sort of threat. If the change is perceived as good
or neutral, the process ends. If the change is perceived to be a
threat, then secondary appraisal occurs. During secondary
appraisal, the individual takes an inventory of available
resources (types and amounts), assesses possible courses of
action and constraints to these actions, and determines what
may provide the greatest likelihood of managing or mitigating
the situation. These appraisals result in the use of certain
strategies to address (i.e., cope with) the given stressor. Feed-
back from the evolving situation then triggers a reappraisal of
the stressor and available resources. The individual adjusts
behaviors, cognitions, and emotions as necessary, and if the
situation resolves, the cycle ends. If the situation continues, the
cycle repeats. Note that the coping process may involve the
individual focusing on (and perhaps adapting) himself or
herself, the external environment, or a combination of the two.
Process theory classifies the various behavioral, cognitive, and
emotional strategies people use. Lazarus and Susan Folkman
598 Coping
categorized coping strategies as either problem-focused or
emotion-focused. Problem-focused coping strategies seek to
reduce stress by dealing directly with the stressor to decrease its
impact. These strategies tend to be action-oriented with the goal
of eliminating or changing the stressor. For example, a woman
who lacks reliable transportation to her job can directly resolve
her situation by replacing her old car. Problem-focused coping
also may involve changing oneself in relation to the stressor,
such as a hiker moving to the other side of a trail to avoid a
snake while on a nature walk. In comparison, the goal of
emotion-focused coping is to reduce the impact of distressing
feelings. This method is more effective than problem-focused
coping for circumstances over which the individual has no
control, such as a severely disabling condition or death of a
loved one. It also often occurs when problem-focused coping
fails to resolve the situation or when the stressor is so great that
problem-focused coping has no real likelihood of helping. For
example, a soldier who loses two buddies in an attack cannot
control this reality, but he may tell funny stories about his
friends to decrease feelings of grief. Emotion-focused coping
need not be linked to the specific event. The grieving soldier
may increase his workouts to counteract the negative effects of
his war experiences, of which death is only one part.
A third category for coping strategies is known as meaning-
focused coping, wherein people attempt to understand their
stressful experiences in a broader context. Also called cognitive
reframing, meaning-focused coping helps people align nega-
tive experiences with their belief systems (spiritual or other-
wise) or worldviews, attempting to make sense of the stressor
by finding a purpose or something good from it. An excellent
example is the young woman dying of breast cancer who made
her sister promise to educate others about the disease. That
promise became Susan G. Komen for the Cure, the world's
largest charity for breast cancer that has raised $1.5 billion
since its inception in 1982. Meaning-focused coping is linked
primarily to traumatic events; the ability to find meaning in
negative events is associated with more positive mental and
physical health outcomes. Finding meaning may be a catalyst
not only for decreasing distress levels but can also result in
cognitive and emotional growth (see, e.g., Richard Tedeschi
and Lawrence Calhoun's concept of posttraumatic growth). At
the same time, not every person who searches for meaning in a
traumatic event is able to make sense of that event in such a
way as to make it meaningful. Some events may be perceived as
so contrary to one's normal experience that they challenge
strong beliefs about the way the world is supposed to be.
Events that are uncontrollable, sudden, and violent, such as a
random homicide, can have these qualities, as can rape or the
sudden death of an infant. These types of events are often
perceived as gravely unfair, and the attempt to find meaning
without being able to do so has been found to be a stressor in
itself.
Current Directions in Coping Theory
In the last decade, powerful statistical software programs have
made possible the rapid analyses of very complex sets of data.
Similarly, the full-text availability of the majority of coping
research (both historic and current) via the World Wide Web
has enabled scholars to integrate this information and test
theories in a manner that previously was quite difficult. Cur-
rent research shows that the situation (e.g., being a single
parent, a medical diagnosis), individual differences (e.g., extra-
vert or introvert, life experiences), and environmental factors
(e.g., physical safety, cultural standards) are all associated with
a person's coping response. Thus, the current refinement of
coping theory focuses on the integration of these factors, as
well as determining whether certain combinations predict
certain coping responses or outcomes.
Coping Strategies
Coping strategies are as numerous and varied as the stressors
that precede them. From Folkman and Lazarus' Ways of Coping
Questionnaire and Charles Carver and colleagues' Coping Orien-
tation of Problem Experience (COPE), some common strategies
or categories for coping responses are accepting the situation or
one's role in it, active/confrontive coping to remove the stressor
or oneself from the stressor, anticipatory coping aimed toward
an expected but uncontrollable event, avoiding/escaping the
stressor or associated feelings of distress, denying the problem
or feelings, disengaging mentally or behaviorally (giving up),
distancing/detaching from the situation or minimizing its sig-
nificance, planning the steps to solve the problem, reinterpreting
the stressor as a positive or growth-oriented experience, seeking
social support (discussed later), controlling one's emotions or
waiting for an appropriate time to act, using substances to dull
feelings, suppressing competing activities until the problem sub-
sides, turning to religion, using humor, and venting emotions.
Individuals also engage in proactive coping. These future-
and action-oriented behaviors can prepare a person not only
for specific stressors, but also for those that are likely to arise in
the normal course of life. Proactive coping includes building
and strengthening all resources (e.g., from practical and aca-
demic knowledge, experiences, and sufficient numbers and
varied kinds of social contacts; see also direct effects hypothesis
of social support). As well, proactive coping involves gaining
skills and abilities to assess the changing environment more
accurately, from signs of a possible stressor, to appropriate
strategies and resource use, to feedback on a given situation.
Though it is impossible to make causal claims, future-oriented
coping tends to be associated with positive outcomes such as
goal achievement and lower levels of distress. However, too
great an emphasis on the future may be a sign of hypervigi-
lance, which has been linked to negative outcomes such as
anxiety and poor information processing.
As with the other components of coping, research classifies
strategies by type, such as problem-focused coping or emotion-
focused coping. One needs to bear in mind, however, that
people's thoughts, feelings, and behaviors seldom fit into
black-or-white categories such as those required for quantita-
tive analytics. For example, a man responds to being laid off by
networking with friends to learn of other job opportunities.
Networking is active coping and planning, both of which are
problem-focused. Networking with friends, however, likely adds
the benefit of support from others to decrease his anxiety, which
is emotion-focused. This illustration also shows that often the
best method of coping is a combination of problem-focused
and emotion-focused strategies. Whether it is a single strategy or
Coping 599
a few enacted together, this approach can both reduce the
impact of the stressor and the negative emotions it may evoke.
Yet, there is no one universal coping strategy that will be effec-
tive in all situations. Coping strategies that lessen distress in one
situation may be ineffective or even detrimental to the individ-
ual in another.
Similarly, one must keep in mind that strict labeling of most
strategies and other coping behaviors can result in erroneous
interpretations. This is especially true when attempting to cate-
gorize responses as effective or ineffective. Seldom is a coping
response all good or all bad, and one must well consider the
many specifics of the situation and the goals of the individual
prior to drawing firm conclusions. For example, considering the
time frame during which the strategy is employed (e.g., the
duration of the stressor) may be helpful in assessing its efficacy.
Denial or substance use (not to be mistaken with abuse or use
of illicit drugs) maybe adaptive in the immediate aftermath of a
traumatic event. However, these strategies rapidly deteriorate
as wise choices for coping over time and should be replaced
with responses that are more appropriate as time progresses. So,
when the laid-off man goes home, he may have a few beers
to wash down his sorrows, but continuing this behavior would
no doubt exacerbate the problem.
Coping Resources
During secondary appraisal, people identify the types and
amounts of resources they can call upon to help manage the
stressor or buffer the intensity of their feelings. The presence
and absence of resources directly influences the appraisal pro-
cess because coping strategies may require specific resources
in order to be an available option or for particular behaviors
to be effective. This is true for both internal resources (e.g., self-
esteem, knowledge, health) and external resources (e.g., money,
friendships, shelter). Coping resources may be tangible or intan-
gible, and the perceived value (intrinsic/extrinsic) of a given
resource can be highly variable depending on the individual,
the specific situation, and even the timing of the situation. For
example, rapid access to a hospital maybe essential immediately
after an injury. During recovery, though, the need for emergency
care declines while the value of a loved one's presence may
increase. The more and varied types of coping resources a person
has, the greater the likelihood of coping success. As well, those
who lack resources or who are unskilled at identifying or using
them are at greater risk for psychological distress and mental
illness. Stevan Hobfoll suggests that it is the threat to one's
resources, whether material goods or personal relationships,
that is the major source of stress in the environment.
Like coping strategies, coping resources have been categor-
ized in the literature. Some are appropriate for problem-focused
coping, others are emotional-focused or meaning-focused, and
like coping strategies, some can address more than one area.
Scholars often use the terms resource and support interchange-
ably, for example, material support.
Material Resources
Many stressors can effectively be addressed or eliminated, often
proactively, with material resources. Material resources typically
are tangible, such as a house, vehicle, clothing, a computer, and
other items used in daily life. In modem society, however, the
key material resource is money. Sufficient and consistent access
to money and its equivalents helps individuals cope in a manner
that no other material resource or coping strategy can match. An
adequate supply of funds can provide proper shelter, healthy
and sufficient food, medical care, and countless other goods
and services that both proactively and concurrently address the
stresses and strains that result when a person lacks necessities.
Socioecological interpretations of deviation amplification illus-
trate the impact of money across multiple aspects of life. Studies
clearly show that those with less money experience a dispropor-
tionately high level of exposure to stressors. Some of these
consequences are generally apparent, such as the lack of ade-
quate food, shelter, or medical care (preventive or otherwise).
Other circumstances are less obvious, however, such as the
positive relationship between poverty and living or working
around environmental toxins.
Social Support
Research conducted over the last few decades has clearly
demonstrated the importance of family and close others in
times of stress. Meta-analyses indicate that the influence of
social relationships on morbidity and mortality is equal
to well-established risk factors such as smoking and obesity.
Early humans evolved as social creatures, quickly recognizing
the value of pooling resources to cope with harsh environmen-
tal demands. Hunter-gatherer societies were more successful
when multiple hunters were available to secure food and ward
off predators, and multiple females more easily tended to the
needs of any offspring. Today, engaging one's social network
during stressful times is recognized as one of the most produc-
tive coping strategies. Indeed, social support has been shown
to be beneficial whether or not one is experiencing increased
stress. Support from others can take many forms, including
instrumental support (e.g., providing groceries or a ride to
work), informational support (e.g., providing information
about an illness), and emotional support (e.g., providing com-
fort and listening). Having the opportunity to talk about an
emotional trauma - when one desires to do so - is associated
with better health. Even the simple knowledge that family,
friends, and others (e.g., counselors, clergy) are available
when needed can buffer the physical and psychological effects
of stress. Though the mechanisms by which social support
works as a defense against the negative effects of stress continue
to be examined, research consistently finds that individuals
who have a supportive network are better able to cope with
stress, and receiving effective social support plays a role in
influencing health and well-being.
Whereas a growing body of work describes these significant
benefits, a broad conclusion that all social interactions are
good for coping would be incorrect. The presence of others
is not always a resource for support and may sometimes be
its own stressor. Seminal work by Karen Rook and others
indicates that not all social interactions are perceived as help-
ful, and some may even lead to unintended negative outcomes
that can be more impactful on the individual than positive
outcomes. For instance, outsiders may have specific expecta-
tions about the process or coping behaviors that should occur
in response to a stressor. A failure to show an expected level of
600 Coping
distress (e.g., length of time to grieve the death of a child), or
not engaging in the expected behaviors, may lead others to
believe that they are seeing signs of pathology. A loved one's
advice may feel more judgmental than assistive, or too many
visitors may exhaust a sick person and prolong the recovery
period. Sometimes the level of support that is expected does
not materialize, or providers offer well-intentioned but ulti-
mately unhelpful comments that minimize the trauma or seem
rejecting and alienating. Researchers have found that for social
support to be effective, it must be matched both with the
environmental demands triggering the distress and the specific
needs of the distressed person.
Religiousness/Spirituality
Many people report a belief in God, though constructions of
God and worship vary greatly, even within a single country. The
study of Western religious philosophies and practices shows
that religion serves two principal functions in coping. First, the
act of believing in a higher power is associated with individuals
finding both emotional comfort and personal meaning from
painful events. This is especially true for situations involving
loss and over which one has no control. For example, a belief in
God may provide comfort upon the death of a loved one, such
as through an image of heaven as beautiful, joyful, and believ-
ing that the loved one is waiting there for reunion. Similarly,
belief in a higher power may help some individuals construct
meaning for events that are dissonant to a worldview of justice
and fairness. Second, regular attendance at religious services
can provide or enhance an individual's social support network.
Having others who share the same beliefs can also be comfort-
ing during difficult life periods. In this same manner, other
religious behaviors, such as being involved in prayer groups
or attending church-related functions, can support strategies
such as positive reframing, and are negatively associated with
typically less effective coping strategies. Religion as a deterrent
to alcohol and drug abuse specifically is apparently achieved via
two mechanisms. The first is a person's behavioral adherence to
the belief held by most religions that the human body should
not be misused (e.g., drinking to excess, drug abuse). The
second deterrent relates to the person's focus of time and atten-
tion on healthy, positive social activities (e.g., a church's soft-
ball team) and away from possible negative influences.
Some Additional Factors
In addition to personality, other characteristics have been
shown to influence the coping process. One such characteristic
is gender. Human subjects research historically focused on the
male gender, and conclusions from these studies (e.g., health
and physiology) were assumed to apply equally to females. An
example specific to stress and coping is Walter B. Cannon's
proverbial fight-or-flight response, wherein a threat or challenge
initiates a rapid sequence of neuroendocrine events to prepare
the individual for maximum effort against danger. While fight-
or-flight is a universal physiologic response, Cannon's model did
not address the actualities that neither fight nor flight is a rea-
sonable behavioral choice for females, especially when they are
tasked with protecting offspring. More recent studies show dis-
tinctively different hormonal responses in females. In particular,
lower levels of testosterone and higher levels of oxytocin trigger
females' behavioral responses to tend-and-befriend, as coined by
Shelley Taylor and colleagues. Tending to the young and
befriending others for mutual support offer females the greatest
likelihood of success in coping with any threat, whether real or
potential. Also to that end, neural (brain) differences in females
better enable the development and maintenance of social net-
works, such as the ability to read body language more accurately
and greater verbal skills relative to males. On the other hand,
females are more likely to appraise situations in a negative light
and tend to rely less on problem-focused strategies than their
male counterparts. These differences are a few examples of how
gender plays a role in the coping process.
Age is another characteristic that influences coping. Studies
have revealed an interesting juxtaposition of generally age-
related losses and gains that older adults have learned to
manipulate to best maintain a sense of control. For example,
older adults are less likely to engage in the active, problem-
focused coping strategies; however, they are more likely to have
a well-established social network on whom to call for assis-
tance. Older people frequently face losses that hold greater
significance than those faced by younger people, from personal
independence to the deaths of loved ones and one's own
declining health. At the same time, age is related to more
positive appraisals of situations, as well as a philosophical
acceptance of uncontrollable circumstances. Often related to
age, a person's level of experience - both in life in general and
with specific or similar events - can influence appraisals. Indi-
viduals may experience the same types of stressors several times
over the life course. Whereas a novel experience requires a
certain amount of energy and resources, repetition of the
same or similar situations maybe related to feelings of comfort
and decreased stress, especially when one has gained wisdom
and mastery over the stressor the first time around.
In addition to enduring characteristics such as gender and
age range, temporal states, such as a person's mood, are asso-
ciated with coping with stress. Appraisals of potentially stress-
ful situations can be influenced by mood to the extent that the
appraisal is, or becomes, consistent with the current mood
state. For example, someone feeling happy may evaluate nega-
tive news in a more positive frame, or may pay less attention
to signs of a potential threat. Conversely, sadness (and in the
longer term, depression) is associated with greater sensitivity
to change and people in negative emotional states tend to
appraise neutral situations as more stressful or threatening
than might be appropriate.
Conclusion
An integral factor in life, stress clearly is unavoidable. Stress,
though, is not inherently bad or negative, and may serve as
an impetus for personal growth. Of importance to human
adaptation and health, then, is understanding the coping pro-
cess as a whole, as opposed to focusing simply on the stressful
circumstance. Two major theoretical perspectives were pre-
sented in this article: coping as a function of personality and
Coping 601
coping as a process. Research has demonstrated that both
individual characteristics and the person-situation process sig-
nificantly contribute, individually and in interaction with each
other, to how a person adapts to change. Features of the
personality influence the coping process by influencing apprai-
sals, strategies, and resources. For nearly all of life's stressful
circumstances, the most successful coping responses typically
include both problem-focused and emotion-focused strategies.
Together, these strategies provide a multifaceted response to
stressors, which often tend to be multifaceted themselves.
Aside from money, perhaps the most important external
resource for coping is social support, which has been shown
to decrease the negative effects of stress both from specific
causes and in life in general. In the next few years, the field of
coping will continue to discover and verify those factors in the
individual, in resources, and in the characteristics of the
stressor itself that predict better coping outcomes. This infor-
mation, in turn, will help guide more effective programs or
treatment for individuals who may be struggling to cope with
a significant trauma or negative event.
See also: Big Five Model and Personality Disorders; Defense
Mechanisms; Hope and Optimism; Id, Ego, and Superego; Learned
Helplessness; Motivation; Perceived Control; Personality, Structure;
Psychological Predictors of Heart Disease; Psychology and Religion;
Stress and Blood Pressure Dysregulation; Stress and Illness.
Further Reading
Aldwin C and Yancura L (2004) Coping and health: A comparison of the stress and
trauma literatures. In: Schnurr P and Green B (eds.) Physical Health Consequences
of Exposure to Extreme Stress, pp. 99-125. Washington, DC: American
Psychological Association.
Aspinwall L and Taylor S (1997) A stitch in time: Self-regulation and proactive coping.
Psychological Bulletin 121: 417-436.
Carver C, Scheier M, and Weintraub J (1989) Assessing coping strategies: A
theoretically based approach. Journal of Personality and Social Psychology
56: 267-283.
Connor-Smith J and Flachsbart C (2007) Relations between personality and coping: A
meta-analysis. Journal of Personality and Social Psychology 93: 1 080-1107.
Hobfoll SE (1989) Conservation of resources: A new attempt at conceptualizing stress.
American Psychologist 44: 513-524.
Lazarus RS (1966) Psychological Stress and the Coping Process. New York:
McGraw-Hill.
Lazarus RS and Folkman S (1 984) Stress, Appraisal, and Coping. New York: Springer.
Rook KS (1984) The negative side of social interaction: Impact on psychological
well-being. Journal of Personality and Social Psychology 46: 1097-1108.
Roth S and Cohen LJ (1986) Approach, avoidance, and coping with stress.
American Psychologist 41 : 81 3-819.
Scheier MF and Carver CS (1985) Optimism, coping, and health: Assessment and
implications of generalized outcome expectancies. Health Psychology 4: 219-247.
Selye H (1 973) The evolution of the stress concept. American Scientist 61 : 692-699.
Taylor S, Klein L, Lewis B, Gruenewald T, Gurung R, and Updegraff J (2000)
Biobehavioral responses to stress in females: Tend-and-befriend, not fight-or-flight.
Psychological Review 1 07: 41 1 -429.
Tedeschi R and Calhoun L (1 995) Trauma and Transformation: Growing in the
Aftermath of Suffering. Thousand Oaks, CA: Sage.
Vollrath M and Torgersen S (2000) Personality types and coping. Personality and
Individual Differences 20: 367-378.
Creative and Imaginative Thinking
M A Runco, University of Georgia, Athens, GA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Convergent thinking Cognition that follows conventional
pathways and is useful for finding correct answers to
closed-ended problems and tasks.
Divergent thinking Cognition that goes in varied and
unconventional directions and often leads to original ideas
and remote associations.
Meta-cognition Tactical and controlled thinking; thinking
about thinking that often contributes to effective creative
problem solving.
Primary process Libidinal thought process,
characteristic of the unconscious and uncensored by
convention and mores.
Secondary process Reality-oriented thought that takes
conventions into account.
Stage models Theories of creative cognition that break the
processes down into phases or steps.
Introduction
Creative and imaginative thinking are related to one another
but far from redundant. This article defines each of them and
highlights exactly how they are related to but separate from one
another. The emphasis here is on thinking, which means that
the most relevant theories and research are found in the cogni-
tive sciences. Still, definitions of creative and imaginative
thinking are far from homogeneous; they are in fact notably
diverse. Some of this is due to the fact that creative thinking
often varies from domain to domain. The obvious example of
this is that it can be more experiential and logical (in a tradi-
tional sense) in mathematics and the sciences and more intui-
tive and subjective in music and the arts.
Diversity is also indicated by the cognitive and extracognitive
processes which have been associated with creative and imagina-
tive thinking. Creative thinking is sometimes intentional, delib-
erate, tactical, and consciously controlled, but other times it is
unintentional, unconscious, and seemingly chaotic. This article
touches on each of these possibilities and reviews the range of
perspectives on creative and imaginative thinking.
Defining Creativity and Imagination
Although there are variations among the domains in which
creativity can be expressed, there is one commonality. Indeed,
all definitions of creativity emphasize originality. Creative
ideas and products must be original; if originality is lacking,
there is no creativity. Most accurately put, originality is neces-
sary but not sufficient for creativity. It is required, but there is
always more than mere originality. Things which are merely
original but not effective or somehow fitting are not creative.
They may in fact be bizzare and useless, odd, weird, or even
unrealistic to the point of psychosis. Psychotic individuals can
be quite original due to their loose contact with shared reality;
but this kind of originality does not constitute creativity since it
is merely original and not effective.
Thus, there are two requirements for creativity: originality
and effectiveness. The former is often operationalized in terms
of novelty or statistical infrequency. In other words, original
things are unique or at least highly unusual. They are unconven-
tional in a literal sense. The second requirement is more difficult
because effectiveness can be objective, as in the case when an
original idea solves a problem in a manner that is widely appre-
ciated, or it can be fairly subjective, as is the case when one
individual sees the esthetic appeal of an idea or work of art, but
others do not see the same thing. A painting may convey mean-
ing very effectively to some people, but not to others. Much the
same can be said about all kinds of creative things; their effec-
tiveness and appropriateness is often fairly subjective.
These two features of creativity go a long way toward dis-
tinguishing it from imagination. That is because imagination is
that which allows an individual to explore ideas and options
that are outside the boundaries of the objective world, ft
has been well described as the capacity to transcend reality. This
can be very useful for creative thinking. Hence, the overlap of
creativity and imagination. More specifically, the capacity to
transcend reality can be used to find or invent original ideas
or perspectives. These may be ineffective, precisely because
they are not constrained by reality. That is how creativity
and imagination differ. The former requires originality and
effectiveness.
There is also a strand of research that defines creativity such
that it always leads to a product of some sort (e.g., an inven-
tion, a work of art, a publication, or perhaps even just an idea),
which makes some sense if you think about the word create
(to bring something into existence). If this product perspective
is held, another distinction is suggested: imagination differs
from creativity in that it may not lead to a product. Imagina-
tion may only allow the transcendence of reality and not result
in a product.
Stage Theories
The bipartite nature of creative thinking (originality and effec-
tiveness) may explain why many theories of creative thinking
look to stages or components. Simplifying, stage models usu-
ally allocate one or more stages specifically to the finding of
602
Creative and Imaginative Thinking 603
original ideas, and then later stages to their refinement or
verification as effective.
Wallas presented a four-stage model of creative thinking
that is still widely cited It describes creative thinking as begin-
ning with preparation, then moving to incubation, illumination,
and finally verification. These four stages are also recognized in
more recent theories of creative thinking, though the labels
sometimes change.
Preparation is a surprisingly important part of the process.
That may be implied by the fact that preparation assumes that
experience is useful. After all, if preparation is influential,
creativity can be enhanced by the right kind of experiences.
Preparation actually takes many forms, depending on the level
and domain of creativity. Preparation may involve learning
and the collection of resources. This kind of preparation can
eventually lead to expertise ; and the highest levels of creativity
(e.g., paradigm shifts) require expertise. Certainly expertise
sometimes makes it difficult to think creatively, as is the case
when it leads to routine and assumption. Both of these make
information processing faster but also can keep the individual
from seeing original ideas. Expertise is sometimes helpful for
creative thinking, and sometimes detrimental.
Another kind of preparation involves the task or problem
rather than the individual. Sometimes the task must be identi-
fied and defined such that creative solutions can be applied
to it. This kind of preparation fits under the label problem
finding, the idea being that problems must be found before
they can be solved. Empirical work has confirmed that prob-
lem finding skills can be reliably assessed and are relatively
independent of the skills used for problem solving. Problem
finding and problem solving can work together, which is prob-
ably the ideal, but some people seem to be better at one than
the other. What may be most important is the implication that
creative thinking is not synonymous with problem solving. It is
not just a special case of problem solving. After all, if creative
thinking is related to problem finding, it can occur before (and
independently of) problem solving.
There are other lines of thought that also support a separa-
tion of creative thinking from problem solving. Creative think-
ing may, for example, occur when there is no problem, as is the
case when someone is proactively creative. Solutions to pro-
blems tend to be reactions to problems, so proactive efforts are
independent of problem solving, at least in the sense that they
precede and may not be dependent on it. Creativity which is
playful or self-expressive rather than an attempt to solve a
problem also support the distinctiveness of creativity from
problem solving.
E. Paul Torrance and J. P. Guilford, two of the most influen-
tial scholars in this field, emphasized that problem finding plays
a central role in creative activity. Torrance, for example, included
problem sensitivity in his definition of creativity. For Guilford,
problem finding was targeted in 'Seeing Problems.' This is a
psychometric task that begins by informing an examinee, "In
each problem of this test you will be given the name of a
common object. You are to state problems that occur to you as
you think about the object. You do not have to think about any
solutions to the problems" (Merrifield and Guilford, 1969: 1).
The object given as an illustration to examinees is 'Candle,' with
examples such as how to light it, keeping it from falling over,
keeping it burning steadily, how long it will burn, and what to do
with the drippings. The actual test includes tree, hammer, wind,
envelope, rug, and sun as the objects for 'seeing problems.'
Guilford is well known for distinguishing divergent thinking
from convergent thinking. The former allows ideation to move
in different and potentially original directions. The latter char-
acterized ideation moving toward the correct or conventional,
and therefore unoriginal, solutions.
Mumford et al. presented a detailed stage model of creative
thinking. They identified stages after an extensive review of the
literature on creativity. The stages in their model are:
problem definition,
information gathering,
information organization,
conceptual combination,
idea generation,
idea evaluation,
implementation planning, and
solution appraisal.
Runco and Chand presented a stage model that included
problem finding as the starting point. This is a unique model
in that it acknowledges the impact of both motivation and
knowledge. The two of these are on one tier of the model, with
(1) problem finding, (2) ideation, and (3) evaluation and valu-
ation on the primary tier. This model acknowledges that cogni-
tion does not work on its own. Many cognitive operations are
only initiated if there is a reason to do so. These may be extrinsic
or intrinsic. Similarly, knowledge can play a role, and it can be
either conceptual, factual, and declarative, or procedural. The
last of these is especially important for creativity that is inten-
tional and tactical. Indeed, procedural knowledge often pro-
vides exactly that - a procedure for finding original ideas
(e.g., 'change your perspective on the problem') or know-how.
A great deal of research has focused on one stage or another.
Many influences on creative thinking, for example, may only
have impact on one stage. Findings suggest that the impact of
expected evaluation varies from stage to stage, which is inter-
esting because evaluation is thought to undermine the intrinsic
motivation which supports creativity. A second line of studies
suggests that the relationship of mood and creative perfor-
mance also varies from stage to stage. This is a significant
finding because, too often, the impact of mood is presumed
to apply across the board, but in fact elation or depression (and
presumably the moods in between) may hinder or facilitate
creative work, depending on the stage in which the individual
is working. Just to name one example, it appears that creative
thinking may suffer from elation or even mild happiness if this
occurs specifically during the evaluation stage of the creative
process. The elated individual may be happy about everything
and his or her 'rose colored glasses' therefore preclude a
detailed, careful examination of options. That individual may
tend toward satisficing, which occurs during problem solving
when one of the first options is accepted, and consideration of
other alternatives terminated.
One of the newer models is called geneplore because it
describes an interplay of generation ('gen') and exploration
('plore'). According to its authors, Ronald Finke, Thomas
Ward, and Steve Smith, creative thinking occurs when the
individual first generates a preinventive form. This is an uncon-
strained but ambiguous cognitive structure. It is subsequently
604 Creative and Imaginative Thinking
extended, elaborated, and evaluated during the exploratory
stage of the process. Finke et al. reported impressive support
for this model in their empirical work.
Uncontrolled Cognition and Creative Syntheses
The geneplore model is intriguing in part because it assumes
that thinking is not entirely controlled, intentional, and
directed. This is a very important idea for creative thinking
because originality may result precisely from phases of the
process which allow uncontrolled ideation. Recall here that
Graham Wallas' stage model included incubation. This is by
definition a time when the individual is either not thinking
about the problem at hand, or, more likely, is working on the
problem but not on a conscious and directed level. The illumi-
nation included in that same model (also known as the
moment of insight or 'ah ha') occurs when the uncontrolled
cognition hits something worthwhile and it is brought back to
the level of consciousness.
The attractive alternative is that creative thinking reflects a
synthesis of controlled and uncontrolled thought, a blend of
conscious and unconscious, primary (libidinal, uncensored)
and secondary (reality-oriented, censored) thought. This
would allow the range of possibilities and motivations of
the uncensored mind to work along with the direction, disci-
pline, knowledge, and critical acumen of the logical mind.
This seems to be what Arieti described as the magic synthesis,
an idea later used by Klaus Hoppe to describe why commis-
surotomy ('split brain') patients have difficulty thinking in a
creative fashion. They are unable to synthesize a freedom of
thought with directed cognition.
Another kind of creative synthesis occurs on the level
of individual ideas or solutions. Arthur Koestler described
how different ideas can come together in what he called the
bisociative process. More recently, Albert Rothenberg demon-
strated how creative thinking may benefit from either Janusian
or homospatial thinking. Janusian thinking was named after
Janus, the Roman God of doorways; he could look in two
directions at once. As is implied by the name, homospatial
thinking allows an individual to bring two images together, in
the mind's eye, to form a meaningful whole. Rothenberg iden-
tified a number of well-known discoveries and insights which
could easily have depended on Janusian and homospatial think-
ing. The Heisenberg Uncertainty Principle, for example, recog-
nizes that the speed and location of a particle cannot be known
at the same time, while Existential Philosophy recognizes both
the infinite and also the absurdity of mortality. Chaos theory
also brings opposites together into 'an orderly disorder' (Glieck,
1987: 15). Most important may be the experimental support
presented by Rothenberg demonstrating that many individuals
do in fact produce creative ideas, solutions, and images when
the synthesis of inputs is experimentally facilitated.
Extra- and Meta-Cognitive Processes
There are, then, a number of ways that creative and imaginative
thinking may benefit from undirected, uncensored, uncon-
trolled thinking. There is at least as much evidence that creative
and imaginative thinking can benefit from controlled, directed,
and deliberate processes. This makes a great deal of sense
given the observation that high-level creative accomplishment
requires work and effort. It can be work and depend on effort.
This again is a reminder of the role of motivation. In fact, there
are several extracognitive influences on creative performance.
For this reason many studies of creativity examine personality
and attitude. One view is that the cognitive capacity is by itself
not sufficient but must be energized and maintained, which
means that the person is motivated to invest the time and effort
into thinking in a creative or imaginative fashion. Yet again
there is a difference between creativity and imagination.
The former is more easily tied to effort, perhaps because it is
not enough to have an original idea. That idea must also be
effective, and verifying or refining the idea may require the
motivation. Imaging alternative realities or experiences, on
the other hand, may require less verification and refinement.
For the same reason it is easy to see the imaginative worlds
and friends of children, even though they do not invest time
into preparation, verification, and refinement.
For similar reasons it is clear that creative thinking can
benefit from intentionally used tactics and strategies. Many of
these have been reported in the literature, including 'work
backward,' 'shift your perspective,' 'take time away from the
task,' 'turn the situation on its head,' and 'break a large prob-
lem down into small ones.' Such tactics and strategies for
creative thought are meta-cognitive in that they are intentional.
The individual is thinking about and trying to guide his or her
own thought. Many programs have been developed to best
communicate tactics to persons interested in creative thinking
and to allow practice with them. In fact, many organizations
do much the same and bring facilitators in for training in
creative thinking and practice with tactics. Organizations tend
to be more interested in creativity than imagination, probably
because the former is clearly connected with effectiveness, and
businesses require effective originality. They often see the role
of creative thinking in innovation. One view is that innovation
is a result of creative thinking. Innovation does seem to require
the same two things, originality and effectiveness, though it
may be that there is more emphasis on the latter in innovation
as compared to creativity.
Conclusions
Although stage theories of creative and imaginative thinking
are common, care must be taken not to delineate them into
entirely discrete stages, phases, or components. It is a bit like
the trade-off of laboratory versus naturalistic research; with
each there is a benefit but also a cost. Laboratory research
offers control, but application to the natural environment is
lost. Similarly, discrete stages may make the process more
manageable and testable, but in reality creativity probably
depends on a spontaneous flow among stages. No doubt
there is recursion rather than a simple step-by-step progres-
sion toward a creative insight.
Recall also that cognitive processes do not operate in isola-
tion. There are extra- and meta-cognitive influences. Motiva-
tion and affect can be quite influential. This is probably true in
a general sense in that some of the directed and intentional
Creative and Imaginative Thinking 605
tactics for creative thinking will not be used unless the individ-
ual is motivated to expend the necessary effort. It is also true in
more specific and idiosyncratic ways, such as when certain
affective disorders provide troubled individuals with advan-
tages for creative thinking.
This article suggested that creative and imaginative thinking
may differ in that the former depends more heavily on effective
thinking. This in turn explains why there are developmental
differences (children having less reason to be effective, often
think in an imaginative way as part of their play) and why
creativity is more often targeted by organizations. They need
to be effectively original, not just original, in order to adapt
to market changes and to perform in an innovative fashion.
It will be interesting as neurophysiological technologies
advance to see if the brain is structured such that there is in
fact both overlap and distinctiveness between creativity and
imagination.
Further Reading
Rickards, Runco MA, and Moger S (eds.) (2009) Companion to Creativity and
Innovation. London: Routledge.
Runco MA (2007) Creativity: Theories and Themes: Research, Development, and
Practice. San Diego, CA: Academic Press.
Singer DG and Singer JL (1990) The House of Makebeiieve: Children’s Play and the
Developing Imagination. Cambridge, MA: Harvard University Press.
Creativity
V Drago, IRCCS Centra San Giovanni di Dio Fatebenefratelli, Brescia, Italy
K M Heilman. University of Florida, Gainesville, FL, USA
Published by Elsevier Inc.
Glossary
Convergent thinking Analytical, usually deductive,
thinking in which ideas are examined for their logical
validity or in which a set of rules is followed.
Depression A psychiatric disorder characterized by an
inability to concentrate, insomnia, loss of appetite,
anhedonia, feelings of extreme sadness, guilt, helplessness
and hopelessness, and thoughts of death.
Divergent thinking Thinking in an unusual way, for
example, to generate several possible solutions to a problem.
Dopamine A monoamine neurotransmitter formed in the
brain by the decarboxylation of dopa and essential to the
normal functioning of the central nervous system.
Norepinephrine A hormone and neurotransmitter, secreted
by the adrenal medulla and the nerve endings of the
sympathetic nervous system to cause vasoconstriction and
increases in heart rate, blood pressure, and the sugar level of
the blood. Also called noradrenaline.
Definitions of Creativity
The construct of creativity has received a great deal of interest
from a variety of disciplines such as psychology and neurosci-
ence, but often there is not much coherence in the findings
across these disciplines. The reason for these inconsistent find-
ings may partially be due to the fact that creativity is a very
complex mental process and thus difficult to study. Different
researchers have not only defined creativity in different ways
but have also approached the study of creativity using different
paradigms and have provided different explanations for the
mechanisms that might underlie creative thinking.
Investigators and experts who judge creativity have not even
come to a consensus about the definition of creativity. When a
person states that something is a creative product or someone
is a creative person, what does that mean? Webster's II Univer-
sity Dictionary (1988) provides several definitions of creativity:
having the power or ability to create, productive, marked by
originality, new. While these definitions do contain elements
of creativity, they are, however, incomplete. For example, there
are many individuals who are musicians, artists, writers, and
scientists who are very productive without being creative.
A person can have technical expertise, work hard, and be
productive but his or her work might not be new or original.
Novelty alone does not make a creative product. For instance,
a list of nonwords and/or colors randomly applied to a canvas
may be novel or original but is not creative.
Another definition of creativity that is widely used in
research is that of Bronowski: Creativity is finding unity in what
appears to be diversity. Great art works have many colors, forms,
as well as textures and great musical works have a large variety
of melodies and rhythms, but creative artists and composers
are able to develop a thread that unifies diverse elements and
display order. Although this definition focuses on the crucial
importance of a 'global approach' within the construct of
creative thinking, it is lacking in concepts such as originality
(novelty) as well as productivity and these concepts are con-
sidered to be essential to the creative process. The concept of
novelty includes two aspects that may be separated: novelty for
a person versus novelty for the world. Researchers often find
that their original or new ideas for future studies have already
been conducted once they perform a review of the literature
and they may learn that they are 'reinventing the wheel.'
Are these instances of creativity, even though the idea was not
new? The answer is 'yes' for that individual, but 'no' for the
society; however, people who perform creative projects need
to be aware of prior creative projects. Computer networks
have been an important advance in allowing creative people
to get to know what has been previously produced.
Many researchers who study creativity believe that, in addi-
tion to being novel, a product must also have some value in
order to be considered creative. Thus, an invention must carry
out the task for which it was designed, a scientific theory must
help us understand the domain in question, a creative work of
art must be appreciated by some audience beyond the artist,
and a creative solution of a problem has to solve the problem.
Csikszentmihalyi and Csikszentmihalyi have presented a
detailed analysis of how novelty and value are both relevant to
creativity. A novel product becomes creative only after it has
become part of the domain, that is, only after it has been posi-
tively valued by the field. In this view, if a product is rejected by
the field, that product is not creative, whether or not it is novel.
Weisberg as well as Heilman, however, disagree with the
concept that products must be valued to be considered creative.
Often, a product is not valued at the time it is produced but
then becomes valued by later generations, or vise versa.
For example, the impressionist painters were criticized when
they first painted using this style. Galileo Galilei was accused
of being a heretic and confined in jail in 1633 when he first
supported the heliocentric theory of Copernicus. It was not
until 180 years later that the Catholic Church officially recog-
nized the Copernican theory as compatible with the Christian
faith. Further, in 1992 the cardinal Poupard finally wrote that
the persecution of Galileo Galilei was a mistake.
For all of these reasons, it may be more useful to separate
the creativity of a product from the value of the product and
to consider a product as creative if it is novel and produced
intentionally.
606
Creativity 607
Another aspect of creativity that many believe is important
is that the product must in some way be beneficial and avail-
able, such that people other than the creator should benefit
from the creative process. Benefit, as used here, does not mean
solely profit and can mean anything from healing (Fleming's
penicillin) to enjoyment (Puccini's Madame Butterfly). For
those who include 'beneficial' in the definition of creativity,
innovative products would not be considered creative if they
are not followed by objective verification, independently of the
value of the idea itself. Thus, for those who believe that 'bene-
ficial' should be a part of the definition of creativity, produc-
tivity is an essential element and if an idea remains solely in the
mind of the researcher or the artist and never becomes a reality
by being produced and judged to be beneficial then there will
be no means of knowing if that idea was indeed creative.
Even to those who believe that ideas, even in the absence of
production can be creative, having a creative idea without a
product, such as a scientific paper, a piece of art, or music, is
like leaving a treasure under the sea.
In this article, we define creativity as the ability to under-
stand (see unity in what appear to be diversity), develop,
construct, and express in a systematic fashion, novel orderly
relationships.
Methods of Studying Creativity
Just as there are difficulties in defining creativity there are also
difficulties in finding the right method to study creativity. Here
we will describe some of the methods that have been used to
study creativity, explaining the limitations and the strengths
that apply to each method.
Autobiographical Self-Reports
The oldest method to study creativity consists of personal reports
from individuals with extraordinary creative accomplishments,
including detailed descriptions of how the individuals con-
ducted their work. These reports come from a range of creative
people across a spectrum of fields, and are in the form of letters,
books, and articles, addresses before scientific societies or other
groups, as well as responses to questionnaires or interviews.
Although these autobiographical self-reports are produced by
people who have been highly creative, these reports are uncon-
trolled, as well as being subjective, and they cannot be fully
assessed for reliability or validity. Even when there is no intent
to deceive, sometimes these reports are even factually inaccurate.
Self-reports are based on introspection and while introspection
can provide some hints as to brain function, neuropsychological
studies, such as those with callosal disconnection, have revealed
that introspection often does not elucidate the brain mechan-
isms underlying a behavior. Finally, self-reports provide us with
only qualitative descriptions of the creative processes, and hence
they cannot test a-priori hypotheses and they do not provide
data that can be used for rigorous scientific analysis.
Biographical Studies
Another way to approach the study of creativity has been
through biographical studies such as Gardner, who collected
data from the biography of many creative individuals such as
Einstein, Picasso, Gandhi, etc. Biographies are based on veri-
fiable historical records and provide a method of directly
studying individuals with the highest level of creative accom-
plishment. One possible limitation of this method, however,
is the quality of the data that is available. Incomplete or
inaccurate records may severely restrict the findings and con-
clusions of these studies. Additionally, biographical studies
have many of the same limitations as do the autobiographical
self-reports discussed above.
Historical Case Studies
Historical case studies are often more focused on the products
of the creator, as opposed to the personal characteristics and
the brain mechanisms that allowed individuals to be creative.
Thus, analysis of historical case studies makes it difficult to
discover the neuropsychological basis of the creative process or
how it can be generalized. Case studies also have the same
limitations as do the autobiographical and biographical stud-
ies as discussed above.
Historiometry
Throughout history there were certain eras and geographic
localities where there was more successful creative productivity
than during other eras and in other locations. These epics and
centers of creativity were often associated with creativity in
certain domains. These observations suggest that environmen-
tal factors can influence creativity. A number of researchers
have applied quantitative methods to historical data in order
to formulate and test casual hypotheses concerning creativity,
using what has been called historiometric analysis. The term
historiometry means 'measuring history.' As an example of
historiometric analysis, Simonton was interested about how
war and other sources of social upheaval might influence
creative productivity. For example, during World War II there
was a flurry of creativity. In medicine, penicillin was first used
and it was learned that it was healthy to mobilize patients
shortly after their surgery. We also learned much about atomic
physics. Simonton divided the last two millennia into 20-year
'epochs' and determined the frequency of creative accomplish-
ments within each of these epochs and assessed the influence
of environmental factors such as war. For example, the fre-
quency of creative accomplishment was based on such mea-
sures as the number of years of active warfare within each
epoch and the number of creative individuals who flourished
during each epoch.
Using statistical methods, Simonton attempted to infer
causal relationships from historical data and concluded that
the occurrence of war involving a nation results in a decrease in
creative accomplishments in that nation for the following
epoch. The strength of the historiometric method is that it
directly involves creative accomplishments at the highest
level. However, this method is also limited by the availability
of the data and although it might provide data about the
influence of environmental factors on creative production it
does not help elucidate the brain mechanisms underlying
creativity.
608 Creativity
Quantitative Case Studies
This method applies quantitative measurements to the analysis
of individuals or groups of individuals over a period of time.
Quantitative approaches to case reports sometimes result in
discoveries that would not have been apparent from a more
qualitative presentation of historical information. However, a
potential problem of this approach consists of the serious
threats to external validity that result from generalizing the
results of a single case to a wider range of individuals and
circumstances. Thus, it is of paramount importance for other
researchers to replicate the findings from case studies using a
range of creative domains, settings, fields, and individuals.
Systematic replication will result in enhanced generalization
and understanding of the creative process.
Empirical Investigations
Empirical investigations of creativity appear to be the premier
means of learning why people are creative. Empirical studies
require the development of an a-priori hypothesis, the devel-
opment of the tests and methods to test the hypothesis, the
analysis of the data, and the interpretation of results. These
studies require the recruitment of participants as well as the
apparatus and procedures that can test the a-priori hypothesis.
These experimental investigations include attempts to directly
manipulate creativity through random subject assignments
and typically require subjects to perform various tests of crea-
tivity under a variety of conditions.
Experimental approaches have the advantage of exerting the
highest level of control over potential confounds, thereby per-
mitting investigators to draw cause and effect relationships.
Although these are widely used methods to study creative
thinking in normal individuals, there are limitations on the
data that are derived from laboratory studies, as well as ques-
tions as to its ecological validity.
As we have discussed, each of the aforementioned methods
have strengths and weaknesses. Thus, depending on the ques-
tion that as been posed and the answer that is sought, research-
ers will use one method or another, while bearing in mind the
potential strengths and limitations of the selected method.
Stages of Creativity
Socrates thought that 'inspired thinking' arrived from divinities,
specifically the muses or the nine daughters of Zeus, each of
whom was responsible for a separate domain. According to this
belief savants are inspired instruments through which Nature
shows humanity the secrets of underlying discoveries. Contem-
porary views, however, regard creative ideas as emanating from
within the individual rather than from outside. Consistent with
this more modern view, many studies have focused on identify-
ing the brain mechanisms that underlie creativity. Helmholtz
and Wallas suggested that creativity has four stages: preparation,
incubation, illumination, and verification.
Preparation
In order to be creative an individual has to acquire the basic
knowledge needed to develop creative ideas and skills
necessary to develop novel and creative products. Kuhn who
wrote a very influential book on creativity noted that many
important discoveries are initiated by the observation of an
anomaly. An example would include a scientist perceiving
significance in an accidental occurrence, such as Alexander
Fleming's discovery of penicillin. Fleming left bacterial cultures
(staphylococci) in his lab exposed to the air and did not close
the windows in his laboratory. The fungus mold penicillin flew
into his laboratory and landed on these cultures. When he
looked at the cultures he noted that the penicillin mold had
killed the adjacent bacteria. Although these types of discoveries
are based on accidents, it is the 'prepared mind' that enables
scientists, such as Fleming, to perceive the importance of the
phenomenon they observed. Fleming was a well-trained bacte-
riologist who during World War I saw many men die of infec-
tion and found that antiseptics would not cure these infections.
Creative imagination takes advantage of the ability of our brain
to displace different life events in a new and original way, and
the more rich and varied the experiences of a person, the more
his or her imagination will have material from which to con-
sider the problem.
Incubation and Illumination
When a person such as a scientist is confronted with a problem,
the creative solution of this problem is often not immediately
apparent. Then after a period of time the solution to the
problem may enter this scientist's consciousness. Thus, it is
highly probable that this creative scientist was incubating his
or her ideas about the phenomena prior to becoming aware of
the solution. This sudden awareness of a solution has been
called the 'Aha!' experience and it is this 'Aha' experience or
epiphany that Hemholtz and Wallas termed illumination.
Whereas Fleming's 'Aha' experience was aided by observing
an anomaly, many scientists and artists develop an illumina-
tion even without observing an anomaly. Henri Poincare
(1854-1912), a world-renowned mathematician and physi-
cist, reported that he had several illuminations when describ-
ing his creative achievements. The occurrence of illuminations
may then be considered evidence for unconscious processing.
If 'Aha!' experiences do not originate from conscious thinking,
then there has to be an unconscious incubation as an expla-
nation for sudden illumination. Support for the concept that
unconscious or subconscious incubation or thinking may
lead to solutions come from more recent work by Strick et al.,
who demonstrated that following distraction from the prob-
lem for a period of time there was an improvement of
some judgments. Weisberg, however, also noted that creativity
does not always require great leaps following unconscious
processing and the processes that lead to many great discov-
eries did not involve subconscious incubation, but rather a
series of conscious steps.
Verification and Production
After illumination, creative people, such as scientists, perform
experiments that attempt to test their new hypotheses, artists
paint their pictures, writers write their books or poems, and
composers write or play their music.
Creativity 609
Intelligence and Creativity
There was a significant change in the direction of research on
creativity around 1950 when Guilford, an expert on intelli-
gence testing, surprised many people by proposing that psy-
chology had not spent enough time examining thinking that
went beyond the kind of cognitive processes and knowledge
measured by intelligence or IQ test, namely, creative thinking.
Definition of Intelligence and Relationship to Creativity
There are many definitions of intelligence, but for most psychol-
ogists, intelligence is the measure of a person's ability to acquire
and apply knowledge. Sternberg and O'Hara suggested four pos-
sible relationships between intelligence and creativity: (1) intelli-
gence is the same as creativity; (2) creativity is a subset of
intelligence; (3) intelligence and creativity are unrelated; (4)
these are overlapping but independent constructs. If intelligence
is considered a measure of a person's cognitive ability to adapt,
then creativity is a gift that might enhances one's ability to use
their cognitive resources to better adapt. Whereas this statement
would hold for certain forms of creativity, such as medical science
(i.e., discovery of penicillin) and engineering, this definition
would not hold true for other forms of creativity, such as the
arts, including music, dance, painting, writing of poetry and
fiction, etc. If, however, the definition was expanded so that
intelligence is both a measure of a person's ability to adapt and
to enhance the quality of life, then the arts could also be included.
One of the founders of intelligence tests, Alfred Binet, must
have initially thought that creativity and intelligence were the
same or closely overlapping constmcts because the first intelli-
gence test he devised in 1896 used inkblots to explore the
imagination of children. Later, according to Sternberg and
O'Hara, he discontinued this inkblot test because he was unable
to develop a reliable means of scoring it. This test, however, has
been subsequently used as a measure of creativity.
Guilford and Christensen, who are among the first to study
and help increase psychologists' interest in creativity, thought
that creativity was a subset of intelligence and Guilford
attempted to develop psychometric tests that could measure
creativity. These tests are similar to those developed by Tor-
rance. Most of these tests assess the ability to develop novel
uses of common objects. For example, subjects would be asked
to name in a fixed time interval the different ways in which
they might be able to use a brick.
Another means of studying the relationship between crea-
tivity and intelligence is studying the intelligence of creative
individuals. Barron and Harrington in their article 'Creativity,
intelligence and personality' suggest that there is a curvilinear
relationship between intelligence and creativity, with intelli-
gence presumably becoming less and less influential as one
moves to the higher levels of intelligence. This observation
suggests that there might be an IQ threshold, such that a
person needs to be above this threshold for him or her to
have sufficient intelligence to acquire the knowledge about
the domain of his or her creativity and to acquire the skills
needed to be creative in that domain. Thus, intelligence is a
necessary but not sufficient component of creativity.
Other investigators, such as Herr et al. and Simonton, have
also studied populations of known creative people and
attempted to learn if there is a strong correlation between
their estimated eminence as creators and their intelligence.
They found that the correlation between intelligence and crea-
tivity is weak. This weak correlation, however, might be related
to the tests that were used to measure intelligence.
Crystallized and Fluid Intelligence
Cattell posited that there are two types of intelligence, which he
termed, 'crystallized' and 'fluid.' Crystallized intelligence is basi-
cally declarative memories, such as knowing that Rome is the
capital of Italy or lexical-semantic knowledge, such as knowing
the meaning of the word 'impale.' In contrast, fluid intelligence
is defined as the ability to solve problems. Most intelligence
tests, such as the Wechsler Adult Intelligence Scale (WAIS),
assess both crystallized (e.g., vocabulary definitions), and fluid
intelligence (e.g., similarities - How are a fly and a tree similar?).
Cattell thought that while crystallized knowledge is important
in creativity, it is fluid intelligence that primarily determines
creativity. Although fluid intelligence may be the best predictor
of creativity, we know of no formal studies that have studied this
relationship and there may different domains of fluid intelli-
gence. In addition, there has been little written about the brain
mechanisms of fluid intelligence and cognitive flexibility.
Divergent and Convergent Thinking in the
Assessment of Creativity
Divergent Thinking
Guilford reasoned that an important step in the creative process
must be a breaking away from the past, which is the function of
what he called divergent thinking. Divergent thinking is the
ability to take different directions from the prevailing modes of
thought or expression. As the name implies, this type of thinking
breaks away from established concepts and produces novel ideas,
which can serve as the basis for the development of a creative
product. Divergent thinking often produces multiple new ideas.
Convergent Thinking
In contrast to divergent thinking, convergent thinking involves
seeing-recognizing the similarity or the commonality of phe-
nomena that appear to be different or combining and joining
two or more different observations or ideas together so that
they provide additional information, for example, putting the
pieces of a puzzle together so they make a complete picture.
One of the best and classic examples of a paradigmatic shift
using convergent reasoning is the Copernican heliocentric the-
ory of planetary motion. Convergent reasoning or thinking is
similar to that which has been termed deductive reasoning.
Unlike divergent thinking with convergent thinking there is
usually a correct answer. Almost all multiple choice tests that
assess reasoning rather than knowledge require the person
taking the test to use convergent reasoning.
Tests Assessing Components of Creativity
Guilford's work was the stimulus to the psychometric stream
of creativity research, which has focused on measuring the
610 Creativity
psychological characteristics of creative people (psychometric
meaning 'measuring the mind'). Guilford and others have used
their tests in attempting to measure the thought processes
underlying creative thinking. Additionally, other investigators
have developed other tests to conduct similar investigations.
As mentioned, creative endeavors require preparation, inno-
vation, and production. Innovative thinking requires that a
person first break away from prior solutions (disengagement),
develop alternative solutions (divergent and associative think-
ing), and then learn which solution works best (convergent
thinking). In this and the next section we will briefly describe
some of the tests that have been used to study creative thinking,
with the understanding that different tests explore different
components of the creative process. The tests mentioned here
represent only a portion of the tests that are widely used to
investigate creativity and there may be many other tests.
The Wisconsin Card Sorting Test
The Wisconsin Card Sorting Test is a test of disengagement and
divergent thinking. The task involves presenting a series of
cards containing stimuli that vary along several dimensions,
including shape, color, and form. The subjects are asked to sort
the cards by matching the stimulus cards with a key card that
has the same characteristic (e.g., shape). Once the subject
demonstrates the ability to sort on one variable the examiner
switches to a new sorting principle and the subject taking this
test must find the new sorting principle (e.g., color). This
process is repeated until all the cards are sorted. The score on
this test represents the ability of subject to successfully disen-
gage and switch sorting strategies. Patients with frontal lobe
damage exhibit impaired performance on this test.
Alternative use of objects task
In this test the subjects are given three different objects (e.g., a
brick) and are asked to name as many different uses for the
object as possible in a specified time. The individuals are
informed that their performance will be scored based on the
number of uses as well as the degree of novelty-innovation for
the uses they identify.
Each specified use is scored as a 1 for a common use (such
as using the brick to build), 2 for a minimally innovative use
(use the brick as a door stop or book end), and a 3 for
moderately or highly innovative uses (grind the brick up and
using it as a cleaning abrasive or as make-up.) The total score is
the sum of the item scores.
When an area of brain is active and its neurons are firing,
more blood is carried to that area. Carlsson et al. conducted a
study looking at cerebral perfusion in low versus high creative
individuals during the alternate use of object task. The results
appear to indicate an increase in bilateral frontal activation in
the highly creative individuals.
Alternative meanings test
Many words in English have multiple meanings. For example,
the word pen can mean a writing instrument or it can be a place
where animals (e.g., pigs) are kept. This test involves providing
subjects with a series of words and having these subjects provide
two sentences using each word, but in each sentence the word
should have a different meaning. After the subject produces two
sentences a new word is given. If the two sentences are not
given within 60s, the examiner provides the next word. The
score is the time taken to successfully complete the test.
Tests Assessing Convergent Thinking
Verbal-word anagram task
Anagram tests involve providing individuals with words where
the letters that comprise the words are arranged out of order.
The task is then to recognize the word that would be spelled if
the letters were in the correct order (such as LPSENID being
identified as SPINDLE). This task has been used in several
studies in our laboratories where we examined the effects of
central norepinephrine blockade on creativity. The anagram
test is often composed of 10 words and contains five 5-letter
words, three 6-letter words, and two 7-letter words. Subjects
are shown the scrambled letters of a word printed on a sheet of
paper and asked to write down on an answer sheet the word
that these letters spelled. The score for each word is the number
of seconds required to identify the correct word. If the subject
makes an error or does not write the correct word in 100s, a
maximum score of 100 is given. The lower the score the better
the ability to perform convergent reasoning.
Verbal-remote associates tests
The remote associates test (RAT) requires respondents to access
the semantic representations for three presented words and
find the one correct fourth word that is linked with the three
presented words. The presented words may be related to the
solution words in a number of ways, they can be semantically
or associatively related, they may be two parts of a compound
word, or they may be words used as synonyms. For example,
the subject may be provided with the words 'falling,' 'famous,'
and 'dust,' with the solution word of 'star.' 'Falling' and 'star'
are associatively linked, 'famous' and 'star' are synonyms, and
'stardust' is a compound word. There are often 20 items in this
task and the response to each item is scored for accuracy and
time to respond.
Tests Assessing Both Divergent and Convergent Thinking
Insight problems
The concept of insight problems began with Gestalt psychol-
ogy. Insight problems are problems which are simple to state
but relatively difficult to solve, and require some sort of
'insight' (creative thinking) to solve them. Such problems initi-
ally seem overwhelming and without solution. However,
insight may suddenly occur, giving a new perspective on the
situation and an answer that clearly fits. There are different
kinds of insight problems, some of them are verbal problems
and others are more based on mathematical or spatial tasks.
An example of an insight problem is the 'Nine Dot Prob-
lem' or 'Thinking Out of the Box.' The subject is asked to
connect nine dots in a 3 x 3 matrix with four connected straight
lines without lifting the pen from the paper (see Figure 1(a)).
The solution can be seen in Figure 1(b).
Revised Torrance Test of Creativity
The standard Abbreviated Torrance Test for Adults (ATTA)
consists of three activities, one verbal and two visual. During
the verbal activity the subject is asked the following question.
Creativity 61 1
(a) (b)
Figure 1 Thinking out of the box.’ The person performing this test is
asked to connect all the dots in (a) with four straight lines, without
removing the pen from the paper. The solution to this problem
is show in (b). In order to correctly perform this test, a person must
go ‘outside the box’ formed by the dots on the perimeter.
"Just suppose you could walk on air or fly without being in an
airplane or similar vehicle. What problems might this create?
List as many as you can."
The second activity consists of giving the subject a paper on
which there are two incomplete drawings. The subjects are then
asked to create meaningful drawings that incorporate these
incomplete figures and to title their drawings. The third activity
consists of giving the subjects a sheet of paper that contains
nine isosceles triangles. The subjects are asked to make as many
pictures as possible using these triangles or lines. The subjects
are told that every picture should have a meaning and a title.
Subjects are given three minutes to complete each activity.
Neurological Diseases, the Right Hemisphere, and
Creativity
As mentioned in the introduction, through investigating the
behavioral changes associated with brain dysfunction, scien-
tists can understand the components that constitute complex
cognitive functions and behaviors. We also mentioned the
critical components of creative thinking including disengage-
ment, divergent, associative, and convergent thinking with
incubation and illumination (finding the thread that unites),
as well as productivity. In the following section we will provide
examples of the effects of neurological diseases on various
cognitive mechanisms that are important in creative thinking.
Bruce Miller and his coworkers wrote a series of reports on
the emergence of visual artistic talents in patients affected by
frontotemporal lobar atrophy (FTLA). Two of the most com-
mon forms of degenerative dementia that are seen in clinics are
Alzheimer's disease and FTLA. Most patients with Alzheimer's
disease initially experience problems with episodic memory
and then develop other cognitive deficits, such as problems
with naming, route finding, constructional apraxia (drawing,
copying), and even performing learned skilled movements
(ideomotor apraxia). The signs and symptoms associated
with frontotemporal or dysexecutive form of FTLA are different
from those seen with Alzheimer's disease. In one form, patients
lose their social skills and engage in socially inappropriate
behaviors. Other patients may become abulic and sit around
all day doing nothing useful. Still other patients with fronto-
temporal dementia might experience language deficits that
cause speech to become progressively less fluent (progressive
nonfluent aphasia) or they might experience difficulty with
naming and comprehension (semantic dementia). Addition-
ally, unlike patients with Alzheimer's disease, patients with
FTLA dementia may have relatively intact visual-spatial
memory.
Brain imaging, such as magnetic resonance imaging (MRI),
of patients with Alzheimer's disease usually demonstrate evi-
dence of atrophy, primarily in the medial temporal lobes and
in the parietal lobes. Conversely, patients with FTLA often
evidence atrophy of the frontal lobes or the anterior temporal
lobes or both. Further, FTLA might be asymmetric and primar-
ily involve only one side of the brain. For example, patients
with the semantic dementia form of FTLA most often have
atrophy of the left anterior temporal lobe.
Microscopic examination of the brain in patients with Alz-
heimer's disease reveals the presence of amyloid depositions
(plaques) and neurofibrillary tangles in the neurons. Patients
with FTLA have a variety of different microscopic changes,
including intraneuronal inclusions called Pick bodies (after the
neurologist who first described this syndrome), ballooned neu-
rons that stain poorly, or even no specific pathological markers.
Although it is often difficult to know, in retrospect, when a
dementing disease has started several of the artists described by
Miller and his coworkers (1998-2000) appeared to have
started drawing or painting prior to the time they had the
symptoms and signs of FTLA dementia or at the time when
the disease was just beginning. What was remarkable about
these patients is that in spite of their dementia they continued
to paint and some even improved their artistic skills.
Whereas the left hemisphere, in right-handed individuals,
is dominant for language and primarily mediates verbal activ-
ities, the right hemisphere appears to be more important for
the type of spatial skills needed when painting. Miller and his
coworkers noted that the brain atrophy of most of these artistic
people was limited to the left frontotemporal lobes, sparing the
right side of the brain, and their creative skills were nonverbal.
Based on the reports by Miller et al. however, we cannot
ascertain whether these patients' talents existed before the
onset of the dementia.
Similar reports have been subsequently described from
other authors. Finney and Heilman conducted a study with a
patient with the progressive nonfluent aphasia form of FTIA by
having independent judges score a variety of artistic qualities of
seven of this patient's paintings completed before the onset
of this disorder and seven paintings competed after the onset of
the disease. The qualities scored included: (1) Novelty: 'How
original is this painting?' (2) Aesthetics: 'How beautiful or
attractive is this painting?' (3) Representation: 'How well is
the subject of this painting rendered?' and (4) Closure: 'How
complete is this painting?' The results of this assessment indi-
cated that the patient's artistic ability neither diminished nor
improved with the onset and progression of the disease, with
the exception of novelty, which declined over time.
Drago et al. conducted a similar investigation by studying
40 paintings of an artist affected by the semantic dementia
form of FTLD. Their findings showed an increase of some
visual artistic skills over time. To help explain why brain dys-
function can result in the emergence of a new or improved
skill, Kapur suggested that facilitation of a function might be
612 Creativity
induced by a lesion that destroys an inhibitory circuit (i.e., a
'paradoxical functional facilitation'). Miller and his coworkers
suggested that perhaps a similar phenomenon was occurring in
their patients and the left anterior temporal lobe degeneration
"contributed to the unexpected emergence of talent . . This
disinhibition hypothesis has also been used to explain creativity
in individuals who are without dementia. For example, Einstein
noted that he often used spatial reasoning to develop his creative
theories. Einstein did not start speaking until the age of three,
was probably dyslexic, and had abnormal development of
his left parietal lobe. It is possible that these left hemisphere
developmental abnormalities allowed his right hemisphere
to develop exceptional spatial skills and talents. Although
these aforementioned studies provide some important insights
concerning visuospatial abilities, this right hemisphere dis-
inhibition hypothesis has not been experimentally tested.
Some support for the hypothesis that the right hemisphere
is important in creativity comes from a recent imaging study by
Asari et al., who examined unique perception using functional
MRI (fMRI). During the study 68 normal subjects were asked
to look at ten ambiguous figures taken from the Rorschach
inkblot test and state what the figures resembled. The re-
sponses were then classified as frequent, infrequent, or unique.
The results indicated a greater activity in the right temporopo-
lar region at the time of the unique response, as compared to
the frequent responses. These results provide support for the
idea that the right hemisphere is integrally involved in the
visuospatial creative process.
Based on the hypothesis of a 'paradoxical functional facili-
tation,' Drago et al. studied an artist affected by Parkinson's
disease (PD) who had a left deep brain stimulator (DBS)
implanted. DBS can induce activation of one or both hemi-
spheres. Thus the purpose of the study was to learn if DBS of
the left hemisphere, presumably inducing left hemisphere acti-
vation, would alter the creativity of a professional artist as well
as her ability to judge somebody else's art. The authors had
the patient perform a test of creativity with the stimulator 'on'
versus 'off.' The patient was also asked to judge the art of other
individuals when the stimulator was both on and off. The
results suggested that left ventral subthalamic nuclei/substantia
nigra reticular deep brain stimulation (STN/SNr DBS) reduces
creativity as well as appreciation of art. The reason for these
alterations is not known, but might be related to enhanced
activation of the left hemisphere and reciprocal deactivation
of the right hemisphere. The right hemisphere is thought to
mediate visuospatial skills and global attention, both of which
might be important in artistic creativity and appreciation of art.
The hemispheric lateralization model grew in the 1970s.
Researchers espousing this model postulated that the nondom-
inant hemisphere was specialized for creative activities such as
art and music. At the beginning of the article when creativity
was defined we raised the concept of 'finding the thread that
unites' and 'finding unity in what appears to be diversity.' The
right hemisphere appears to be more important in global than
local processing and a global approach may be important in
finding the thread that unites. The right hemisphere is also
important in mediating visual-spatial functions and as men-
tioned, many extremely creative people, such as Einstein, claim
to have used visual-spatial strategies to help find creative
solutions. Thus, the hemispheric lateralization model suggests
that the right hemisphere mediates functions that are crucial
for creative processing.
An EEG study conducted by Jausovec and Jausovec recorded
from 115 normal participants during resting, 'eyes open' and
'eyes closed' conditions. He found that when his subjects were
stratified across measures of creativity and intelligence, EEG
coherence (obtained during 'rest' with eyes open) was signifi-
cantly related to creativity scores, particularly across the right
hemisphere. Coherence provides information about the cooper-
ation of different areas of the brain as well as the functional
relations between brain regions. Finding 'the thread that unites'
might require the cooperation of a variety of modular networks
that are distributed in different anatomic areas and thus mea-
sures of coherence might provide evidence of modular coordina-
tion, ie associative thinking that is important in creativity.
The Role of the Frontal Lobes and the Influence of
Dopamine on Creativity
According to Beals, Thomas Edison said that being a creative
genius required "ninety-nine percent perspiration and one
percent inspiration." Almost all creative people, independent
of the domain in which they create, may be characterized by
perseverance and persistence.
Goal-oriented behaviors or volition permeates almost all
aspects of creativity, and the major organ of volition appears to
be the frontal lobes. Clinically, patients with degeneration or
injury of the frontal lobes present with a loss of initiative and
drive, and this loss is called 'abulia.' Harlow, in 1868, reported
the famous case of Phineas Gage, one of the best-known
descriptions of a person who developed abulia from a brain
injury. The ability to have long-term goals and to suppress
biological drives when they interfere with long-term goals, as
well as the ability to persist and not be distracted, is what
Heilman calls 'frontal intelligence.' Frontal intelligence is one
of the major factors underlying success in any profession,
including those that require creativity. However, the frontal
lobe networks have another function that appears to be impor-
tant for creativity, namely, disengagement and divergent think-
ing. As we mentioned above, divergent thinking is the ability to
take a different direction from the prevailing modes of thought
or expression and as also mentioned, Zangwell and Milner
suggested that frontal lobe damage or dysfunction would dis-
rupt divergent thinking.
Electromagnetic studies demonstrate that frontal lobe
activity appear to be important in creativity. Comparing sub-
jects with high and low creativity has revealed that the highly
creative individuals possess higher baseline frontal lobe activ-
ity and greater increases in frontal lobe activity while per-
forming creative tasks. Further, there is preliminary evidence
that transcranial magnetic stimulation applied over frontal
lobes can increase frontal activation as well as creativity in
normal subjects during both drawing and writing tasks. There
are also case reports of patients whose creativity increased after
receiving subcortical deep brain stimulation with the electro-
des placed near the nucleus accumbens. The connections of the
nucleus accumbens to the frontal and temporal lobes, and its
role in limbic generation of drives and reward may help
explain this effect. The dorsolateral prefrontal cortex supports
Creativity 613
the planning and organization of artistic efforts, while the
cingulate cortex modulates attention, drive, and emotion.
Also, the motor and premotor frontal regions (along with the
basal ganglia and cerebellum) carry out the precise movements
needed for great art, dance, and playing of musical instru-
ments. In addition, there is some evidence suggesting that
creative drive may be controlled through interactions among
the temporal lobes, the frontal lobes, and the limbic system.
Particularly, creative drive increases with temporal lobe dys-
function and increasing dopaminergic activity. In contrast,
reduced creativity (creative block) is associated with deficits
in frontal lobe activity, or decreasing dopaminergic activity.
Creative people have also been shown to have higher
baseline arousal and heightened responses to sensory stimu-
lation. Dopamine decreases latent inhibition, a behavioral
index of the ability to habituate to sensation and raises base-
line arousal. Thus, the focused aspect of creative drive may be
modulated by mesolimbic dopaminergic activity. Dopamine
also mediates reward seeking activity, ranging from gambling
and cocaine addiction to the appreciation of beautiful faces
and music. Conversely, too much dopamine may cause exces-
sively focused, highly complex motor stereotypes, such as
repeatedly disassembling and reassembling one's motorcycle
engine. Dopamine may also play a role in creative discovery
through its effect on novelty seeking. An allele of the D4
receptors has been postulated, somewhat controversially, to
be a novelty-seeking gene.
A recent study by Drago et al. examined patients affected
by PD performance on a test of creativity. The patients were
grouped based on side of onset, with the hypothesis that when
compared to healthy controls, patients with a right hemibody
onset would be more impaired in verbal creativity and patients
with left hemibody onset would have more problems with
visuospatial creativity. The results indicated when compared to
control participants there was a decline in verbal creativity in the
group of patients with right hemibody onset PD.
The neurobiology of creativity has also been addressed using
single photon emission computerized tomography (SPECT),
positron emission tomography (PET), and fMRI. Chavez-Eakle
et al. used SPECT to study 12 highly creative subjects while
performing figural and verbal creative tasks. These authors
found a positive relationship between the creativity index and
cerebral blood flow in the right post central gyrus, bilateral rectus
gyri, right inferior parietal lobule, and right parahippocampal
gyrus. These findings were interpreted as supporting a 'highly
distributed brain system' underlying creativity. PET was used to
study normal subjects as they performed verbal creativity tasks,
with significant brain activation being observed in the left parie-
tal-temporal brain regions (Brodmann's areas 39 and 40),
which are considered to be 'crucial' to the creative process.
An fMRI study attempted to localize creative story generation
within the brains of a cohort of eight normal subjects. The
results indicated that when creative story generation was con-
trasted to uncreative story generation, significant activations
were observed within bilateral medial frontal gyri (Brodmann's
areas 9 and 10) and the left anterior cingulate gyrus (BA 32).
Across all studies, however, no clear consensus has emerged as to
whether frontal or more posterior brain regions are more central
to creative processes. There appears to be no single creativity
node, but rather a complex of interactive modular networks.
The Role of White Matter Connectivity in Creativity
Creativity was defined in the beginning of this chapter as the
ability to understand, develop, and express in a systematic
fashion novel, orderly relationships. This ability might require
intra- and interhemispheric communication between several
brain modules. Works of scientific or artistic creativity often
require the skills and knowledge mediated by both cerebral
hemispheres. For example, the sculptor must imagine the rota-
tion of spatial images mediated by the right hemisphere while
he or she uses the motor skills mediated by the left hemisphere.
Thus, intra and interhemispheric communication might be
crucial for integrating the knowledge and skills that are impor-
tant for creative innovation.
Corpus Callosum
The largest structure connecting independent modular systems
is the corpus callosum. Lewis administered the Rorschach test
to eight patients affected by medication-resistant epilepsy
before and after they had undergone a cerebral commissurot-
omy and noted that disconnection of the two cerebral hemi-
spheres tended to destroy creativity as measured by this test.
Frederic Bremer suggested that the corpus callosum subserves
the highest and most elaborate activities of the brain, in a
word, creativity.
The corpus callosum is primarily comprised of myelinated
axons whose cells bodies are in the pyramidal layers of the
cerebral cortex. The cerebral connections important for creativ-
ity, however, might not only be interhemispheric, but also
intrahemispheric.
Intrahemispheric Connectivity
In addition to the myelinated axons that carry information
between the hemispheres, there are white matter connections
between the thalamus, basal ganglia, brain stem, and the cor-
tex. Myelinated axons also carry information between cortical
regions in the same hemisphere. These intrahemispheric con-
nections facilitate intrahemispheric communication, which
also might be important for creative innovation because wide-
spread connectivity allows creative people to combine the
representations of ideas that have been previously isolated.
In addition, these intrahemispheric white matter connections,
that include connections between the frontal lobes and the
posterior association areas in the temporal, parietal, and occip-
ital lobes, might allow the selective activation or inhibition of
cognitive modules.
Connectionist, or parallel distributed processing (PDP)
models, suggest that there are processing units or nodes that
are similar to the neurons found in the brain and PDP
models strongly emulate the fundamental properties of neural
networks. Information in PDP networks is stored in the
strengths of connections between units (as in the brain). Con-
cepts are represented as patterns of activity involving many
units (i.e., as distributed representations) and large number
of units linked by a set of connections define a domain of
knowledge from which any one of a large number of concepts
can be generated.
614 Creativity
Psychological studies of priming effects on the latency of
lexical decision tasks are a good demonstration of the ability
of this model to account for empirical results. In a lexical
decision task, words are flashed, one at a time, on a screen
and the subject has to indicate, as rapidly as possible, if the
word displayed is a real word or a pseudo word. In a priming
paradigm, before the target word or a nonword appears on the
screen, a prime word appears. If the prime is strongly related
to the target (e.g., the prime is stripes and the target is tiger),
the response time for word (e.g., tiger) is less than when the
prime and target are unrelated (e.g., doctor-zebra). This reduc-
tion of reaction time to primed words is thought to occur
because when a related prime appears it generates a distributed
concept representation that involves activation of many of the
units that define the distributed representation of the target.
As much of the target's distributed representation is already
activated when the target appears, response latency to the target
is reduced.
Mednick suggested that in generating associative responses
to a stimulus, creative individuals are characterized by a more
flat associative hierarchy than are less creative individuals.
Hence, creative people might have the ability to activate more
highly distributed networks and this ability allows them to see
'the thread that unites.'
Relaxation, Depression, and Creativity
Sleep and Arousal
Several scientists reported being able to solve difficult scientific
problems during sleep or when they were either falling asleep
or awakening from sleep. One of the most famous examples
of this phenomenon is August Kekule, who in 1865, while
attempting to understand the structure of benzene, had a
vision, during a dreamy state, of a snake chasing its own tail.
This vision provided Kekule with the idea that benzene was a
ring-like structure.
Before and after sleep, people who are actively working on a
problem describe moments of insight when they were able
to solve previously insoluble problems. Often the solution to
difficult problems is aided by the ability to make remote asso-
ciations and these moments of insight often come at time
when the person is relaxed and at rest. Eysenck suggested that
during conscious problem solving, cortical arousal is often
high. This high level of cortical arousal narrows associative
fields and suppresses the ability to make remote associations.
A lowering of arousal, however, might allow these remote
associations to emerge.
Support for the postulate that the level of arousal might
determine the size of neural networks comes from some
research performed by Contreras and Llinas. Using high-
speed optical imaging, they performed electrical stimulation
of subcortical white matters in slices of brain from a guinea pig
and recorded the portions of the neocortex that were activated
by this subcortical stimulation. They found that with low-
frequency stimulation cortical activation was initially limited
but then after few milliseconds the activation spreads into
nearby areas. After high-frequency stimulation the excitation
remained fixed to a small column of neurons that were directly
above the stimulating electrode. Recordings from the neurons
around the excited column during rapid stimulation revealed
increased inhibitory synaptic activity that probably inhibited
the spread of activation.
Depression and Bipolar Disorder
Several researchers have investigated the relationship between
depression and bipolar disorders in otherwise healthy creative
individuals who were or are composers, scientists, artists, and
writers and found a high incidence of affective disorders in this
population and in their families.
To find a connection between all these different states that
influence creativity, such as relaxation, depression, and resting,
it has been hypothesized that alteration of the brain's neuro-
transmitter systems, primarily a reduction of the catechol-
amine norepinephrine might enhance creative thinking.
As mentioned above, the simultaneous and full activation
of multiple modular networks appear to be important in
creative innovation. Partial support for the postulate that
catecholamines modulate the size of neuronal networks
comes from the priming study of Kischka and coworkers
who used a lexical priming task. As mentioned, either real
words or pseudo words were flashed on a screen and the
participants were asked to press a computer key as rapidly as
possible when they determined that the word on the screen
was a real word, but not to press the key if a pseudo word was
seen. Sometimes these real words and pseudo words were
preceded by another word that is called 'a prime.' If this
preceding word is related to the real target word, the prime
word will help the participants more rapidly recognize the
real target word. The closer the prime and the target word
are associated the more rapid will be the recognition of the
real target word (direct priming). In contrast, if the words
are weakly or distantly related the priming will be less robust
(indirect priming) and the reaction will be only slightly
reduced by these primes. When Kischka et al. administered
levodopa to normal participants, the indirect priming effect
decreased, suggesting that dopamine reduces the spread of
semantic activation induced by the prime. Although Kischka
et al. attributed this effect to the dopaminergic system, levo-
dopa is also precursor of norepinephrine, and the adminis-
tration of levodopa to these individuals may have also
increased the level of norepinephrine.
To study the influence of norepinephrine on cognitive
flexibility, Beversdorf et al. tested the ability of normal
participants to solve anagrams when treated with placebo,
ephedrine, or propanolol. Ephedrine increases the level of
norepinephrine whereas propanolol is a beta-noradrenergic
blocker and thus interferes with the influence of norepineph-
rine in the brain. They found that performance on the ana-
grams task was better after participants took propanolol than
after they took ephedrine.
During stressful conditions, where normally there is an
increase of catecholamines such as norepinephrine, the perfor-
mance on the Remote Association Test declines. Further sup-
port for the postulate that high levels of catecholamine might
reduce associative and convergent reasoning comes from a
study where students with test anxiety improved their score
on the Scholastic Aptitude Test when they took the beta-adren-
ergic blocker propanolol.
Creativity 615
When performing the Controlled Oral Word Association
Test subjects are asked to produce as many words as they can
that begin with a certain letter in one minute. Foster et al.
performed a study examining lexical-semantic activation in a
population of Parkinson's patients. They sought to investigate
the effects of PD, which is associated with dopamine and
norepinephrine depletion, on spreading activation in the
lexical-semantic networks. If these lexical-semantic networks
are constricted, subjects would be more likely to use high-
frequency words and if the network was expanded the subjects
would be more likely to recall low-frequency words. Hence,
Foster et al. examined the average word frequency on the words
produced on the Controlled Oral Word Association Test in
patients with PD and compared their performance to that of
controls. These investigators found that the participants with
PD exhibited a lower average word frequency than did control
subjects, suggesting the subjects with PD had a more expansive
activation of their lexical-semantic networks than did the con-
trol participants.
Ghacibeh et al. performed a study examining the effect of
vagus nerve stimulation versus sham stimulation on creativity
and cognitive flexibility in people affected by medically
intractable partial epilepsy. Their hypothesis was that since
vagus nerve stimulation activates the neurons in the locus
coeruleus (LC), and results in increased release of brain nor-
epinephrine, vagus nerve stimulation should reduce creativity
and cognitive flexibility. These investigators had their patients
perform tests of creativity such as the Torrance Test as well as
cognitive control tests such as the Hopkins Verbal Learning
Test. They found that when compared to sham stimulation,
vagal stimulation reduced creativity but did not impair mem-
ory. These investigators concluded that the reason vagus nerve
stimulation impairs cognitive flexibility and creative thinking
is probably related to increased activity of the LC-central
noradrenergic system that increases the signal-to-noise ratio
and improves the brain's ability to attend to sensory input,
but decreases its ability to recruit large-scaled networks.
Depression is associated with a reduction of brain norepi-
nephrine and based on this literature, it would be reasonable
to think that depression with a decrease in norepinephrine
would facilitate creative innovation; however, as we men-
tioned earlier, the verification and productivity portion of
the creative process do need high level of arousal and thus it
might be that depressed individuals have very creative ideas
but have difficulty producing creative products until their
depression remits.
Summary
Creativity is defined as the ability to understand, develop,
construct, and express in a systematic fashion, novel orderly
relationships. There are three or four major steps in the creative
process, the first step is preparation, the second is innovation,
including incubation and illumination, and the final is pro-
ductivity. There are several means of studying how the brain
allows humans to be creative and these include studies of
creative people and patients with neurological diseases, as
well as physiological studies (including functional imaging
and electrophysiology (EEG and evoked potentials)), cognitive
psychological studies, and pharmacological studies. In regard
to preparation, to be creative a person needs an adequate
level of general intelligence as well as domain-specific knowl-
edge and special skills that are dependent upon the domain
of creativity. In the human brain knowledge is, for the
most part, stored in the temporal, parietal, and occipital asso-
ciation cortex. In the human brain some forms of knowledge
(language) are stored in the dominant left hemisphere whereas
other forms of knowledge (visuospatial) are stored in the
right hemisphere. There is some evidence that either faulty
development or degeneration of one portion of one hemi-
sphere might disinhibit other areas of the brain, particularly
in the other hemisphere, and this disinhibition might enhance
domain-specific creativity. Domain-specific knowledge and
special skills alone are, however, are not sufficient for creativ-
ity. The ability to disengage from accepted solutions and
develop alternative solutions or divergent thinking is an
important aspect of creativity. Clinical as well as functional
imaging studies suggest that the frontal lobes are critical for
divergent thinking. The frontal lobes have strong connections
with the posterior cortical association areas where concepts
and various forms of knowledge are stored. White matter con-
nects the right with the left hemisphere and white matter path-
ways connect the frontal lobes with the more posterior cortical
association areas. These connections might allow the selective
activation and inhibition of portions of posterior neocortex in
both hemispheres and thus be important for developing alter-
native solutions. Although extensive knowledge and divergent
thinking are critical for creative innovation they alone are
insufficient for allowing a person to find 'the thread that uni-
tes.' Finding unity requires the binding of different forms of
knowledge and these different forms of knowledge are often
stored in separate cortical modules. Thus, creative innovation
requires both the coactivation and communication between
regions of the brain and thus is highly dependent on white
matter pathways, both inter- and intrahemispheric. In addi-
tion, creative innovation most often occurs during levels of low
arousal. High arousal is associated with high levels of brain
norepinephrine produced by high rates of LC firing. This high
rate of firing normally allows people to focus on external
stimuli by restricting the breadth of activated concept
representations. Low levels of norepinephrine shift the brain
toward intrinsic neuronal activation with an increase in the
size of distributed concept representations and coactivation
across modular networks, a state that enhances creative inno-
vation. Many people with depression are creative and this
relationship may be related to the alterations of neuro-
transmitters such as norepinephrine that are associated with
depression. In addition to being important in divergent
thinking, the frontal lobes also help control the activity of the
LC-norepinephrine system. Highly creative people may have
brains that are capable of storing extensive specialized knowl-
edge in their posterior association cortex, are capable of frontal
mediated divergent thinking, and have an ability to modulate
the frontal lobe-LC (norepinephrine) system allowing them to
find the thread that unites.
See also: Creative and Imaginative Thinking; Depression; Human
Intelligence.
616 Creativity
Further Reading
Abraham A and Windmann S (2007) Creative cognition: The diverse operations
and the prospect of applying a cognitive neuroscience perspective. Methods
42: 38-48.
Aharon I, Etcoff N, Ariely D, Chabris CF, O’Connor E, and Breiter HC (2001) Beautiful
faces have variable reward value: fMRI and behavioral evidence. Neuron
32: 537-551.
Asari T, Konishi S, Jimura K, Chikazoe J, Nakamura N, and Miyashita Y (2008)
Amygdalar enlargement associated with unique perception. Cortex.
Beals G (1996) In Thomas Edison’s Home Page, www.thomas-edison.com.
Bechtereva NP, Korotkov AD, Pakhomov SV, Roudas MS, Starchenko MG, and
Medvedev SV (2004) PET study of brain maintenance of verbal creative activity.
International Journal of Psychophysiology 53: 11-20.
Benton A and Tranel D (1993) Visuoperceptual, visuospatial, and visuoconstructive
disorders. In: Heilman KM and Valenstein E (eds.) Clinical Neuropsychology,
3rd edn., pp. 165-213. New York: Oxford University Press.
Berg EA (1948) A simple objective technique for measuring flexibility in thinking.
Journal of General Psychology 39: 15-22.
Beversdorf DQ, Hughes JD, Steinberg BA, Lewis LD, and Heilman KM (1999)
Noradrenergic modulation of cognitive flexibility in problem solving. Neuroreport
10(13): 2763-2767.
Breiter HC, Aharon I, Kahneman D, Dale A, and Shizgal P (2001) Functional imaging of
neural responses to expectancy and experience of monetary gains and losses.
Neuron 30: 619-639.
Bronowski J (1972) Science and Human Values. New York: Harper and Row.
Carlsson I, Wendt PE, and Risberg J (2000) On the neurobiology of creativity.
Differences in frontal activity between high and low creative subjects.
Neuropsychologia 38: 873-885.
Catell RB (1963) The theory of fluid and crystallized intelligence: A critical test. Journal
of Educational Psychology 54:1-22.
Chavez-Eakle RA, Graff-Guerrero A, Garcia-Reyna JC, Vaugier V, and Cruz-Fuentes C
(2007) Cerebral blood flow associated with creative performance: A comparative
study. Neuroimage 15: 519-528.
Contreras D and Llinas R (2001) Voltage-sensitive dye imaging of neocortical
spatiotemporal dynamics to afferent activation frequency. Journal of Neuroscience
21(23): 9403-9413.
Csikszentmihalyi M (1996) Creativity: Flow and the Psychology of Discovery and
Invention. New York: Harper Perennial.
Csikszentmihalyi M and Csikszentmihalyi IS (eds.) (1988) Optimal Experience:
Psychological Studies of Flow in Consciousness. Cambridge: Cambridge University
Press.
Dow GT and Mayer RE (2004) Teaching students to solve insight problems.
Evidence for domain specificity in training. Creativity Research Journal
16(4): 389-402.
Drago V, Foster PS, Okun MS, et al. (2009a) Turning off artistic ability: The
influence of left DBS in art production. Journal of Neurological Sciences
281: 116-121.
Drago V, Foster PS, Skidmore FM, and Heilman KM (2009b) Creativity in Parkinson
disease as a function of right versus left hemibody onset. Journal of the
Neurological Sciences 276: 179-183.
Drago V, Foster PS, Trifiletti D, FitzGerald DB, Crucian GP, and Heilman KM (2006)
‘What’s inside the art?’ The influence of fronto-temporal dementia in art production.
Neurology 2006(67): 1285-1287.
Ellenbroek BA, Budde S, and Cools AR (1996) Prepulse inhibition and latent
inhibition: The role of dopamine in the medial prefrontal cortex. Neuroscience
75: 535-542.
Eysenck HJ (1995) Genius: The Natural History of Creativity. New York: Cambridge
University Press.
Faigel HC (1991) The effect of beta blockade on stress-induced cognitive dysfunction in
adolescents. Clinical Pediatrics (Phila) 30: 441-445.
Feist GJ (1999) The influence of personality on artistic and scientific creativity.
In: Strenberg RJ (ed.) Handbook of Creativity, pp. 273-296. Cambridge:
Cambridge University Press.
Fernandez HH and Friedman JH (1999) Punding on L-dopa. Movement Disorders
14: 836-838.
Finney GR and Heilman KM (2007) Artwork before and after onset of progressive
nonfluent aphasia. Cognitive and Behavioral Neurology 20: 7-10.
Flaherty AW (2005) Frontotemporal and dopaminergic control of idea generation and
creative drive. Journal of Comparative Neurology 493: 1 47-1 53.
Foster PS, Drago V, FitzGerald DB, Skoblar BM, Crucian GP, and Heilman KM (2008)
Spreading activation of lexical-semantic networks in Parkinson’s disease.
Neuropsychologia 46: 1908-1914.
Gardner H (1993) Creating Minds: An Anatomy of Creativity Seen Through the Lives of
Freud, Einstein, Picasso, Stravinsky, Eliot, Graham, and Gandhi. New York: Basic.
Ghacibeh GA, Shenker Jl, Shenal B, Uthman BM, and Heilman KM (2006) Effect of
vagus nerve stimulation on creativity and cognitive flexibility. Epilepsy and Behavior
8(4): 720-725.
Ghiselin B (1952) The Creative Process: A Symposium. New York: Mentor.
Goff K and Torrance EP (2002) Abbreviated Torrance Test for Adults Manual.
Bensenville, IL: Scholastic Testing Service.
Guilford JP (1950) Creativity. American Psychologist 5: 444-454.
Guilford JP (1967) Creativity: Yesterday, today, and tomorrow. Journal of Creative
Behavior 1(1): 3-14.
Guilford JP and Christensen PR (1973) The one-way relation between creative potential
and IQ. Journal of Creative Behavior 7(4): 247-252.
Guilford JP, Christensen PR, Merrifield PR, and Wilson RC (1978) Alternate Uses:
Manual of Instructions and Interpretation. Orange, CA: Sheridan Psychological
Services.
Hayes JR (1989) Cognitive processes in creativity. In: Glover JA, Ronning RR, and
Reynolds CR (eds.) Handbook of Creativity, pp. 135-145. New York: Plenum.
Heilman KM (2005) Creativity and the Brain. New York: Psychology Press.
Heilman KM, Nadeau SE, and Beversdorf DO (2003) Creative innovation: Possible brain
mechanisms. Neurocase 9: 369-379.
Helmholtz H (1826) In: Eysenck H and Genius (eds.) Vortrage and reden. Brunschweig :
Vieweg, p. 308. Cambridge: Cambridge University Press.
Howard-Jones PA, Blakemore SJ, Samuel EA, Summers IR, and Claxton G (2005)
Semantic divergence and creative story generation: An fMRI investigation.
Brain Research. Cognitive Brain Research 25: 240-250.
James W (1890) The Principle of Psychology. New York: Dover.
Jausovec N and Jausovec K (2000) Differences in resting EEG related to ability.
Brain Topography 12(3): 229-240.
Kapur N (1996) Paradoxical functional facilitation in brain behavior research: A critical
review. Brain 119: 1775-1790.
Keltikangas-Jarvinen L, Elovainio M, Kivimaki M, Lichtermann D, Ekelund J, and
Peltonen L (2003) Association between the type 4 dopamine receptor gene
polymorphism and novelty seeking. Psychosomatic Medicine 65: 471-476.
Kischka U, Kammer T, Maier S, Weisbrod M, Thimm M, and Spitzer M (1996)
Dopaminergic modulation of semantic network activation. Neuropsychologia
34(11): 1107-1113.
Kraepelin E (1921) Maniac-Depressive Insanity and Paranoia. London: Churchill
Livingstone.
Kuhn TS (1996) The Structure of Scientific Revolution, 3rd edn. Chicago: University of
Chicago Press.
Lewis RT (1979) Organic signs, creativity, and personality characteristics of patients
following cerebral commissurotomy. Clinical Neuropsychology 1(4): 29-33.
Martindale C (1990) The Clockwork Muse: The Predictability of Artistic Change.
New York: Basic.
Martindale C (1999) Biological bases of creativity. In: Stenrnberg RJ (ed.) Handbook of
Creativity, pp. 137-152. New York: Cambridge University Press.
Martindale C and Greenough J (1973) The differential effect of increased arousal on
creative and intellectual performance. Journal of Genetic Psychology 123: 329-335.
Martindale C, Smith SM, Ward TB, and Finke RA (1995) Creativity and connectionism.
In: Smith SM, Ward TB, and Finke RA (eds.) The Creative Cognition Approach,
pp. 249-268. Cambridge, MA: MIT Press.
McCarley RW (1982) REM sleep and depression: Common neurobiological control
mechanisms. American Journal of Psychiatry 139: 565-570.
Mednick S (1962) The associative basis of the creative process. Psychological Review
69(3): 220-232.
Miller BL, Boone K, Cummings J, Read S, and Mishkin F (2000) Functional correlates
of musical and visual ability in frontotemporal dementia. British Journal of
Psychiatry MG: 458-463.
Miller BL, Cummings J, Mishkin F, et al. (1998) Emergence of artistic talent in
frontotemporal dementia. Neurology 5: 978-982.
Miller BL, Ponton M, Benson JL, Cummings J, and Mena I (1996) Enhanced artistic
creativity with temporal lobe degeneration. Lancet 348: 1744-1745.
Milner B (1984) Behavioural effects of frontal-lobe lesions in man. Trends in
Neurosciences 1: 403-407.
Milner B and Petrides M (1984) Behavioural effects of frontal-lobe lesions in man.
Trends in Neurosciences 7(11): 403-407.
Plucker JA and Renzulli JS (1999) Psychometric approaches to the study of human
creativity. In: Sternberg JR (ed.) Handbook of Creativity, pp. 35-61. Cambridge:
Cambridge University Press.
Robertson LC, Lamb MR, and Knight RT (1988) Effects of lesions of temporal-parietal
junction on perceptual and attentional processing in humans. Journal of
Neuroscience 8(10): 3757-3769.
Creativity 617
Rorschach H (1927) Rorschach Test - Psychodiagnostic Plates. Cambridge,
MA: Hogrefe Publishing Corp.
Rumelhart DE, Smolensky P, McClelland JL, and Hinton GE (1986) Schemata and
sequential thought processes in PDP models. In: McClelland JL, Rumelhart DE, and
T.P. Research Group (eds.) Parallel Distributed Processing, vol. 2, pp. 7-57.
Cambridge, MA: MIT Press.
Savitz JB and Ramesar RS (2004) Genetic variants implicated in personality: A review of
the more promising candidates. American Journal of Medical Genetic. PartB
Neuropsychiatric Genetic 15: 20-32.
Simonton DK (1994) Greatness: Who Makes History and Why. New York: Guilford
Press.
Simonton DK (1999) Origins of Genius: Darwinian Perspectives on Creativity. Oxford:
New York.
Snyder A, Bossomaier T, and Mitchell DJ (2004) Concept formation: ‘Object’ attributes
dynamically inhibited from conscious awareness. Journal of Integrative
Neuroscience 3: 31-46.
Sternberg RJ and O’Hara LA (1999) Creativity and intelligence. In: Sternberg RJ (ed.)
Handbook of Creativity, pp. 251-272. New York: Cambridge University Press.
Strick M, Dijksterhuis A, and van Baaren RB (2010) Unconscious-thought effects take
place off-line, not on-line. Psychological Science 21(4): 484-488.
Torrance EP (1974) The Torrance Test of Creative Thinking. Bensenville, IL: Scholastic
Testing Service.
Wallas G (1926) The Art of Thought. London: Cape.
Wechsler D (1981/2010) WAiS-R Manual. New York: Psychological Corporation.
Psychological Science 21(4), 484-488.
Weisberg R (1 986) Creativity: Genius and Other Myths. New York: W.H. Freeman.
Weisberg RW (1993) Creativity: Beyond the Myth of Genius. New York: W.H. Freeman.
Weisberg RW (1994) Genius and madness? A quasi-experimental test of the
hypothesis that manc-depression increases creativity. Psychological Science
5: 361-367.
Zangwell 0L (1966) Psychological deficits associated with frontal lobe lesions.
International Journal of Neurology 5: 395-402.
Crisis Management
A M Legg and K Sweeny, University of California, Riverside, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Crisis Any negative event or situation that a person or
organization perceives as such.
Crisis decision theory A framework developed by Kate
Sweeny in which that describes the decision-making
processes people undergo when faced with a crisis.
Crisis management The process a person or organization
uses to deal with a crisis. The process includes preparation
and prevention processes, detection of crisis, responding,
and postcrisis responding.
Crisis signal detection The process through which people
and organizations recognize or predict crises.
Damage containment The process through which people
and organizations attempt to minimize the damage
associated with a crisis.
Proactive coping A set of behaviors including acquiring
resources, recognizing potential negative events, and
taking action before an event occurs that prevents the
negative event from happening or changes its course
or outcome.
Introduction
Crises arise in many forms. For example, a company suffers a
privacy breach in which their customers' financial information
is accessed by outside parties. A woman realizes that her car is
no longer in its parking space and wonders if someone stole
or towed the car. A teacher overhears a student threatening to
bring a gun to school. How do people manage crises such as
these? This article focuses on the decision-making processes
associated with managing crises in a variety of circumstances.
Although much of the crisis management literature focuses on
organizational behavior, this article also devotes substantial
attention to the way individuals manage personal crises and
draws connections between crisis management in individuals
and organizations.
Definition of Crisis
The definition of a crisis depends in part on whether it is a crisis
for an individual or an organization. The business and man-
agement literature typically defines a crisis as an event that
damages a company's financial situation or reputation. In the
case of medical research on crisis management, the definition
of a crisis is any potentially life-threatening situation such as
anaphylaxis or cardiac arrest. Further, education researchers
define a crisis as any event that threatens the safety, and well-
being of the students, or faculty, and staff or the students'
ability to learn. However, when focusing on crises occurring
at the individual or small group level, the definition becomes
much broader and more subjective and includes any negative
event or situation that a person perceives to be a crisis. The
severity of individual crises can vary and do not necessarily
threaten the person's life, safety, or well-being, or impact finan-
cial stability or social reputation. In other words, a crisis is, to a
large extent, is in the eye of the beholder.
A variety of crisis management models put forth in the past
few decades address a wide range of factors associated with
preparing for, preventing, and responding to crises. Despite
the diverse research areas from which these models derive
(e.g., health, education, social psychology, and organizational
behavior), they include many common and overlapping fea-
tures. This article provides an overview of the full process of
crisis management, including the detection of crises, the prep-
aration for and prevention of crises, the decision-making pro-
cesses engaged in crisis response, and recovery from crises. The
goal of this article is to draw parallels between the various crisis
management models and to merge previously disconnected
areas in which crisis management has been studied.
Detection of Potential Crises
A significant segment of the crisis management literature
comes from the fields of management and business and
focuses on early detection of potential crises. Signal detection
and issues management, two frequently used terms that refer to
this stage of crisis management, involve information gathering,
monitoring, and scanning of the environment in an attempt
to predict potential crises. Additionally, Lisa Aspinwall and
Shelley Taylor's proactive coping framework provides a social
psychological conceptualization of how individuals acquire
resources and attempt to recognize stressors that could poten-
tially lead to a crisis.
Regardless of whether a crisis affects an individual or orga-
nization, four types of information can signal an impending
crisis, according to frameworks described by Ian Mitroff and
Gus Anagnos. Signals can be internal or external and derive
from technological sources or from people. Internal technologi-
cal signals monitor internal operations, such as virus scans on a
computer. External technological signals can come from outside
organizations such as government agencies or weather moni-
toring institutions, and they typically alert people to impend-
ing natural disasters or other external crises. Internal people
signals involve information provided by personnel within an
organization but may also refer to an individual's thoughts and
emotions. For instance, a woman who finds a lump in her
breast would likely respond to her awareness of the problem
(i.e., internal information) by making an appointment to get
a mammogram. Finally, external people signals come from
618
Crisis Management 619
community members, customers, or friends and family. For
example, a friend may communicate news of the death of a
mutual friend, or community members may alert their city
government about increased gang activity in a part of their city.
Even when signals exist to cue individuals or organizations
to the potential for a crisis, the question remains as to whether
they will seek out and acknowledge this information. In fact,
research finds that although some people seek out information
in the face of a potential crisis, many others prefer to avoid such
information. At the individual level, information seeking in the
face of a crisis involves activities like reading about symptoms,
prognosis, and treatment for a medical illness or comparing
home insurance policies to gain the best coverage. Businesses,
too, can seek out information to aid in their prediction of
potential crises. For instance, they may collect stock market
trend data to foretell financial woes or examine employee satis-
faction surveys to anticipate potential employee discontent.
Information avoidance, on the other hand, is a defensive
strategy to avoid acquiring unwanted or aversive information.
For example, a large percentage (up to 55%) of people tested
for HIV do not return to the testing clinic to receive their
results. In such cases, choosing to avoid information may
postpone the onset of a potential crisis (e.g., facing a diagnosis
of HIV). Although attending to information that potentially
predicts an impending crisis is typically quite beneficial, indi-
viduals and organizations may ignore accessible information
for a variety of reasons. That is, people are most likely to avoid
information when it could lead to an unwanted or difficult
change in their attitudes, beliefs, or actions, or when the infor-
mation will induce aversive or negative emotions. Similarly,
people also avoid crisis-provoking information if they perceive
that they have little control over the outcomes of the informa-
tion, if they lack sufficient coping resources to deal with the
information, or if the information is difficult to access or hard
to interpret.
As an illustration of the potential consequences of infor-
mation avoidance, the crisis associated with the Space
Shuttle Challenger disaster provides a prime example of cri-
sis-predicting information avoidance. The conclusions of the
investigating committee, the Rogers Commission, determined
that there was a plethora of available information, namely
indicators of defective parts on the shuttle and expert warnings
against launching the shuttle, that NASA officials chose to
ignore and that ultimately contributed to the occurrence of
the tragedy. Although information avoidance may protect peo-
ple from unpleasant feelings of anxiety and fear, the conse-
quences in the context of crisis management can be dire.
Even if individuals or organizations are receptive to infor-
mation that might indicate an impending crisis, they may face
another barrier to crisis detection. Ian Mitroff notes that
a psychological bias can occur in organizations such that they
perceive an unrealistically low risk of experiencing a crisis.
Organizations with this perspective tend to take fewer prepara-
tory and preventative actions and thus, ironically, become
more crisis prone. Similarly, individuals who are unrealistically
optimistic about their likelihood of, for example, facing a
health crisis also tend to engage in fewer preventative health
actions. Thus, a critical aspect of crisis management is over-
coming the biases that prevent individuals and organizations
from realistically assessing their risk of experiencing a crisis.
Preparation and Prevention of Crises
Effective crisis management requires preparatory and preven-
tative actions. A number of models include preparation and
prevention as key components of crisis management, and
many business and management scholars argue that the prepa-
ratory and prevention stage is the most critical for good crisis
management. The goals of the preparation and prevention
stage of crisis management are preparing oneself and others
for a potential crisis and developing and implementing contin-
gency plans for crisis responding.
Even when individuals and organizations recognize the
necessity of preventative action, they often have limited
resources to prepare for all possible crises. Mitroff and Anagnos'
business frameworks of crisis management typically distinguish
different categories of crises. The seven most frequently dis-
cussed crises are economic (e.g., labor strikes or market
crashes), informational (e.g., loss of customer information or
other confidential documentation), physical (e.g., loss of mate-
rials or facilities), human resource related (e.g., loss of staff or
decrease in productivity), reputational (e.g., slander or gossip),
psychopathic acts (e.g., violence, kidnapping, or terrorism),
and natural disasters. The developers of such crisis frameworks
recommend that organizations prepare for at least one crisis
in each category.
Preparation for crises often entails simulation and training
experiences. For example, in the case of hurricane preparation
an organization might establish a meeting destination in case
of evacuation, determine how members will communicate
with each other, and post information about where emergency
supplies are located. Similarly, schools conduct fire, earth-
quake, and tornado drills in order to prepare faculty, staff,
and students to respond to natural disasters. Other preventive
actions include implementing strong security systems, sched-
uling, planning and training workshops, engaging in mental
health counseling, and encouraging activities that promote
well-being, among others.
Unfortunately, individuals and organizations at times do
not engage in preparatory or preventative actions even when
they detect the possibility of a crisis. For example, women often
fail to get recommended mammograms to proactively scan for
breast cancer. Similarly, prior to the 2007 Virginia Tech cam-
pus shooting many college campuses failed to invest resources
in preventative programs focusing on college student mental
health and response plans for violence in the school. Similar to
information avoidance, organizations engage in a number of
defense mechanisms that hinder proper proactive responding.
Mitroff and Anagnos point out several of the typically used
defense mechanisms of crisis-prone organizations. For exam-
ple, both individuals and organizations can fall victim to the
'just world' thinking by believing that good people/organiza-
tions do not face crises, or that all people/organizations in
crisis must have done something to deserve their fate. Simi-
larly, organizations often have a sense of perceived grandiosity,
or the feeling that they are immune to crises. Both the 'just
world' thinking and grandiosity can lead to denial in the face of
crises, which undermines the motivation to prepare and
respond effectively. Other defense mechanisms that can under-
mine the motivation to prepare for a crisis include disavowal
620 Crisis Management
(i.e., thinking that a crisis will not significantly impact the
organization), projection (i.e., thoughts about how evil people
or organizations want to hurt the targeted organization), intel-
lectualization (i.e., believing that the chances of a crisis are
statistically small), and compartmentalizadon (i.e., thinking
that a crisis can only affect a small part of the organization
and not the entire company).
When a Crisis Happens
Despite the best efforts to detect and prepare for crises, they
may nonetheless occur, and often unexpectedly. When a crisis
occurs, people must then decide how best to respond. Kate
Sweeny proposed the crisis decision theory to conceptualize the
decision-making process that individuals and organizations
use when dealing with a crisis. The following sections of
this article explore how people make choices when responding
to a crisis. Specifically, crisis decision theory proposes three
stages of responding: assessment of severity, determination
of response options, and evaluation of response options.
Although crisis decision theory primarily describes individual's
decision-making processes in the face of a crisis, the theory is
broad enough to encompass the crisis response options that
confront business, medical, and education organizations.
Assessment of Severity
As mentioned at the outset of this article, a crisis can be defined
as any negative event that a person or organization perceives as
a crisis. This broad definition can encompass such a wide
variety of crises that the severity of crises also varies widely.
Crisis decision theory describes that three types of information
inform perceptions of a crisis's severity: information about the
causes of a crisis, information about how the crisis compares to
other outcomes, and information about the potential conse-
quences of the crisis.
Crisis decision theory further proposes that the severity of a
given crisis affects the extent to which people invest time and
effort in responding to the crisis. According to the theory,
responding occurs in an inverted U-shaped pattern such that
nonsevere crises and extremely severe crises both result in low
response effort. For example, one employee stealing a single pen
from the office supply closet (a nonsevere event) is not likely to
warrant a response from the management. At the other extreme,
individuals or organizations can become paralyzed by very
severe crises. A diagnosis of a severe and incurable disease may
be so traumatic for the patient that s/he will not seek out ways to
cope with or respond to the crisis. Similarly, an organization
facing a financial crisis that could lead to the downfall of the
entire organization may be unable to face the implications of
the crisis, and thus unable to respond. Accordingly, moderately
severe crises are most likely to prompt an effective response.
Determination of Response Options
Some crises may lack the severity necessary to prompt a response.
However, when an individual or organization deems an event
severe enough to warrant a response, they then turn to the second
stage of crisis decision theory: determining available response
options. At this stage, people focus on two components of the
crisis: the controllability of the consequences of the crisis and
the feasibility of the available response options. Turning first
to controllability, a crisis may have short- or long-term conse-
quences that are controllable or uncontrollable. For example,
the company responsible for the Gulf oil spill in 2010 faced
some consequences that they could mitigate or control (e.g., the
financial impact of the crisis on industries in the region) and
other consequences over which they had little control (e.g., the
effectiveness of various attempts to stop the oil from spilling
into the Gulf) . Crisis decision theory proposes that to the extent
that individuals or organizations evaluate the consequences of
a crisis as controllable, they will consider more active response
options (i.e., responses that tackle the crisis itself). If the con-
sequences of a crisis are primarily uncontrollable, individuals
and organizations are left with few response options other
than protecting themselves from the full impact of the crisis.
In addition, the extent of available resources limits crisis
response options by making many options unfeasible. If indi-
viduals or organizations were effective at the preparation and
prevention stage of crisis management, they are likely to have
more response options available once a crisis occurs. For exam-
ple, a school that employs security guards on campus has more
response options in the face of a violent incident than does a
school without extra security resources. Likewise, an elderly
person who gathers a social support network in advance of
the diagnosis of a terminal illness will have more options when
facing the illness than will a person who is not prepared for
the diagnosis.
Evaluation of Response Options
Having determined the available options for responding to a
crisis, individuals and organizations must then evaluate the
potential effectiveness of their various response options. This
stage of crisis management is essentially a cost-benefit analysis
of the resources required to engage in each response and the
consequences associated with each response, including both
direct and indirect consequences. All else being equal, people
typically prefer the response option that requires the fewest
resources (e.g., time, money, effort, etc.).
However, people must also consider the consequences of
each response option. Direct consequences are outcomes of a
response that affect the status of the crisis (i.e., make the crisis
better or worse). Crisis decision theory describes three impor-
tant aspects of direct consequences: the likelihood that the
consequences of a response will be positive and avoid making
the crisis worse, the magnitude of the potential consequences,
and whether the consequences are reversible.
Indirect consequences of responses to a crisis are any out-
comes of the response that do not affect the status of the crisis.
Indirect consequences include public image consequences, con-
sequences in other areas of life or infrastructure, and conse-
quences for other people or organizations. Public image
consequences are especially important to organizations facing
public crises. It is at this stage in crisis management that busi-
nesses and other organizations must decide whether to commu-
nicate with the media about their current crisis. Even when the
media is already aware of the crisis, organizations must deter-
mine how their response to the crisis will affect the media's
Crisis Management 621
coverage of the crisis and ultimately their public image. Simi-
larly, social psychology reveals that individuals constantly mon-
itor the impressions they make on others, and these impression
management concerns come into play when people evaluate
their response options in the face of a crisis. People and organi-
zations also evaluate how various responses to a crisis are likely
to affect other aspects of their lives or other parts of their organi-
zation. For example, choosing to lay off a number of low-level
employees may mean that senior-level employees work more
hours, take on more responsibilities, and carry a heavier work
burden. In addition, responses to a crisis can have far-reaching
consequences for people other than the individual or organiza-
tion facing the crisis. For example, an organization's response to
a crisis could affect employees and their families, customers,
suppliers, and stockholders.
Selection of a Response
Finally, individuals and organizations facing a crisis ultimately
must choose a response. Although the chosen response typi-
cally reflects the cost-benefit evaluation just described, people
may also prefer responses that they used in the past, regardless
of their prior efficacy. Put simply, the known is often more
comfortable than the unknown, even if the familiar response is
imperfect from a rational, cost-benefit perspective.
In addition to selecting an appropriate response to the
crisis, individuals and organizations may also benefit from
engaging in damage containment. The goal of damage contain-
ment is to minimize the impact of the crisis and prevent the
situation from negatively affecting other areas of an indivi-
dual's life, or other departments, procedures, and individuals
within an organization. For example, damage containment
played an important role in the reaction of the company
responsible for the 2010 Gulf oil spill, despite the fact that
the company was heavily criticized for not engaging in suffi-
cient damage containment at the onset of the crisis. In the
months following the onset of the oil spill, the company
engaged in damage containment by circulating press releases
and producing television commercials that attempted to repair
the company's reputation. In addition, they implemented a
plan to provide monetary aid for people suffering financial
loss due to the spill. Despite these efforts, critics still argued
that the delayed response severely harmed the organization's
reputation as well as the financial stability and well-being
of individuals and businesses impacted by the oil spill. For
instance, environmentalist groups accused the company of
being responsible for withholding information regarding the
severity of the oil spill that potentially slowed the response
action to help save wildlife in the Gulf region. To reiterate, the
goal of damage containment is to minimize the crisis's impact,
and thus BP's efficacy in achieving that goal is perhaps
debatable.
Postcrisis Responding
Fortunately, most crises eventually resolve and come to an
end. However, crisis management may not cease with the end
of the crisis. In fact, many crisis management models include
the recovery period as an important phase in the experience
of a crisis.
Recovering from a crisis involves numerous strategies
including replenishing resources, attempts to recover losses,
and revising crisis management programs. In addition, people
must deal with the emotional consequences of experiencing a
crisis. For example, after traumatic school shootings such as the
Virginia Tech massacre or the Columbine shootings, schools
typically hire counselors to talk with students and faculty about
their reactions to the crisis.
Can individuals and organizations truly and fully recover
from a crisis? Research from the health domain suggests that
they can. Although cancer patients report lower quality of life
during the acute phases of their illness, cancer patients in
remission report similar quality of life as that of cancer-free
individuals. Similarly, some schools and businesses report
greater cohesion and a stronger sense of community between
students, faculty, and/or employees following a crisis. This
enhanced sense of well-being and connectedness can serve to
replenish resources such as energy and mental health that may
have been depleted during the crisis.
Reflecting on and learning from the crisis are critical pro-
cesses to prepare for future crises. Although reflecting on a
crisis may provoke anxiety, especially in the case of severe or
traumatic crises, reflection may also promote renewed confi-
dence and feelings of self-efficacy regarding the ability to han-
dle future crises. Furthermore, people's perceptions of the
causes and implications of a crisis may change once the crisis
is over. It may be difficult to evaluate all aspects of a crisis while
in the midst of it, so reevaluating the crisis once it ends can
provide insight into the most effective ways to prevent future
crises and to mitigate the consequences of crises that are
unavoidable.
See also: Expectation; Leadership; Organizational Behavior; Perceived
Control; Planning; Problem Solving; Uncertainty.
Further Reading
Aspinwall LG and Taylor SE (1997) A stitch in time: Self-regulation and proactive
coping. Psychological Bulletin 121(3): 417-436.
Comfort LK (2007) Crisis management in hindsight: Cognition, communication,
coordination, and control. Public Administration Review 67(Supplement 1 ,
Special Issue): 189-197.
Egelhoff WG and Sen F (1992) An information-processing model of crisis management.
Management Communication Quarterly 5(4): 443-484.
Elsubbaugh S, Fildes R, and Rose MB (2004) Preparation for crisis management:
A proposed model and empirical evidence. Journal of Contingencies and Crisis
Management 12(3): 112-127.
Gonzalez-Herrero A and Pratt CB (1995) How to manage a crisis before - or
whenever - it hits. Public Relations Quarterly 40(1): 25-29.
Mitroff I (2005) Why Some Companies Emerge Stronger and Better from a
Crisis: 7 Essential Lessons tor Surviving Disaster. New York, NY:
AMAC0M.
Mitroff II and Anagnos G (2000) Managing Crises Before They Happen: What
Every Executive and Manager Needs to Know About Crisis Management. New York,
NY: AMAC0M.
Mitroff II, Shrivastava P, and Udwadia FE (1987) Effective crisis management.
The Academy of Management Executive 1(4): 283-292.
Pauchant TC and Mitroff II (1992) Transforming the Crisis-Prone Organization:
Preventing Individual, Organizational, and Environmental Tragedies. San Francisco,
CA: Jossey-Bass.
Pearson CM and Clair JA (1998) Reframing crisis management. The Academy of
Management Review 23(1): 59-76.
622 Crisis Management
Pearson CM and Mitroff II (1993) From crisis prone to crisis prepared:
A framework for crisis management. The Executive 7(1): 48-59.
Rock ML (2000) Effective crisis management planning: Creating
a collaborative framework. Education and Treatment ot Children
23(3): 248-264.
Runciman WB, Kluger MT, Morris RW, Paix AD, Watterson LM, and
Webb RK (2005) Crisis management during anaesthesia: The development
of an anaesthetic crisis management manual. Quality and Safety in Health Care
14(1): 1-12.
Sweeny K (2008) Crisis decision theory: Decisions in the face of negative events.
Psychological Bulletin 134(1): 61-76.
Sweeny K, Melnyk D, Malone W, and Shepperd JA (2010) Information
avoidance: Who, what, when and why. Review ot General Psychology
14(4): 340-353.
Cross-Cultural Adaptation
Y Y Kim, University of Oklahoma, Norman, OK, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Acculturation The process of acquiring a new culture's
symbols and practices by an individual whose prior cultural
learning has taken place elsewhere.
Adaptation The process of internal modification of
individuals that enables them to attain a relatively stable,
reciprocal, and functional relationship with a particular
environment.
Assimilation The absorption process of foreign-born
individuals into the native population by cultural, social,
and economic mainstreaming.
Communication The process by which information is
exchanged between individuals through the encoding and
decoding activities employing verbal and nonverbal
symbols, signs, or behavior.
Culture shock Psychological and physiological stress
responses to being placed in an unfamiliar cultural
environment.
Deculturation The process by which individuals at least
temporarily suspend, or unlearn over time, some of the
symbols and habitual practices of one's native culture and
replace them with new symbols and practices of the host
culture.
Enculturation The primary socialization process
whereby individuals learn and internalize their native
culture.
Background
Millions of immigrants and refugees change homes each year
seeking better hopes of safety, freedom, or economic better-
ment, or simply looking for a more desirable environment in
which to live. Many others temporarily relocate in a foreign
land to carry out religious activities, or to serve as diplomats,
military personnel, and other governmental, intergovern-
mental, and nongovernmental agency employees on overseas
assignments. Researchers, professors, and students visit and
study at foreign academic institutions, along with the employ-
ees of multinational corporations. Increasingly, individual
accountants, teachers, construction workers, athletes, artists,
musicians, and writers seek employment in foreign lands on
their own. Although varied in individual circumstances, all of
these groups of people and many others share the common
project called cross-cultural adaptation. They find themselves
straddled between two worlds, the familiar milieu of the home
culture and their new locus in the host society, and recognize
that many of their previously held beliefs, taken-for-granted
assumptions, and routinized behaviors no longer relevant or
effectual.
Long-Term Adaptation
The process of cross-cultural adaptation among long-term
settlers has been one of the most salient research issues
across social science disciplines. Studies addressing this phe-
nomenon have been continuous and extensive, particularly
in the United States and more recently in European countries.
In the 1930s, the Social Science Research Council appointed
a Subcommittee on Acculturation and charged it with the task
of analyzing and defining the parameters for this new field
of inquiry within the domain of cultural anthropology. As a
result, the concept, acculturation, was formally adopted as a
new area of study dealing with those phenomena that result
when individuals and groups from one culture come into
direct contact with another culture.
Subsequent academic inquiry into cross-cultural adaptation
has evolved into a vast and diverse field. Such has been the case
due to the application of concepts, definitions, and methodol-
ogies pertaining to differing disciplinary and individual per-
spectives and research foci. Many sociologists, for example,
have approached immigrant adaptation focusing on macrole-
vel issues pertaining to the patterns and processes in which
minority groups are integrated into the political, social, and
economic structure of the new, or host, society. Studies in
urban sociology and urban anthropology have added the
dimension of ethnographic explorations into the issues related
to political, social, and economic dynamics in immigrant or
ethnic communities and neighborhoods.
In comparison, studies in social psychology, sociolinguis-
tics, and communication have concentrated on the microlevel
phenomenon of long-term adaptive change in individual
immigrants. In this approach, emphases have been placed on
the psychological, social, and cultural integration of individual
newcomers into their host society. By and large, such studies
have revealed that an individual's adaptive change is cumula-
tive and progressive over time, thereby affirming the traditional
'melting-pot' perspective on the history of immigration in the
United States.
The linear-progressive understanding of long-term adapta-
tion, however, has been increasingly challenged since the
1960s and 1970s. With the rise of the ideology of pluralism in
the United States and elsewhere, pluralistic conceptions of long-
term adaptation have been proposed to reflect the viewpoint that
assimilation into the dominant mainstream culture on the part
of immigrants and ethnic minorities need not, or should not,
take place. Instead, scholars upholding or sympathetic to this
view have emphasized alternative choices for individuals or
groups such as maintaining a distinct identity or seeking a dual
identity with respect to their original culture and the host society.
623
624 Cross-Cultural Adaptation
Short-Term Adaptation
Studies of temporary sojourners have been largely exempt
from the above-described ideological controversy over the
phenomenon of assimilation. Since the 1960s, studies of
sojourner adaptation have been motivated mainly by practical,
problem-centered concerns of easing the temporary but often
difficult transition into a new cultural environment. The
problem-centered perspective is evident in the many studies
that have addressed the phenomenon of culture shock. This
concept, as well as related terms such as transition shock,
cultural fatigue, and self-shock have been employed to investi-
gate various stress reactions to cultural dislocation, from irrita-
bility, insomnia, and an acute sense of loss, to feelings of
anxiety, insecurity, and impotence.
While focused on the problematic nature of life in a foreign
land, sojourn research findings have shown discernible pat-
terns of change in psychological states over time. A variety of
studies have reported a 'U-curve' of psychological adaptation
during a sojourn, depicting the general tendency for sojourners
to experience initial optimism and elation, the subsequent
emotional dip or trough, followed by a gradual recovery to
higher adaptation levels. This U-curve description has been
extended to the 'W-curve,' that includes the reentry phase
during which the sojourner, once again, undergo a similar
psychological adaptation process upon returning home. De-
spite their intuitive appeal, the U-curve and W-curve hypoth-
eses have produced inconsistent findings when applied to
different research contexts. Comprehensive reviews of pertinent
research have concluded that evidence for the theories' claims
are inconclusive and overgeneralized.
Sojourner adaptation studies have also highlighted some of
the positive and beneficial aspects of crossing cultures. Argu-
ments have been made that the cultural shock experience must
be viewed in a broader context of profound learning and
growth that leads to an increased self-awareness. Highlighting
such positive aspects of sojourn, some studies have shown that
the intensity and magnitude of initial culture shock is either
unrelated to, or even positively related to, the level of sub-
sequent psychological adjustment and social and professional
effectiveness within the host environment.
The Process of Cross-Cultural Adaptation
Clearly, there are differences between long-term and short-
term adaptation associated with the duration of the adaptation
process itself. Unlike immigrants, most short-term sojourners
are more likely to limit their contact with the new culture to
mostly peripheral areas, such as to pursue a career, obtain a
degree, or merely enhance one's prestige in the eyes of the folks
at home. Compared to long-term settlers, their commitment to
the host environment is likely to be weaker given their more
specific and narrowly defined goals for sojourn.
Differences such as these, however, are largely a matter of
degree, and not of kind. That is, immigrants and sojourners
alike undergo at least some level of the common project of
cross-cultural adaptation as they begin their life in the host
society as strangers. As strangers, all newcomers embark on a
cross-cultural journey as outsiders. They are challenged to
undergo at least some level of modifications in their original
cultural habits, at least to the extent that is called for by the
specific circumstance under which they have entered the host
culture, so as to navigate their unfamiliar milieu and carry out
their daily life activities.
This notion of strangers can be extended to include those
who return home after a sojourn in a foreign culture or subcul-
ture. Generally, readapting to one's original culture is less
demanding than adapting to a foreign culture. Yet, to the extent
that the returnee has been changed by the sojourn experience,
and to the extent that the original culture has changed during
the sojourn, the returnee, once again, goes through essentially
the same process of cross-cultural adaptation. Strangers further
include those native-born individuals who cross subcultural or
ethnic boundaries within a society and face significant adaptive
pressures from the new subcultural environment. Such would
be the case, for example, for American Indians leaving an
Indian reservation to find employment in an urban environ-
ment. Also included in the domain of strangers are those indi-
viduals who find their local community rapidly changing in
its demographic composition. For example, existing African
American inner city residents may find themselves changing
from being a majority group to becoming a minority group
due to a significant influx of Asian immigrants.
The following discussion addresses the essential nature
of the adaptation process that applies to all of the above-
described categories of strangers who share the common chal-
lenge of the practical necessity to bridge the gap between their
internalized cultural habits and those of the new environment.
Enculturation, Acculturation, Deculturation, Assimilation
The unwritten task of every culture is to organize, integrate, and
maintain the home world of the individual primarily in the
formative years of childhood. Children grow up in their native
culture learning what they need to know to function acceptably
in a given culture. Through continuous interaction with the
various aspects of the environment, they are socialized to live
in the company of others around them who share a similar
image of reality and self. This process is commonly called
enculturation.
In many ways, entering a new culture is like starting an
enculturation process all over again. Only this time, strangers
are faced with situations that deviate from the familiar and
internalized original cultural script. They become more aware
of the previously taken-for-granted habits of mind. Now, as
strangers, they discover that they lack a sufficient level of
understanding of the new communication system of the host
society. They are challenged to suspend, or even abandon,
some of the cultural patterns that have symbolized who they
are and what they are. Such inner conflicts, in turn, make
strangers susceptible to external influence and compel them
to learn the new cultural system. This activity of new learning is
the essence of the process known as acculturation, that is, the
learning and acquisition of the new cultural patterns and prac-
tices, particularly in those areas of direct relevance to their
everyday living.
Acculturation, however, is not a process in which new
cultural elements are simply added to prior internal condi-
tions. As new learning accumulates, some deculturation also
Cross-Cultural Adaptation 625
takes place, at least in the sense that new cultural responses are
adopted in situations that previously would have evoked old
ones. As the interplay of acculturation and deculturation con-
tinues for a prolonged period, strangers undergo an internal
transformation, from visible changes in superficial areas such
as overt role behavior to more intangible but more profound
changes such as changes in previously held cultural beliefs,
values, outlooks on life, and even cultural identity.
The ultimate directionality of the internal change driven by
the continual processes of acculturation and deculturation is
toward assimilation, a state of the maximum possible conver-
gence of our internal and external conditions to those of the
natives. Whether by choice or by circumstance, individual
settlers vary in the distance that they travel in their own adap-
tation process. For most strangers, assimilation remains a life-
time goal than an obtainable outcome, one that often requires
the efforts of multiple generations.
Stress, Adaptation, and Growth
Acculturation and deculturation processes inevitably produce
in individual strangers stress, the experience of temporary inter-
nal disturbance and even breakdown in extreme cases. As parts
of their internal organization undergo adaptive modifications,
strangers are temporarily in a state of disequilibrium mani-
fested in varying levels of confusion and anxiety. The state of
internal flux is often met by the tendency to use various defense
mechanisms such as denial, hostility, cynicism, avoidance, and
withdrawal, all of which are particularly acute during the initial
phase of sojourn or immigration, as indicated by the difficulties
and disruptions documented in culture shock studies.
At the same time, however, stress experiences are the very
force that drives strangers to adaptation. The interplay of
stress and adaptation can be understood as a dialectic be-
tween integration and disintegration, between progression
and regression, or between permanence and change. Each stress
experience presents strangers with an opportunity to recreate
themselves. It is through the impetus of stress that they are
compelled to engage in activities of new learning that enables
them to handle the transactions of daily living. Stress, in this
regard, is intrinsic to complex open systems and essential in the
adaptation process - one that allows for self-(re)organization
and self-renewal. The state of misfit and a heightened awareness
in the state of stress serve as the very forces that propel indivi-
duals to overcome the predicament and partake in the act of
adaptation through active new learning.
Out of these activities, some aspects of the environment
may be incorporated into an individual's internal structure,
gradually increasing its overall fitness to the external realities.
Over time, most strangers manage to achieve an increasing
capacity to detect similarities and differences between the
new surrounding and the home culture and are better able to
manage their changed circumstances. What follows a success-
ful, long-term, and cumulative management of the stress-
adaptation disequilibrium is a subtle and often imperceptible
psychological growth, a form of internal change in the direc-
tion of increased perceptual and intellectual complexity with
respect to the host culture.
Together, stress, adaptation, and growth constitute a three-
pronged conceptual representation of the psychological
underpinning of the cross-cultural adaptation process. As
depicted in Figure 1, the stress-adaptation-growth process
continues as long as there are new environmental challenges,
with the overall forward and upward movement in the direc-
tion of greater adaptation and growth. The process does not
unfold in a smooth, arrow-like linear progression, but in a
cyclic and fluctuating pattern of drawback to leap: each stress-
ful experience is responded to with a temporary setback,
which, in turn, activates adaptive energy to reorganize and
reengage in the activities of cultural learning and internal
change, bringing about a new self-reintegration. Large and
sudden changes occur during the initial phase when the sever-
ity of difficulties and disruptions are likely to be high. Over a
prolonged period of undergoing internal change, the fluctua-
tions of stress and adaptation are likely to become less intense,
leading to an overall calming in the individual experiences of
interacting with the host environment.
Factors Influencing the Process
We now turn to the question of differential adaptation rates, or
speeds, at which cross-cultural adaptation occurs in individual
cases. Although strangers, by and large, manage cross-cultural
challenges successfully over time, the degree and pace of their
individual adaptive change vary. Some may suffer more from
hostility or indifference among the natives that makes it diffi-
cult to find ways to overcome the challenges they face in the
host environment. Some may resist the idea of having to
change their original cultural habits more, thereby raising psy-
chological barriers to work against their own adaptive change.
Others may be less capable of dealing with states of adaptive
stress, leading to a more extreme withdrawal and alienation
from the host society or a decision to return to their home
country prematurely.
The fact that individual cases of cross-cultural adaptation
vary in terms of the level of success begs the question, 'What
factors facilitate or impede the adaptation process?'
This question is addressed below through an examination
of the key factors of the communicative interface between the
individual and the host environment, of the host environment,
and of the individual's own background predisposition.
Adaptation
Figure 1 The stress-adaptation-growth dynamic. Reproduced from
Kim YY (2001 ) Becoming Intercultural: An Integrative Theory of
Communication and Cross-Cultural Adaptation, p. 59. Thousand Oaks,
CA: Sage.
626 Cross-Cultural Adaptation
Communication Factors
Like all human-social activities, cross-cultural adaptation
occurs in and through the communication interface between
the stranger and the surrounding environment. Just as the
natives acquire their capacity to function in their social world
through communicative interactions throughout their lives, it
is through communication that strangers come to learn and
internalize the significant symbols and practices of the host
culture. Through communicative engagement, nonnatives gain
insights and skills that are necessary to achieving a level of
success in their adaptive endeavor.
Communication activities comprise two basic, inseparable
dimensions, personal communication and social communica-
tion. Through personal (or intrapersonal) communication,
individuals organize themselves internally with respect to
their surroundings, developing ways of engaging themselves
in social communication activities. If the personal communi-
cation can be compared to the 'off-line functions' of computer
systems, social communication carries out the 'on-line' func-
tions of interfacing with the computer environment through
input-output transactions of messages.
Personal communication
In the context of cross-cultural adaptation, personal communi-
cation can be best understood in terms of 'host communica-
tion competence,' that is, the overall internal capacity of a
stranger to decode and encode information in accordance
with the host cultural practices of communication. Host com-
munication competence serves as the main engine that drives
the adaptation process. Successful adaptation of strangers can
be realized only when their internal communication systems
sufficiently overlap with those of the natives. An individual's
level of host communication competence at a given time
reflects his or her overall capacity to participate in host social
communication activities, whereas the lack of such compe-
tence manifests itself in various forms of miscommunication,
social inadequacies, and, in some cases, marginalization.
Host communication competence comprise three interre-
lated subcategories: cognitive, affective, and operational. Cog-
nitive competence refers to an individual's overall intellectual
capacity to deal with the host culture, including the knowledge
of the host culture, history, social institutions, and rules of
interpersonal conduct. Knowledge of the host language, in
particular, serves as the primary conduit for adaptation, one
that enables strangers to access the accumulated records of the
host culture and learn to think in the way the native speakers
think. Knowledge of the local language means not just techni-
cal knowledge such as phonetics, syntax, and vocabulary. It
also includes pragmatic knowledge about the everyday use of
the language including the many subtleties in the way the
language is spoken and interpreted in various formal and
informal social contexts.
Affective competence refers to the emotional and motiva-
tional capacity to deal with the various challenges of living in
the host environment. Specifically, it includes positive attitude
(acceptance of the host culture) and adaptation motivation
(the willingness to learn and participate in it). A positive,
willing, and flexible self-other orientation helps engender
greater openness and lessen unwarranted prejudicial cynicism
toward new cultural experiences. The affective competence also
includes the development of a capacity to understand, and
participate in, the local people's emotional and esthetic sensi-
bilities, thereby making it possible for strangers to establish
a deeper-level psychological connection with the natives.
Although initially difficult, strangers become better able to
participate in the surrounding landscape of art, food, music,
sports, and other everyday experiences of fun, joy, humor, and
happiness, and of anger, despair, and disappointment. In
this visceral connection, strangers are on their way to being
'insiders' who are capable of forming close relationships with
the natives.
The cognitive and affective capabilities work side by side the
operational competence, the capacity to express outwardly by
choosing a 'right' combination of verbal and nonverbal acts in
specific social transactions of the host environment. As sttan-
gers become increasingly competent cognitively and affec-
tively, they are better able to act in greater consonance with
the natives' actions and display greater resourcefulness with
which to negotiate and reconcile cultural differences and to
form appropriate and effective action plans for accomplishing
personal and social goals.
Social communication
Host communication competence is directly and reciprocally
linked to participation in the social communication activities
of the host society. An individual's host social communication
activities are enabled or constrained by his or her capacity to
communicate effectively and appropriately. Conversely, the
individual's host communication competence cannot be devel-
oped without participating in the social communication
processes.
Interpersonal communication activities with natives offer
opportunities for learning the host cultural practices and for
engaging in 'corrective exchanges' with respect to the use of the
host communication system, including its verbal and nonver-
bal codes. Through experiences of direct engagement with
native-born individuals, strangers obtain vital information
and insight into the mindsets and behaviors of the local peo-
ple. The natives' behaviors provide strangers with points of
reference or a check and validation of their own communica-
tion behaviors. By conferring with the natives, strangers can
confirm or reject presumed meanings and motives in natives'
communication behaviors. They learn not only what to do but
also some of the relevant cultural and personal values.
Notwithstanding cross-cultural barriers, strangers develop
at least a limited number of interpersonal ties with host
nationals over time. As their host communication competence
improves, so does their self-confidence to participate in
host interpersonal communication activities. Through active
engagement in host interpersonal communication activities,
strangers also begin the process of constructing a set of satisfy-
ing and supportive relationships with natives. Newly arrived
immigrants and sojourners tend to be drawn to coethnics
when they are available nearby. Over time, however, more
ties with local people are likely to be incorporated into their
personal networks, replacing some of the coethnic ties.
Cross-cultural adaptation is further facilitated by participa-
tion in mass communication activities of the host society.
Through a wide range of mediated and public forms of
Cross-Cultural Adaptation 627
communication such as radio, television, newspaper, maga-
zine, movie, art, literature, music, and drama, strangers interact
with their host milieu without direct interpersonal involve-
ments. Such mass communication activities help broaden the
scope of cultural learning opportunities beyond the immediate
social context with which a stranger routinely comes into
contact. Compared to host interpersonal communication
activities, host mass communication activities render less
opportunity for direct feedback, serving as an important source
of cultural and language learning, particularly during the early
phases of the adaptation process.
In many societies and communities today, strangers' inter-
personal and mass communication activities involve their
coethnics or conationals and home cultural experiences, as
well. Some form of ethnic mutual-aid or self-help organiza-
tions, including religious organizations, may be available to
render assistance to those who need material, informational,
emotional, and other forms of social support. In addition,
ethnic mass media such as newspapers, radio stations, and
television programs may be accessible via the Internet or in
prerecorded audio- and videotapes and computer disks.
Findings from studies of long-term cross-cultural adapta-
tion generally indicate that participation in ethnic inter-
personal and mass communication activities can be helpful
to the initial phase of the cross-cultural adaptation process
when newly arrived strangers lack host communication com-
petence and access to host interpersonal resources. Beyond
the initial phase, however, heavy and prolonged reliance
on coethnics have been found to be either an insignificant
influence on, or impede, the long-term adaptation process
with respect to the host society at large.
Environmental Factors
The adaptation function of the strangers' personal and social
(interpersonal, mass) communication cannot be fully ex-
plained in isolation from the conditions of the host environ-
ment. Since different societies and communities present
different environments for cross-cultural adaptation, a given
stranger can be more successful in adapting to a certain envi-
ronment than to another one.
Of various environmental characteristics, three key factors
are discussed below as particularly important to individual
strangers' adaptive efforts (1) host receptivity, (2) host confor-
mity pressure, and (3) ethnic group strength. These three fac-
tors help define the relative degrees of 'push and pull' that a
given host environment presents to strangers.
Host receptivity
Host receptivity refers to the degree to which the receiving
environment welcomes and accepts strangers into its interper-
sonal networks and offer them various forms of informational,
technical, material, and emotional support. A society or a
community can be more receptive toward certain groups of
strangers while unwelcoming toward certain others. For exam-
ple, Canadian visitors arriving in a small town in the United
States are likely to find a largely receptive environment. On the
other hand, the same town may show less receptivity toward
visitors from lesser known and vastly different cultures. Differ-
ences in host receptivity toward different groups of strangers
can be attributed to a number of plausible reasons such as the
nature of the relationship, friendly or hostile, between the host
country and the stranger's home country; cultural and ideolog-
ical difference and incompatibility between the two cultures;
and the racial/ethnic prejudice predominantly held by the
society against strangers in general or the particular group.
Host conformity pressure
Along with receptivity, individuals face differing levels of con-
formity pressure from the host environment. Conformity pres-
sure refers to the extent to which the society challenges
strangers to adopt the normative patterns of the host culture.
Different host environments show different levels of accep-
tance and appreciation of strangers and their ethnic character-
istics. In general, people in heterogeneous and cosmopolitan
societies such as the United States tend to hold more plural-
istic and tolerant attitudes toward ethnic differences, thereby
exerting less pressure on strangers to change their habitual
ways. Even within a country, ethnically more heterogeneous
metropolitan areas tend to display lower levels of host confor-
mity pressure than do smaller, ethnically more homogeneous
rural towns.
Ethnic group strength
The third environmental factor that influences strangers' cross-
cultural adaptation is the strength of their ethnic group in the
context of other groups in the host society or community.
Depending on relative group size, status, or power, stronger
ethnic groups are likely to provide their members with a more
vibrant subculture and practical services to its members. In
doing so, however, a strong ethnic community is likely to
encourage the maintenance of ethnic culture and communica-
tion, and even exert a subtle or explicit pressure to conform to
the ethnic cultural norms. A long-term consequence of such
ethnic community influences is likely to be an impediment,
intended or not, to individual community members' active
participation in host social communication processes and
long-term cross-cultural adaptation to the host culture at large.
Predisposition Factors
Each stranger begins the cross-cultural adaptation process with
a different set of preexisting characteristics. The internal condi-
tions with which they begin their life in the host society help
set the parameters for the way they relate to the new environ-
ment and their own subsequent adaptive changes. The various
predispositional differences in which strangers differ in their
background conditions can be organized into three categories:
(1) preparedness, (2) ethnic proximity/distance, and (3) per-
sonality predisposition. Together, these characteristics help
define the degree of a stranger's 'adaptive potential.'
Preparedness for change
Strangers come to their new environment with differing levels
of preparedness for dealing with that environment. Prepared-
ness is directly linked to the level of readiness to develop host
communication competence. Influencing the strangers' readi-
ness are a wide range of formal and informal learning experi-
ences they may have had prior to moving to the host society.
Included in such activities are the schooling and training in,
628 Cross-Cultural Adaptation
and the media exposure to, the host language and culture, as
well as their prior cross-cultural adaptation experiences, partic-
ularly contacts with members of the host society. In addition,
the strangers' preparedness is often influenced by whether their
move to the host society is voluntary or involuntary and for
how long. Voluntary, long-term immigrants, for example, are
likely to enter the host society with greater readiness and
willingness for making necessary efforts to adapt, compared
to the temporary visitors who unwillingly relocate for reasons
other than their own volition.
Ethnic proximity/distance
The ethnicity of a stranger plays a role in the cross-cultural
adaptation process by serving as a certain level of advantage
or handicap. It does so by affecting the ease or difficulty with
which the stranger is able to develop the communication com-
petence in a given host society, and to participate in its social
communication activities. For instance, many Japanese busi-
ness executives in the United States face a greater amount of
challenge in overcoming their language barrier than are their
counterparts from the United Kingdom. The strangers' visual
ethnic markers (such as height, skin color, and facial features)
and audible ethnic markers (such as accents and other speech
patterns) influence the degree to which the natives are psycho-
logically prepared to welcome them into their interpersonal
networks. Accordingly, ethnicity influences the cross-cultural
adaptation process in at least two interrelated manners. First,
each stranger faces a level of linguistic and cultural distance for
the stranger to overcome in order to develop host communica-
tion competence and to participate in the host social commu-
nication activities. Second, each stranger creates a certain level
of psychological distance (or affinity) in the minds of the
natives, which, in turn, would affect the natives' receptivity
toward the stranger.
Adaptive personality
Along with preparedness and ethnicity, strangers enter a host
environment with a personality, or a set of more or less
enduring traits of sensibilities. Their existing personality serves
as the inner resource, based on which they pursue new cultural
experiences with varying degrees of enthusiasm and success. Of
particular interest are three interrelated personality resources
that would help facilitate the strangers' adaptation by enabling
them to endure stressful challenges and to maximize new
learning: (1) openness, (2) strength, and (3) positivity. Open-
ness is an internal posture that is receptive to new information.
Openness enables strangers to minimize their resistance and to
maximize their willingness to attend to the new and changed
circumstances. It is a psychological orientation that allows for
greater tolerance for ambiguity, intellectual flexibility, and cul-
tural and interpersonal sensitivity. The strength of personality
reflects the inner quality of resilience, hardiness, patience, and
persistence. Individuals with high levels of personality strength
tend to be stimulated by new challenges and remain confident.
Openness and strength of personality is closely associated with
positivity, an affirmative and optimistic outlook of mind, or
the internal capacity to defy negative prediction. Strangers with
a positive personality are better able to endure stressful events
with a belief in the possibilities of life in general as opposed to
being overcome by unwarranted defeatist cynicism.
Facets of Adaptive Change
Through the interactive workings of the above-described fac-
tors of the communication activities, of the environment, and
of the background of an individual stranger, the process of
cross-cultural adaptation unfolds. Emerging in the adaptation
process, over time, are three interrelated facets of internal
change in the stranger (1) increased functional fitness in carry-
ing out daily transactions; (2) improved psychological health
in dealing with the environment; and (3) emergence of an
intercultural identity orientation. These three facets are interre-
lated developmental continua, in which individual strangers
can be placed at different locations reflecting the different
levels of adaptive change at a given point in time.
Functional Fitness
As previously noted, strangers conduct continuous experi-
ments in the host environment. They instinctively strive to
'know their way around,' so that they may effectively control
their own behavior and the behavior of others in the host
environment. Through repeated activities resulting in new
learning and internal reorganizing, strangers achieve an in-
creasing functional fitness between their internal responses
and the external demands of the host environment. Well
adapted strangers would be those who have accomplished a
desired level of appropriate and effective communication rela-
tionship to the host environment - particularly with those
individuals with whom the strangers carry out their daily activ-
ities. As they achieve increased communication competence,
they are also better able to meet their personal and social needs.
Psychological Health
Everyone requires the ongoing validation of his or her social
experience. Being unable to meet this basic human need can
lead to symptoms of mental, emotional, and physical distur-
bance. The shifting of the self-world relationship brings about
heightened levels of inner conflict due to the gap between the
internal and external reality. In the absence of adequate host
communication competence, engagement in host social com-
munication activities, and functional fitness, strangers are
subject to frustration, leading to the symptoms of poor psycho-
logical health. The very meaning of the term culture shock
employed in studies of temporary sojourners suggests lack of
psychological health, stemming from the poor fit of a stran-
ger's intrapsychic system to that of the natives. Most strangers,
however, are able to achieve, over time, a higher level of
psychological health and a subsiding level of disturbances in
dealing with the host environment. They achieve an increased
internal integration - a sense of inner cohesiveness and
confidence.
Intercultural Identity
Long-term adaptive change also includes the gradual emer-
gence of an intercultural identity, a psychological orientation
toward self and others that is no longer clearly and rigidly
defined by either the original cultural identity or the iden-
tity of the host culture. As the host cultural elements are
Cross-Cultural Adaptation 629
increasingly incorporated into their self-concept, the stranger's
self-other orientation becomes more flexible. While rooted
in the original cultural identity, the stranger's identity is broad-
ened to incorporate new dimensions of perception and
experience in which the original cultural identity begins to
lose its distinctiveness and rigidity. More than likely, this
kind of identity transformation is unlikely to be a matter of
conscious decision but simply an aspect of the natural psy-
chological evolution beyond the boundaries of childhood
enculturation.
Terms close to intercultural identity include 'metaidentity,'
'cosmopolitan identity,' or 'transcultural identity,' all of which
indicate an identity orientation that is less dualistic and more
inclusive, less categorical and more individuated, less parochial
and more universalized. The development of an intercultural
identity does not come about without certain 'costs' that may
be deemed too high by some. The process can be filled with
ambivalence and internal conflict between one's loyalty to the
original identity and a necessity to embrace a new one that is
rooted in, embracing, and not discarding, the original cultural
identity, just as acquiring knowledge and skill in the host
language does not necessarily result in corresponding loss in
the original language.
Conclusion
Cross-cultural adaptation is, in essence, a largely unconscious
process of cultural learning and personal change that takes
place over time. It is a dynamic unfolding of the natural
human tendency to struggle for an internal equilibrium in
the face of adversarial environmental conditions. It is a process
in which, over a prolonged period, a newcomer can potentially
become a fully functioning insider. The process unfolds in the
fluctuating experiences of stress, adaptation, and growth, a
process in which the newcomer advances toward achieving a
higher level of personal and social efficacy with respect to the
environment that was once new and unfamiliar. The pursuit of
cross-cultural adaptation opens up a path of personal develop-
ment, in which individuals stretch themselves out of the famil-
iar and reach for a deepened and more inclusive understanding
of human conditions, including their own.
Central to this process is the individual's ability to commu-
nicate in accordance to the norms and practices of the host
culture and his active engagement in its social communication
processes. Multiple forces are simultaneously at work sur-
rounding the communicative interface between the individual
and the host environment, from the conditions of the environ-
ment and the ethnic and personal predispositions of the indi-
vidual. Like a locomotive engine, each unit operating in this
process affects, and is affected by, the workings of all other
units. Some of the factors may be more pertinent than others in
specific cases of cross-cultural adaptation. In some cases, adap-
tive successes may be almost entirely due to the openness,
strength, and positivity of the stranger's personality, which
enable the stranger to overcome even the most severely unre-
ceptive host environment. In other cases, very little adaptive
change may take place in strangers whose ethnic community
offers an almost complete insulation from having to face the
host cultural challenges.
Should strangers choose to adapt successfully, they would
benefit from being prepared and willing to face the stressful
experiences that are part and parcel of the adaptation process.
They would need to concentrate on acquiring new cultural
communication practices and be willing to put aside some of
the old ones, recognizing the critical importance of host com-
munication competence as the fundamental mechanism that
drives their own adaptation. They would also need to partici-
pate actively in the interpersonal and mass communication
processes of the host environment. In addition, through culti-
vating the adaptive personality of openness, strength, and
positivity, they are likely to be more successful in overcoming
temporary setbacks and attaining greater success in pursuing
their personal and social goals.
With the advent of globalization, we no longer have to leave
home to experience new cultural learning and cross-cultural
adaptation. For many people in the world, physical distance no
longer dictates the extent of exposure to the images and sounds
of once distant cultures. Many urban centers present their own
contexts of cross-cultural adaptation, the natives are routinely
coming in face-to-face contacts with nonnatives. Such encoun-
ters often compel everyone involved to put aside and even
unlearn at least some of the original cultural patterns. The
rapidly changing reality can be threatening to many people,
provoking a keen sense of unsettling discontinuity, malaise,
and nostalgia for the age of certainty, permanence, and a fixed
and unitary cultural identity.
In the end, then, cross-cultural adaptation is a journey that
compels individuals to make choices and to be accountable for
the outcomes. Those who have successfully crossed cultural
and subcultural boundaries are likely to be those who chose
to adapt and to be changed by that choice. Although their
tribulations can be staggering, they have worked through the
setbacks and come out victorious with an increased capacity to
see others, themselves, and situations in a new light. They are
the ones who bear witness to the promise of the cross-cultural
adaptation process. Their personal achievements are a tribute
to the ever-present human capacity to adapt - the capacity
to face challenges, learn from them, and evolve into a greater
self-integration that defies the simplistic and conventional
categorizations of people and reveals a new, intercultural way
of being in the world.
See also: Evolutionary Social Psychology; Perceptual Development;
Social Support; Stress and Illness.
Further Reading
Adler PS (1987) Culture shock and the cross-cultural learning experience. In: Luce LF
and Smith EC (eds.) Toward Internationalism, pp. 24-35. Cambridge, MA: Newbury
Original work published 1972.
Berry JW (1980) Acculturation as varieties of adaptation. In: Padilla A (ed.)
Acculturation: Theory, Models and Some New Finding, pp. 9-25. Boulder, CO:
Westview.
Berry JW (2008) Globalization and acculturation. International Journal of Intercultural
Relations 32: 328-336.
Furnham A and Bochner S (1 986) Culture Shock: Psychological Reactions to Unfamiliar
Environment. New York: Routledge.
Herskovits M (1958) Acculturation: The Study of Culture Contact. Gloucester, MA:
Peter Smith.
630 Cross-Cultural Adaptation
Kim YY (1988) Communication and Cross-Cultural Adaptation: An Integrative Theory.
Clevedon, UK: Multilingual Matters.
Kim YY (2001 ) Becoming Intercultural: An Integrative Theory of Communication and
Cross-Cultural Adaptation. Thousand Oaks, CA: Sage.
Kim YY (2003) Intercultural adaptation: An interdisciplinary overview. In: Gudykunst W
(ed.) Cross-Cultural and Intercultural Communication, pp. 243-257. Thousand
Oaks, CA: Sage.
Kim YY (2008) Intercultural personhood: Globalization and a way of being. International
Journal of Intercultural Relations 32: 359-368.
Kim YY, Lujan P, and Dixon L (1998) T can walk both ways': Identity integration
of American Indians in Oklahoma. Human Communication Research 25: 252-274.
Redfield R, Linton R, and Herskovits M (1936) Outline for the study of acculturation.
American Anthropologist 38: 149-152.
Taft R (1977) Coping with unfamiliar cultures. In: Warren N (ed.) Studies in
Cross-Cultural Psychology, vol. 1, pp. 121-153. London: Academic Press.
Ward C (2001) The A, B, Cs of acculturation. In: Matsumoto D (ed.) The Handbook of
Culture and Psychology, pp. 411-445. New York: Oxford University Press.
Ward C, Bochner S, and Furnham A (2001) The Psychology of Culture Shock, 2nd edn.
Philadelphia, PA: Routledge.
Crowd Psychology
S Reicher, University of St. Andrews, Fife, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Collective empowerment An increased sense that the group
has the ability to do what it wishes, even against the
opposition of outgroups.
Collective self-objectification The transformation of the
group's idea of how society should be organized into the
actual organization of society.
Contagion The notion that people automatically follow the
ideas and emotions of those around them.
Crowd A large face-to-face assembly of people who consider
themselves to be members of a common social group.
Deindividuation The idea that identity is lost in the group
and that, as a consequence, people lose control over their
behavior.
Elaborated social identity model of crowds (ESIM) A
theory of how crowd identities and crowd action develop
through the intergroup interaction between crowd members
and outgroups.
Emergent norm theory A theory of how norms emerge in
crowds through the interactions of members and the actions
of prominent individuals or 'keynoters.'
Social identity One's sense of oneself as a member of a
particular social group (e.g., 'I am an American') along with
the significance attached to that membership.
Social identity model of crowds (SIM) A theory of how
crowd behavior derives from the transition from individual
to social identity and the process of defining
identity-appropriate behavior in context.
Introduction
In his 1988 textbook on Group Processes, Rupert Brown notes
that "there is a widespread view both in our society and in our
discipline that groups are bad for you." If this is true of groups
in general, it is particularly true of crowds. Indeed, the very
language used to denote crowds - terms such as 'the mad mob,'
'the herd,' 'the rabble' - speaks in equal measure of fear and
derision. This is entirely understandable if one considers the
contexts in which a science of crowd psychology emerged.
Crowd psychology centers on the question of how people
are transformed when they assemble together in large num-
bers. Although, as we shall see, when authors refer to crowds,
they are generally referring not simply to large numbers of
individuals but rather to assemblies of people who share
some common sense of who they are and some common
purpose: crowds of demonstrators, crowds of sports fans, revo-
lutionary crowds. We shall also see that there are important
psychological differences between masses who are merely
copresent and copresent masses who perceive themselves as
sharing a group membership.
Interest in crowds goes as far back as one cares to look - and
the general consensus seems to be that whatever transforma-
tions are wrought by crowds, they are inevitably negative.
Herodotus, sometimes styled 'the father of history,' declared
that "there is nothing less understanding and more proud than
the blind mass". He was but the first of a long line of eminent
thinkers who made disparaging comments about crowds and
crowd psychology. However, a distinctive crowd science devel-
oped only during the late nineteenth century.
This period was dominated by a crisis of social order.
Industrialization had transformed a largely agrarian popula-
tion into urban masses. New forms of social organization such
as trade unions and new political movements, such as syndi-
calism, anarchism and, above all, socialism, were leading these
masses to challenge gaping social inequalities. How, then, were
the elite to maintain their dominance? How were they to face
down the challenge of the masses and even engage them in
defense of the existing order? These were the dominant ques-
tions of the day.
But if the masses in general were seen as a potential threat
to the status quo, the crowd was the radicalized mass in action.
It was the potential catastrophe made actual. And so the crowd
became a dense symbol which regrouped all the bourgeois
fears and fantasies of chaos. If social discipline seemed under
threat from alcoholism, the crowd was characterized as drunken,
either metaphorically or literally. If the patriarchal order seemed
under assault from independent women, the crowd was
described metaphorically as feminine, while the worst of crowds
were seen to be crowds of women - such as the 'petroleuses' of
the 1871 Paris Commune.
Indeed, the Paris Commune has a particular prominence
in the history of crowd psychology. For, to take progression
one step further, where the masses were a potential threat and
crowds in general an actual threat to the elite, the Commune
crowds had, at least temporarily, overthrown the elite and
created what has been described as the first socialist republic
in history. Those of the elite who had experienced the Com-
mune had witnessed a future they did not like and which they
would do everything to prevent. A key part of this was to find
ways of mitigating against crowd action. It is hardly surprising,
then, that French theorists dominated early crowd science and
that they presupposed that crowd psychology and crowd
action are a purely negative phenomena.
I shall start by analyzing this early crowd science and
showing how its social concerns are embedded deep within its
core assumptions. I shall then outline alternative approaches
which, rather than seeking to pathologize crowd psychology
(and hence crowd action), pay attention to the perspectives of
crowd participants, analyze the patterns of crowd action, and
seek to understand the relationship between the two. Above
all, I shall seek to demonstrate that the early crowd theorists
631
632 Crowd Psychology
have done us a disservice by characterizing crowd members as
mindless and crowds as irrelevant to the normal functioning of
society. For crowds provide a particularly productive site in
which to understand how people are constituted and act as
social subjects. Crowds are critical to the formation of the
social identities and social relationships which regulate every-
day life.
Classic Crowd Science: A Tale of Loss
Early crowd science was a vibrant field. Many theorists such
as Fournial, Tarde, Sighele, Rossi, and others debated crowd
behavior, not only in academic texts but also in the society
magazines of the day. But these names have largely disap-
peared. Only one remains, that of Gustave Le Bon. His book
The Crowd: A Study of the Popidar Mind, first published in 1895,
is still cited in psychology journals, in official reports, and in
the popular media, ft has been described as the most influen-
tial psychology text of all time in that it not only analyzed mass
behavior, but also that it helped create the mass politics of the
twentieth and twenty-first centuries.
Le Bon - both the man himself and the basis for his
enduring influence - exemplify the general themes of fear,
hostility, and repression which frame the emergence of crowd
science. Le Bon's first experience of crowds was as the head of
ambulance services in the Paris Commune. That experience
bred him in a deep hatred and contempt for the masses. The
rest of his career was devoted to understanding how crowds
could be tamed and engaged for - not against - the nation.
Indeed, the text of which he was most proud was not 'The
crowd' but a far more obscure book called 'La Psychologie
Politique et la Defence Social' published in 1910.
What made Le Bon stand out from his fellow theorists was
not so much his conceptual originality as his practicality.
Where others documented the bestiality of crowds and did
little more than throw their hands up in horror, Le Bon sought
to show how leaders could take advantage of crowd psychol-
ogy and use it to their own ends. His book was only in small
part a theoretical analysis. The larger part reads like a primer
for leaders who wish to exploit the power of crowds. And the
leaders he sought to educate were the right wing leaders of his
time - for whom he ran a regular weekly luncheon club.
Moreover, these leaders reciprocated his enthusiasm. The list
of those who praised Le Bon reads like a roll call of the auto-
crats of the early twentieth century. Mussolini, for instance,
after praising Le Bon to the skies, declares that he built
the principles of the (fascist) Italian state on the basis of 'The
crowd.'
When it comes to Le Bon's actual crowd psychology, the
predominant theme is one of loss. His starting point lies in
the notion that, on entering the crowd, people become anony-
mous and lose their sense of personal identity. This first process
submergence is termed submergence. Submergence, in turn, leads
people in crowds to lose control over what they feel and do.
Because they can no longer access the personal values and
standards that allow them to judge what is and is not appropri-
ate, crowd members simply follow passing ideas and emotions -
particularly emotions, since the self as the seat of the intellect
has been occluded. This second process is termed contagion.
Finally, submergence and contagion lead the crowd members
to lose their civilized standards. Because people's conscious
access to internal standards has been blocked, the emotions
and ideas that govern action come predominantly from the
'racial unconscious' - an atavistic residue which people share
in common from a distant past. And because this residue is
ancient and primitive, so are the actions to which it leads. This
third and last process is termed suggestion.
Altogether, the picture that emerges of crowd psychology
and of crowd behavior is unremittingly bleak. Crowds, says
Le Bon, are only powerful for destruction. The crowd member,
he declares, is a barbarian who descends several rungs on the
ladder of civilization simply by virtue of becoming part of the
crowd. Most revealingly, in a passage that reveals much of both
Le Bon's 'science' and his politics, he remarks that
among the special characteristics of crowds there are several - such
as impulsiveness, irritability, incapacity to reason, the absence of
judgement and of the critical spirit, the exaggeration of the senti-
ments and others besides - which are almost always observed in
beings belonging to inferior forms of evolution - in women children
and savages for instance.
There is only one exception to this analytic negativity.
Le Bon acknowledges that crowd members can be extraordi-
narily heroic at times. Moreover, this flows not only from the
loss of self (and hence a loss of concern for self-preservation)
but also from the one element of gain in Le Bon's account. That
is, on becoming submerged in the mass, people might forget
their individuality, but they become part of something much
bigger and hence they evince a sense of immense power which
allows them to do almost anything - thus making crowds the
ultimate nightmare for those who have stakes in the social
order: total power without any sense of responsibility.
Le Bon's work continues to be cited in contemporary texts.
What is more, his work continues to have relevance through
the development of deindividuation theory. This approach is
derived directly from the idea of submergence and is most
centrally concerned with the effects of anonymity (or 'percep-
tual immersion') in a group. There are a number of variants of
the theory. One, most associated with Zimbardo, suggests that
anonymity leads to a lowered concern with social standards
and a lowered threshold for exhibiting antisocial behaviors.
A second, associated with Diener, suggests that immersion in a
group leads to a lowered awareness of internal standards and
increased domination by external stimuli. In effect, people
become uncontrolled and simply go along with whatever is
suggested to them. A third variant effectively integrates the
other two and argues that, in the group, people become less
aware of both internal and social standards and hence both
more uncontrolled and more willing to behave in antisocial
ways. For all these differences, all the variants equate collective
participation with the loss of self and of the loss of self with
the loss of reasoned or socially responsible action.
Not surprisingly, Le Bon's approach evoked strong opposi-
tion from early on. Floyd Allport in particular strongly attacked
the notion of a 'racial unconscious' or 'group mind' as an
empty and meaningless concept. His 1924 book, often (mis-
takenly) seen as the first social psychology text, and certainly
a seminal influence in the development of the discipline in
North America, is written largely as a rebuttal of Le Bon's ideas.
Crowd Psychology 633
However, paradoxically, it is arguable that Le Bon's influence
was enhanced rather than diminished by those, like Allport,
who challenged him. For, in challenging his explanation of
crowd action, they accept his description of such action as
mindless and destructive. What is more, the difference at the
explanatory level is more a matter of displacing rather than
discarding the Le Bonian narrative of loss.
To be more concrete, Allport - and the so-called 'conver-
gence' approach to crowd action which derived from his work -
suggests that crowd action reflects the preexisting individual
characteristics of crowd members. If crowds are violent or
destructive, it is because flawed individuals are drawn to such
events. All that has happened here is that the locus of crowd
pathology and loss has shifted from the group to the individual
level. People become crowd members because they lack some-
thing already, they do not come to lack something because
they have become crowd members.
For all the sharp debates that divided classic crowd science,
then, the commonalities are ultimately more telling. Crowds
either create a pathological psychology or else attract indivi-
duals with psychopathologies. As a result, crowds are patho-
logical entities. There is no place here for understanding how
social and cultural factors might shape the understandings and
actions of people in collective settings. There is no basis for
addressing, let alone understanding the socially meaningful
nature of crowd action.
Crowd Patterns and Crowd Norms
There are many criticisms that can be made of Le Bon's work
and classic crowd science more generally. But at their root lies
yet another loss - the loss of context. In both Allport's and
Le Bon's accounts, the focus is exclusively on the crowd.
Although the crowds that they reference are drawn from impor-
tant social groups - crowds of socialists or trades unionists or
miners - they write as if mere physical assembly is enough to
explain how crowd members behave. They make no reference
to the broad social context of struggles between these groups
and others for power and resources in society. Equally, they
make no reference to the immediate context of struggles
between these groups and others (the army, the police, private
security forces) on the streets. And once the context is lost, we
can no longer see how action might make sense as an attempt to
challenge the other or as a reaction to the acts of the other.
Instead, behavior can only be attributed as reflecting something
inherent about crowd members or crowd process, something
universal that holds for all time and all places. In short, decon-
textualization pathologizes and essentializes and eternalizes
the results of specific social processes.
By the latter half of the twentieth century, however, things
were changing. Academia was changing from an elite system to
more of a mass system. Academics who experienced crowds
were more likely to be positioned as engaged insiders as horri-
fied outsiders, whether (in the United States) within the civil
rights movement, the anti-Vietnam protests, or (in Europe) in
the events of 1968 and its aftermath. And, from this perspec-
tive, crowds were to be welcomed rather than to be feared, they
were healthy rather than pathological phenomena, and, above
all, they were meaningful rather than meaningless.
This was true across the humanities and human sciences.
A series of seminal historical studies demonstrated that crowd
members are not those who stand at the margins or against
society - in more pejorative parlance, 'riff-raff.' Rather, they
characteristically come from more stable, integrated, and
'respectable' layers. They also showed that crowd action, far
from being inchoate and universally destructive, is typically
highly patterned and that the patterns are highly socially mean-
ingful. Crowds act in ways that reflect shared social belief
systems. Indeed for many historians, who face the problem
of recovering the perspective of those groups in society who,
in the past, did not leave written records, crowds provide a
uniquely valuable resource.
Similar conclusions were drawn by those political scientists
and sociologists studying contemporary events such as the
wave of black urban riots across many US cities in the 1960s
and 1970s. The massive report of the US Riot Commission in
1968, using a database of 1200 interviews from 20 cities, noted
that the average rioter was more socially integrated and better
educated than the average black person. The evidence also
shows that, far from being random or impulsive, riots were
articulate protests against specific grievances. Finally, and per-
haps most importantly, a close examination of events showed
that crowds were neither wild nor orgiastic. They were discrim-
inating in their choice of targets, they only attacked those state
officials they viewed as enemies (notably, the police), they only
looted shops of outsiders against which they had concrete
grievances. Fogelson sums up this evidence in a phrase that
contrasts markedly and memorably with the prevalent image
of mob madness: "Restraint and selectivity were among the
most crucial features of the riots."
In sum, even the most violent of crowd events are subject to
normative constraint, even if the norms (attacking particular
people and property) are at odds with those that regulate
everyday life. These conclusions, drawn from the field, have
been corroborated by experimental evidence. Thus, a meta-
analysis of deindividuation studies suggests that, far from act-
ing in uncontrolled or antisocial ways, those made anonymous
in groups tend to increase their adherence to situational
norms.
How then are such norms created? The first answer was
provided by Turner & Killian's aptly named 'emergent norm
theory' (ENT) first published in 1957. This theory suggests that
crowds do not become homogenous entities in an instant.
Rather, there is an extended period where people mill about.
They are addressed by many would-be influence agents, or
'keynoters.' Gradually, particular keynoters who are more
striking than others begin to gain sway and norms begin to
spread through the crowd. Homogeneity, like normativity, is
therefore an emergent property of encounters among crowd
members.
ENT provided a nuanced and compelling portrait of the
micro interactions in the crowd out of which shared under-
standings emerge. This is its legacy from the symbolic interac-
tionist tradition. However, it is less successful in explaining
why particular keynoters become successful leaders and why
some norms rather than others come to shape crowd action.
This can be traced to Turner & Killian's reliance on a desocia-
lized model of influence deriving from the small group psy-
chology of the time. Such models root influence in the
634 Crowd Psychology
personal qualities and interpersonal relations between individ-
ual crowd members. As a consequence, it is hard to see how
group (and crowd) norms relate to broad cultural belief sys-
tems. Such a model can explain the patterned nature of crowd
action; however, it cannot explain why such patterns should be
socially meaningful and reflect broader conceptions of society.
In effect, the micro is divorced from the macrosocial. Although
ENT and other normative accounts mark a great step forward
from classic crowd psychology, they, like Le Bon, still face
the problem of relating crowd action to the larger social con-
text in which it occurs.
The Social Identity Model and Intragroup
Dynamics in Crowds
Over recent decades, the social identity has become the domi-
nant psychological approach to group processes in general.
This approach aims explicitly to understand how ideological
and structural factors shape group action and it is formulated
to be as relevant to large-scale social categories (such as nation,
religion, or 'race') as to small groups of familiars. It is under-
standable, then, that social identity tenets have proved success-
ful in addressing the specific issues of crowd psychology.
Indeed, the social identity model (SIM) of crowd behavior
has, by now, itself become the dominant contemporary
approach to crowd psychology.
SIM contests the classic models of crowd psychology as loss
through the very concept of social identity which lies at its
heart. It will be recalled that Le Bon's analysis starts from the
idea that selfhood is lost upon entering a crowd. This derives
from a singular and individualistic notion of selfhood: people
have a unique personal identity which is the sole basis of
behavioral control. If this personal identity is compromised
(as in the crowd), then control over behavior is lost. By con-
trast, social identity theorists view self as a rich and varied
system which exists at different levels of abstraction. Some-
times, I think of myself in terms of how I, as an individual,
differ from other individuals (personal identity: 'I' vs. 'you').
At other times, I may think of myself in terms of belonging to
a social category - I am an American, a Catholic, a woman, a
Mets fan ... - and in terms of how my group differs from other
groups (social identity - 'we' vs. 'they').
The foundational assumption of social identity research is
that when people act as group members, they shift from acting
in terms of personal identity to acting in terms of social iden-
tity. That is, the eclipse of one's sense of oneself as a unique
individual is not a loss but a refocusing of identity. Hence, as
intimated at the start, an important psychological distinction
is implicitly made between a physical group - a set of people
who happen to be copresent at the same place and time - and
a psychological group - a set of people who share a sense of
themselves as belonging to the same social category. The dif-
ference between the two can best be illustrated by a simple
thought experiment: imagine you are on a crowded commuter
train heading home. People are squeezed together in a carriage,
but each remains psychologically an individual, seeking to
avoid eye contact with others, reading the same paper as others
but resenting anyone looking at theirs, and feeling discomfort
or even disgust at any physical contact with others. Then, the
train stops. After a long delay, there is an insufficient excuse for
what has happened. Now, people begin to think of themselves
together as aggrieved commuters against the train company.
And that shared identification transforms their behavior. They
start turning toward one another, talking to one another, even
sharing their sandwiches.
In a similar vein, SIM draws an explicit distinction between
physical crowds based on copresence (aggregates) and psycho-
logical crowds based on social identification. These, as has
been argued previously, are the types of assembly that all
crowd psychologists have been concerned with (even if this is
not made explicit in their models) and therefore, this is the
type of assembly for which the term 'crowd' will henceforth be
reserved. SIM proposes three transformations that occur when
people join a crowd and these are detailed in the following
section.
The Cognitive Transformation
Just as identity is not lost in crowds but refocused from the
personal to the social level, so control is not lost but shifted
from personal norms and values to those that define the rele-
vant social category. This means that the behavior of crowds
will vary as a function of what categories are involved. The
norms and values - and hence the actions - of, say, a crowd of
environmentalist protestors will be different from those of a
crowd of soccer fans which, in turn will be different from those
of a Catholic crowd welcoming the Pope. The process of con-
formity to group standards may be general, but the behaviors it
leads to will always depend upon contextually relevant belief
systems.
There are two important implications of this process which
need to be stressed. The one is that, insofar as crowd members
act in terms of their social identities and that social identities
invoke category-based belief systems, here is a mechanism for
understanding how the behaviors of specific crowds relate to
broad conceptions of society - a key requirement for any
adequate model of crowd action. The other is that this process
is not mechanical or routinized. Crowd situations are typically
either novel or uncertain - and made all the more so by the fact
that they are interactions in which the behavior of the other
side is unpredictable (a point developed in much more detail
in the next section). Consequently, one cannot just apply
preformed rules and norms of action. Rather, one has to
determine what the general category means in the concrete
situation.
The creation of situationally appropriate norms can happen
through a process of induction. That is, people infer group
norms from the behaviors of those who are unambiguously
ingroup members - as long, that is, that these behaviors are
consonant with the broad terms of group identity. So, for
instance, in a British urban riot, when someone threw a stone
at the police, it was followed by a hail of stones. However,
when someone stoned a bus, not only did others not follow,
but they actively intervened to prevent further stoning. Or,
equally, when someone smashed the windows of a bank (one
of the external institutions seen to keep the local community in
penury) others joined in, but when the windows of a local shop
were smashed, others spontaneously defended the shop from
looting. This is very different from the process of contagion.
Crowd Psychology 635
It also elaborates upon ENT, both by showing how norm
creation can be rapid and responsive to events and by showing
that, while there is space for variability in the interpretation of
what is appropriate, there are clear limits set by social identity
to the norms that form in crowd events.
This is not to say that the deliberative processes described
by emergent norm theorists are irrelevant, merely that they are
not always necessary. Where there is time - particularly at the
start of events - people do mill around, listening to others and
trying to determine what they should be doing. But even here,
social identity is cmcial in framing the emergence of norms.
The keynoter who is influential (i.e., the leader) is the one
who can best represent his or her proposals as instantiations
of 'who we are' and 'what we believe in' - of group identity that
is. Conversely, the basis on which members deliberate is not
simply 'what should we do here,' but rather 'what should we
do here as environmentalists' (or whatever the situationally
relevant category might be). Altogether, the task of keynoters/
leaders and group members is jointly to develop a situationally
appropriate elaboration of a preexisting identity.
The Relational Transformation
In crowds, the relationships between individual members
undergo a profound shift. This is not simply as a function
of the fact that crowd members identify with a social category.
It occurs to the extent that people share a common identity
with others in the crowd and also that they are aware of this
commonality. Under such conditions, fellow crowd members
shift from being 'other' at the individual level to being part of
a shared collective self. Hence, personal boundaries between
people are dissolved and the boundary shifts to that between
ingroup ('us') and outgroup ('them').
On the one hand, then, the sense of difference to others in
the group is dissolved. We expect them to share similar norms,
values, and beliefs with ourselves. We' will all see things the
same way and strive for the same goals. As a consequence, we
will both trust and respect our fellow ingroup members more
and we will look for agreement rather than disagreement with
them. We will also feel more confident and validated in our
own beliefs. Unlike everyday life where venturing an opinion,
especially a strong stance, may evoke the disagreement, disap-
proval, even rejection by the other, in crowds, we can express
ourselves in the full expectation that others will support our
views. More than that, for those in the midst of a crowd, they
will be surrounded by others who chant the same chants as
them, who look and dress like them, who even smell like them:
their sensuous universe is one that affirms their identity.
On the other hand, our sense of 'interest' and of 'fate' is
extended to encompass other group members. That is, our self
being the collective self, then 'self-interest' becomes a matter of
advancing the interest of the group as a whole, and a detriment
to any group member becomes a detriment to our (collective)
selves. Correspondingly, the act of 'self-defense' is a matter of
defending not only one's own body from injury but also that
of any other group member from attack, especially where that
comes from the outgroup. Often, in crowds, people will put
themselves at considerable risk to help and support fellow
group members who, for instance, are under threat of being
attacked or seized by the police.
In combination, these various consequences of the rela-
tional transformation that occurs in crowds create the condi-
tions where people can coordinate their actions. People are
able to integrate their efforts, to support one another, and to
initiate action in the confidence that they will be supported by
others. Whereas, outside the crowd, people may feel isolated
and unable to enact their beliefs due to the power of others, in
the crowd they feel united and able to overcome any potential
resistance of others. That is, they feel empowered as group
members. In this sense, Le Bon provides an important insight
about crowds. However, contra Le Bon, this empowerment is
highly focussed. It is a power to translate the norms and values
of the group into objective social practices. Accordingly, Drury
and Reicher refer to this process as collective self-objectification.
To paraphrase the historian of the French Revolution, Georges
Lefebvre, perhaps it is only in crowds that people are able to
become the subjects of history.
The Affective Transformation
One criticism of the early social identity approach to crowd
psychology was that it was a very cold approach which empha-
sized the mindful and meaningful nature of crowd action but
which neglected the emotional dimension. Empirically, this
would be to leave out one of the most striking things about
crowd events: they are often highly passionate affairs. People
who watch sports at a stadium scream and shout and chant far
more than at home or alone. Any model that leaves this out is
necessarily deficient. Conceptually, this would be to buy into
the Le Bonian duality between reason and emotion (one which
is widespread within and beyond psychology) and simply
reverse the polarity: from all emotion and no reason to all
reason and no emotion.
There are many bases for the strength of emotional experi-
ence in crowds. Many of the relevant processes have been
clarified by recent research in the area of collective emotions.
That is, when people identify highly with a group, they can feel
things not only because of what they have done or what was
done to them as individuals, but because what their group has
done or what was done to members of their group. Thus, for
instance, acts by outgroups (such as the police) which are seen
either to attack the group or else to deny the group its legiti-
mate rights evoke strong feelings of anger. However, when
talking about their feelings, crowd members lay particular
stress on the importance of empowerment and collective self-
objectification discussed previously. After living one's life in a
world made by others and having to trim one's behavior to
what is acceptable to others, at last, in crowds, people can
shape the world in their own terms. This is not only an occa-
sion for joy, it also leads to increased commitment to the group
and even to increased individual well-being.
Once again, then, we can find value in Le Bon's work - in
this instance, his emphasis on passion in crowds. But the
explanation of this passion is the precise opposite of Le
Bon's. It does not derive from a loss of mind, a loss of judge-
ment, and a loss of individual agency. Rather it flows from the
fact that, perhaps uniquely in our experience, ordinary people
are able to make their own history. Insofar as we define our-
selves in terms of our group membership and that, through
joint action, our groups are able to impact on social reality,
636 Crowd Psychology
then we feel ourselves to be agents. And because crowds make
us agents, we are passionate about them.
The Elaborated Social Identity Model and
Intergroup Dynamics in Crowds
The focus of the social identity model is predominantly upon
intragroup relations in crowds: the model analyzes how indi-
viduals and interindividual relations are transformed within
crowds and how this impacts on what they do and feel. And
yet, throughout the analysis, there is an implicit recognition of
the importance of the outgroup - whether it be a matter of the
salient presence of an outgroup transforming a disparate set of
people into a psychological crowd with shared identity (as in
the example of the train carriage), of the perceived illegitimacy
of outgroup action inspiring crowd solidarity and crowd reac-
tion, or else of crowds empowering their members and hence
transforming relations with the outgroup.
Nonetheless, there remains the danger that SIM, like pre-
ceding models of crowd behavior, will seek to analyze crowd
action without (explicitly) recognizing that crowd events are
typically interactions between at least two groups (one group
of sports fans against another, the police against rioters, com-
pany security guards against strikers) and that it will be impos-
sible to fully understand the behavior of any one party without
looking at both and the ways each impacts on the other. To
make the same point slightly differently, if the core problem
with classic crowd psychology lies with decontextualizing
crowd action, then it can be argued that SIM links crowd action
to the broader context (by rooting crowd action in broad
category-based beliefs, norms, and values) but that it neglects
the immediate interactive context.
The elaborated SIM (ESIM) of crowd action extends SIM by
explicitly analyzing the intergroup dynamics of crowd events
and placing intragroup dynamics in this interactive context.
Moreover, through its focus on intergroup dynamics, ESIM lays
more stress on change processes in crowds than the earlier
social identity work.
ESIM derives from a common pattern of action observed
across a number of crowd events, including such different
groups as students, environmental protesters, antitax protes-
tors, and football fans. In each case, events started with heter-
ogenous crowds comprising multiple psychological groups,
most of whom were opposed to conflict and violence. How-
ever, all those present were perceived as at least a potential
threat by the police who then acted to contain the crowd and
prevent them from progressing as they wished. This common
experience of being denied what they perceived to be their
legitimate rights led to the formation of a single psychological
group united around opposition to the police. Moreover, this
unified crowd felt thereby empowered to contest police
actions, which in turn validated the police's original concep-
tion of the crowd as dangerous. In this way, a spiral of escalat-
ing conflict occurred, often resulting in overt violence.
In order to explain this pattern, ESIM conceptualizes social
identity as a representation of the nature of social relations
in our world and where we stand within them along with the
behaviors that are proper and possible given that position.
Many protestors, for instance, see themselves as living in a
liberal democratic society where the police are neutral arbiters
who uphold the social order. They themselves may be opposed
to certain other groups but they do not view the police as an
outgroup. Most of the time, the different parties to a crowd
event share the same representations of one another. Hence,
their interactions serve to confirm and stabilize the views of
participants.
However, under certain (rare) circumstances, there is an
asymmetry, such as when protestors do not see the police as
an outgroup but the police see all protestors as an outgroup (in
part because of the continuing dominance of models which
portray all crowd members as dangerous). What is more, when
one of these parties (the police) has the power to act upon its
perception (by erecting cordons and stopping crowd members
from doing what they want), then the perception of one group
becomes an experiential reality for the other. Protestors experience
being positioned as 'oppositional.' And, from this experience, they
begin to reconceptualize their relationship to the police and
hence their own identity. That is, they begin to see themselves as
oppositional.
This shift of identification sets in train a whole series of
associated changes. First, new norms and values emerge. Those
behaviors that make sense in a world where the police and state
are neutral - persuasion and patience - no longer make sense
where the police and state stand against one. Hence, those
voices (leaders/keynoters) calling for conflict who were earlier
shunned are listened to more, and there is greater potential for
more than a small minority to become involved in conflict. In
other words, outgroup actions alter the relative success of con-
tending sources of influence in the ingroup. The irony, often, is
that the police, fearing violence, act in ways that make those
calling for violence more impactful.
Second, as crowd members redefine their own identities,
their relations to others outside as well as inside the actual
crowd may shift. Thus, in seeing themselves as oppositional,
a common ingroup bond may be formed with other oppo-
sitional groups which, previously, had been rejected. For
instance, in one antiroads campaign, an early division between
local protestors who simply wanted to preserve their amenities
and committed environmentalists who were opposed to the
overall roads program was overcome after the police were
equally hostile to all involved in trying to stop the tree on
the town green being felled. But the 'locals' not only came to
see themselves as part of a wider protest and hence included
the environmentalists as ingroup, they also started to see
groups like striking miners, black people, and feminists as
part of a common cause against injustice.
Third, even the goals and the concept of success of crowd
members can shift. To continue with the aforementioned
example, locals originally conceptualized success as saving
their green and its tree. But as they began to see the police
and state as arraigned against them and in favor of the roads
lobby, they looked at the act of simply standing up against the
police and publicizing the nature of what they were up against
as an end in itself. Even if the tree was cut down, in this latter
sense, the protest had been a success, it was celebrated as such
and it led to increased commitment to future protests.
Theoretically, the core message here is that crowd identities
and hence crowd action are not simple givens but rather pro-
ducts of intergroup interaction. It is important to recognize this
Crowd Psychology 637
in order to understand how identities are reproduced and
come to be seen as stable or even natural and in understanding
the conditions where rupture and change occur. Practically, the
core message is that the interventions of the authorities,
the police in particular, do not simply serve to contain collec-
tive violence. They are a critical part of the escalation or the de-
escalation of violence. In recent years, ESIM has been applied
practically to design forms of police intervention which distin-
guish between groups in the crowd and which seek to facilitate
those with lawful intentions. In this way, the influence of those
promoting violence can be reduced and the prospects of peace-
ful democratic protest maximized.
Conclusion
I began this piece by noting how crowds have long been seen as
aberrant - a social aberration caused by psychological aberra-
tion - and by suggesting that this has prevented us from
learning about both society and psychology from the study of
crowds. By now, it should be clear that one cannot understand
crowd action if one divorces it from its social context. Even
if crowd action is occasionally extreme, this reflects long-
standing patterns of social belief about what is acceptable
and unacceptable. And crowd action does more than reflect
existing social beliefs and social identities. Crowds play a criti-
cal role in forming the identities of participants and creat-
ing new unexpected identities in the crucible of collective
interaction.
But even here, we are only beginning to appreciate the
wider importance of crowds. There are two ways, at least, in
which crowd psychology has much to contribute. First, at a
theoretical level, crowds provide a particularly clear example of
how intergroup dynamics frame intragroup relations. How-
ever, this is not because crowds are unique but rather because
(a) in crowds, the intergroup dimension is particularly obvious
since the police are physically present, whereas in everyday life,
the impact of outgroups may be more subtle (enshrined, for
instance, in the ways that institutions are structured) and less
easy to see, and (b) the copresence of both ingroup members
and outgroup members in crowds renders the impact of the
inter- on the intragroup immediate, general, and hence easy
to see. However, it is reasonable to suppose that similar pro-
cesses occur in everyday life, albeit more distributed through
multiple encounters between individuals or small numbers of
ingroup and outgroup members, and then disseminated slowly
through the telling of stories which are disseminated through
social networks. In this way, ESIM may provide a model
for intergroup relations in general, and the practical insights
of the model may help us understand the conditions under
which members of certain groups disengage from authority
and support conflict against authority.
Second, if crowds are important to the formation of wider
social identities, this need not be limited simply to those who
participate in them. For crowds are often the embodied reality
of broad social categories that are abstractions - groups
like nations, classes, religions, 'races,' - which are too large
for all the members to congregate and which are therefore, in
Anderson's terms, 'imagined communities.' What happens
in crowd events is therefore often emblematic for all members
of the relevant categories whether they themselves were there
or not and can therefore impact on the identities of all. Such
events tell them about the world, about who they are, and how
they are treated in society. What is more, crowd events are high-
profile. They are likely to be covered on the television, in the
papers, and hence be visible to the broader category member-
ship. In this way, for instance, the US riots of the 1960s
impacted on all black people in the United States just like the
British riots of the 1980s affected the overall views of black
people in the United Kingdom.
To sum up, crowd psychology has come a long way over the
last century or so. There is still a long way to go. But at least we
now recognize that when many members of a social group
assemble together, then what they do has much to tell us
about our society and where it is going. At least the value of
pursuing crowd studies is more apparent than it once was.
See also: Group Dynamics; Individualism; Leadership; Persuasion.
Further Reading
Barrows S (1981) Distorting Mirrors. New Haven, CT: Yale University Press.
Davis NZ (1978) The rites of violence: Religious riot in sixteenth century France.
Past and Present 59: 51-91 .
Drury J and Reicher SD (2005) Explaining enduring empowerment: A comparative study
of collective action and psychological outcomes. European Journal of Social
Psychology 35: 35-58.
Drury J and Reicher SD (2009) Collective psychological empowerment as a model of
social change: Researching crowds and power. Journal of Social Issues
65: 707-725.
Le Bon G (1895/1947) The Crowd: A Study of the Popular Mind. London: Ernest Benn.
McPhail C (1 991 ) The Myth of the Madding Crowd. Social Institutions and Social
Change. New York: Aldine De Gruyter.
Nye RA (1975) The Origins of Crowd Psychology. London: Sage.
Reicher SD (1984) The St. Pauls riot: An explanation of the limits of crowd action in
terms of a social identity model. European Journal of Social Psychology 14: 1-21 .
Reicher SD (2001) Crowd dynamics. In: Hogg M and Tindale S (eds.) Blackwell
Handbook of Social Psychology: Group Processes. Oxford: Blackwell.
Reicher SD, Spears R, and Postmes T (1995) A social identity model of deindividuation
phenomena. European Review of Social Psychology 6: 1 61 -1 98.
Rude G (1964) The Crowd in History. New York: Wiley.
Stott C, Hutchison P, and Drury J (2001) ‘Hooligans’ abroad? Inter-group dynamics,
social identity and participation in collective ‘disorder’ at the 1998 World Cup
Finals. British Journal of Social Psychology 40: 359-384.
Thompson EP (1971) The moral economy of the English crowd in the eighteenth
century. Past and Present 50: 76-136.
Turner RH and Kilian LM (1972) Collective Behaviour, 2nd edn. Englewood Cliffs, NJ:
Prentice-Hall.
US Riot Commission (1 968) Report of the National Advisory Commission on Civil
Disorders. New York: Bantam.
Crowding: Effects on Health and Behavior
S J Lepore, Temple University, Philadelphia, PA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Community density Number of people in a community
area or proportion of people per available dwellings or space
in a community.
Crowding A negative psychological state that results from
perceiving that there are too many people present in the
available space.
Density Number of people in a specified amount of space
or proportion of people per available space.
Household density Number of people in a residential
dwelling or proportion of people per available rooms or
space in a residential dwelling.
Distinction Between Density and Crowding
It is important to distinguish between the subjective, psycholog-
ical experience of crowding and the objective, environmental
source of the crowding experience: high-population density.
Daniel Stokols has noted that density is a property of the
physical environment whereas crowding is primarily a psycho-
logical experience. The negative effects of high density on
human health and behavior are strongest and most prevalent
when individuals are uncomfortable or stiessed by the high
density. That is, high density might only affect health and
behavior when individuals appraise the high-density setting as
being crowded. Density is an important antecedent to the expe-
rience of crowding but is often not sufficient to explain every-
one's feeling of crowding or a particular individual's experience
of crowding in different settings or at different times.
Density is typically measured by calculating a ratio score
of the number of people to a given amount of space. There
are two broad types of density studied by social scientists,
household density and community density (see Table 1). House-
hold density can be determined in several ways. For example,
one could calculate the number of people per room or square
footage within a household. According to the US census, house-
holds with more than 1.0 persons-per-room are overpopulated.
Absolute number of people in a household in a residence also
can be used as indicators of objective levels of household crowd-
ing. Community density measures reflect the amount of space
for a given population over a wider space than an individual's
residence. Community density can be determined by calculating
the number of people per acre, people per square mile, or
people per census tract. Occasionally, community density will
be defined as the ratio of dwellings or buildings to a given area
or the total number of dwellings in a community area.
Public, health officials, city planners, housing developers,
and policy analysts in charge of setting housing standards
are particularly interested in understanding the effects of
household and community density levels on human health
and behavior. In general, ratio measures of density are better
predictors of health, especially mental health, than are abso-
lute measures, such as number of persons or number of rooms.
In addition, measures of density in a dwelling, such as persons
per room, are better predictors of individual health and behav-
ior problems than are community measures of density, such as
dwellings per square mile or persons per acre. These latter two
findings may be explained by the fact that individuals will tend
to have greater difficulty escaping from or avoiding unwanted
interactions with other people when they are inside a highly
populated dwelling with little available space than when they
are in the outside world or in a highly populated dwelling with
lots of space. The different behavioral and health effects of
density in a dwelling versus density outside of a dwelling are
discussed in more detail below.
Role of Social and Personal Characteristics
The correspondence between density and the psychological
experience of crowding often depends upon the individual
and the social situation. For example, people can be exposed
to high levels of density in a bustling city street, a cramped
apartment, a sporting event, a concert, a supermarket, or a
political rally. Some of these crowded situations are exciting
and inviting and some are threatening and foreboding. While
many people would concede that the throngs of people cheer-
ing at a football stadium contribute to the excitement of the
sport, few would agree that the clanging and banging of shop-
ping carts and waiting in long lines to make a purchase in a
crowded supermarket are enjoyable experiences.
Why do different situations evoke different experiences of
and feelings toward high density? There are, of course, impor-
tant differences between high density in different settings, like
stadiums, or supermarkets. For example, in a stadium the cheers
and enthusiasm of a crowd can be stimulating and help a
spectator to have a good time. In the supermarket, the presence
of many people can interfere with or constrain a shopper's
movement through the supermarket and his or her ability
to finish shopping. When high density thwarts goal-directed
behavior, it is more likely to be experienced as sUessful and
crowded than when it does not block goal-directed behaviors.
Interference with goal-directed behaviors can diminish indivi-
duals' actual and perceived control over their environment. Lack
of conUol over the environment can cause some people to feel
psychologically disUessed. The role of conUol in explaining the
negative effects of crowding is discussed in more detail below.
There are also wide differences in peoples' reactions to high
density. For example, men and women appear to experience
638
Crowding: Effects on Health and Behavior 639
Table 1 Different environmental sources of crowding
Types of household density
Number of people per household area 3
Number of rooms per household square footage per household
Number of persons per room in a household
Types of community density
Number of residents per community area*
Number of households per community area
Number of commercial buildings per residential community area
Number of multiunit housing structures per community area
Proportion of households with more than one person per room in a
community area
Proportion of households with five or more persons in a community area
Number of persons per 10000ft 2 of residential space in a
community area
Number of persons living on a street per 1000ft
^Household areas can be measured in square footage or meters.
*Community areas can be measured in acres, square miles, or by census tract.
high density quite differently. When men and women are
required to interact in small groups, men are more uncomfort-
able and less social than are women. Thus, men appear to
prefer a larger amount of personal space or physical distance
between themselves and other people than do women. Differ-
ent cultural groups also may experience density differently.
For example, American college students appear to be less toler-
ant of residential crowding than are adult males in India or
Turkish college students. Many Chinese and Japanese families
appear to be relatively unaffected by living in very high-density
homes. Among different American ethnic groups, household
density tends to have a stronger negative effect on the mental
health and social relations of black Americans than white
Americans, and only a weak effect on Americans of Hispanic
descent. On the other hand, researchers also have found
equally strong positive associations between household
density on psychological distress across diverse racial and eth-
nic groups, including African Americans, Anglo Americans,
Vietnamese Americans, and Mexican Americans.
Why do individuals have unique reactions to similar levels
of density? One explanation is that people develop methods of
coping with the crowding. For example, gender differences in
reaction to high density might be explained by adaptation
to different levels of closeness in interpersonal interactions.
In comparison with males, females might be socialized to
expect and to engage in contact with others in closer physical
proximity. As they mature, males might become more accus-
tomed than females to having large interpersonal distances
between themselves and others. Thus, experience with high-
density settings potentially can diminish the negative psycho-
logical experience of crowding because individuals can learn
how to cope with the undesirable aspects of high density or
they become accustomed to close physical contact with others.
Effects of Crowding on Health and Behavior
There has been a long history of interest in the effects of
crowding on human health and behavior. The spread of vari-
ous diseases and social pathologies is often attributed to the
vast numbers of people living in urban areas. High-population
density is often cited as a reason for the very high rates of
mental and physical health problems and various deviant
social behaviors found in cities. Living in close proximity to
others can obviously facilitate the spread of infectious diseases.
However, transmission of infection through close contact with
others is not the only way in which high density contributes to
social and biological health problems.
Observations of the negative effects of overpopulation in
nonhuman animal species have fueled some of the concern
over the effects of high density and crowding in human popu-
lations. Much of what has been learned about density and
disease processes has been through experimentation with ani-
mals because of ethical constraints in experimenting with
humans. However, even in animal populations it is difficult
to ascertain the exact causes for the negative effects of density
on health and behavior. Take, for example, the observation
that high density is associated with higher rates of mortality
in many animal species. Many animal species exhibit cycles of
population escalation followed by a sudden and tremendous
mortality, or a 'population crash.' Different pathways have
been identified that may link density to mortality. For example,
food shortages or the rapid spread of disease, as in plagues
and other epidemics, could explain population crashes in high-
density settings. However, starvation and contagious disease
do not fully explain the effects of high-population density on
mortality. One group of scientists observed that deer living
on an island multiplied quite rapidly until their population
reached about one deer per acre, then their mortality rate
skyrocketed. The deer population crashed, even though the
deer had plenty of food and water and showed no signs of
contagious disease spread. Examinations of the deer's internal
organs, however, revealed signs of stress-related disease pro-
cesses, which suggested another explanation of why high den-
sity can increase mortality: social stress.
Social stress can be caused by uncontrollable, threatening,
unwanted, or otherwise negative social contacts and interac-
tions with other organisms in an environment. High density
can increase social stress, which can induce fighting between
animals, interfere with reproductive behaviors, and cause
unhealthy metabolic disturbances. Under high-density condi-
tions, rodents' reproduction rates drop, cannibalism and
deviant sexual behavior increase, and other social and biologi-
cal pathologies all increase. In some animal studies, higher
mortality among animals in high-density pens appears to be
caused by overactive adrenal glandular systems. Many of the
biological pathologies manifest in crowded rodents are similar
to those observed in many species of animals after they have
been exposed to noxious environmental stimuli, or stressors.
Thus, it appears that the stress-related disease processes that
might be caused by high-population density could contribute
to death and illness in different animal species.
It is tempting to draw analogies between overpopulated
animal populations and overpopulation that occurs in human
settlements. For example, birth and mortality rates of high-
density human communities could be compared to similar
outcomes in high-density rat colonies. However, many out-
comes, such as criminal behavior in humans versus aggression
in rats, are not so directly comparable. In addition, it is more
difficult to prove that density is the cause of deviant behaviors
in humans than it is to do so in animals. In animal research,
640 Crowding: Effects on Health and Behavior
scientists can control the effects of external factors other than
high density that could influence the behaviors of crowded
animals. Researchers studying human crowding can seldom
control other factors, such as poverty and noise, which tend
to accompany high density and influence humans' health and
behaviors independent of density. In social science terms, the
uncontrollable factors could cause a 'spurious,' or illusory,
relation between density and human health and behaviors.
Some researchers use statistical techniques to attempt to exam-
ine the effects of density independent of other social factors.
The problem with making such statistical judgments is that
researchers never know for certain whether they have identified
all possible factors that could influence both density and the
outcome of interest. These points should be kept in mind when
reading the next section on the evidence relating high density
to human health and behaviors.
Chronic Versus Acute Crowding
The effects of high-population density on humans do not
appear to be as dramatic as in animals. This is partly due to
the sophisticated and complex ways which humans experience
and adapt to noxious environmental stimuli like high density.
For example, humans can hoard food supplies or increase food
production to avoid starvation under high-population condi-
tions. Humans also can modify their environments, perhaps
through scheduling or architectural interventions, to minimize
social stress in high-density settings. Nevertheless, social
research has revealed some relations between density and vari-
ous health and social problems in humans.
In discussing the effects of crowding on humans, it is
important to distinguish between chronic crowding and acute
crowding. Chronic crowding takes place in settings where peo-
ple tend to spend much of their time, like work places, resi-
dential settings, or institutional settings such as dormitories,
prisons, and military barracks. Acute crowding takes place in
settings where people tend to spend very little time, like stores,
elevators, sidewalks, restaurants, theaters, stadiums, and other
public places. The effects of acute crowding also have been
examined by researchers in laboratory settings modified to
represent different levels of density.
Effects of chronic community crowding on social pathology
Social pathology can be defined as those phenomena that
contribute to the demise of a society, typically by reducing
its population, but also by disrupting its institutions and
social relations. Thus, high rates of crime, mortality, accidents,
disease, and divorce are indicators of social pathology. In the
minds of many, social pathologies are linked to large cities,
where they seem to proliferate and concentrate. Because large
cities are both highly populated and full of social pathology,
scientists have attempted to determine whether community
crowding is at the root of the pathology evident in cities.
Interest in the relation between community density and
pathology has been apparent since at least the end of the
nineteenth century. Along with the industrial revolution
came a rapid growth in cities throughout the western world.
Some social theorists thought that the diversity of people,
the personal anonymity, and high levels of individual auton-
omy existing between people in large cities would lead to
psychological distress and anomie. In contrast, people from
small towns and agrarian societies were expected to have
richer social lives and greater morale because of familiarity
and close interaction with similar others. Other social theorists
argued that the high density of cities would expose people
to overwhelming amounts of stimulation. In response to the
stimulus overload, city-people would socially withdraw. Social
withdrawal could be a strategy for reducing stimulus overload.
By reducing concern for others and by interacting at a superfi-
cial level, there would be fewer stimulus inputs to cope with
in day-to-day life. However, there would naturally be social
costs if everyone acted this way, including apathy, frustration,
conflict, and competition.
Contemporary social scientists pursue many of the same
questions regarding community crowding and pathology as
did their counterparts from a hundred years ago. Typically,
researchers researching on crowding investigate whether areas
with high levels of community density also have high concen-
trations of social, psychological, and biological pathologies
or problems. Community population density has been studied
in relation to rates of death, infant mortality, perinatal mortal-
ity, accidental death, suicide, tuberculosis, venereal disease,
mental hospitalization, birth, illegitimate birth, juvenile delin-
quency, imprisonment, crimes, public welfare, admissions to
general hospitals, and divorce. The current evidence suggests
that there is little or no relation between population density
and major indicators of social pathology, such as mortality,
crime, and juvenile delinquency. One research group observed
that a higher ratio of persons per acre was associated with
slightly elevated rates of mortality, fertility, juvenile delin-
quency, admissions to mental hospitals, and public assistance.
However, the researchers also noted that certain ethnic and
economic groups were over represented in the high-density
areas. Thus, factors such as poverty, rather than density, could
have caused the higher rates of pathology observed among
individuals living in high-density areas. Indeed, when the
researchers controlled for the effects of social class and ethnic
background on the pathological outcomes, the relations
between density and the outcomes disappeared.
On the other hand, it is possible that some community-
crowding studies have underestimated the effects of high
density on human pathology. Aggregate measures of density,
such as persons per square mile, and aggregate measures
of pathology, such as number of hospital admissions, do not
precisely reveal the exposure to high density or its effects on
individuals. For example, a person living in a high-density com-
munity might spend most of his or her waking hours at a job in
a community that has a low level of density. Or, a person from a
low-density suburb might work all day in a high-density city.
The actual exposure of these respective individuals to high den-
sity is different than what one would expect based on the density
of their communities. In one instance, the negative effects of
living in a high-density community could be underestimated.
In the other instance, the benefits of living in a low-density
community could be overestimated. If there are many of
these peculiar cases in a study population, then an aggregate
measure of community density will not be a good estimate
of exposure to crowding. Nor would such a measure be useful
for examining the effects of crowding on human health and
behavior. There are also problems with aggregate measures of
Crowding: Effects on Health and Behavior 641
pathology. The principle problem is that data on social pathol-
ogy originate from official public records, which can be incom-
plete and inaccurate.
To make matters more complicated, researchers can never
know whether the relations between density and pathology are
overestimated or underestimated when analyzing aggregate
data. That is, the data errors caused by using aggregate mea-
sures could make the effects of density on pathology look
stronger or weaker than they are in reality. One way around
the problems associated with aggregate data is to study the
effects of high density on individuals rather than on whole
communities. That is, one could carefully measure individuals'
exposure to density and their health and behaviors. This is
usually done by surveying individuals about the levels of den-
sity in their households and about their health, behavior, and
psychological well-being. Findings from this type of research
are discussed in the next section.
Effects of chronic household crowding on health
and behavior
Household crowding stems from high density in the residential
environment. Residential environments include individuals'
homes and apartments, as well as institutional settings such
as prisons, dormitories, and military bases. Household density
appears to have a wide range of effects on human health,
behavior, and general well-being. Research in institutional set-
tings such as prisons and dormitories is especially compelling
because it helps to control for selection biases: prisoners and
college students typically cannot choose or control whether
they will be in a more or less crowded room.
Prisoners in high-density cells, for instance, report more
negative moods, discomfort, and illness symptoms than those
in single-person cells or relatively low-density cells. Disciplinary
problems, psychiatric commitment rates, suicide rates, and
death rates also appear to increase in prison populations that
grow in size without increases in the size of prison facilities.
Crowding also is a frequent problem in student populations.
This often occurs because of a shortage of desirable housing
near colleges or because students often double-up in apartments
to save money on rent. In comparison with students in low-
density dormitories or off-campus apartments, students in
high-density residences feel more crowded, have more unwanted
social contact and interactions, have more frequent negative
moods, are less happy, and try to avoid interactions by socially
withdrawing. In comparison with their uncrowded counter-
parts, crowded students are less sensitive to others' needs, less
willing to help others, and less aware that others are available to
provide emotional support or help to them when they need it.
People living in high-density residences can become insen-
sitive to social cues even when they are not in high-density
setting. That is, the social insensitivity cultivated in high-
density environments can carry over into low-density settings.
College students who are withdrawn and insensitive to social
cues in their high-density residences also act this way in low-
density laboratory settings. Crowded students exhibit their
withdrawn behaviors by sitting far away from others, not initi-
ating conversation, making little eye contact with others, and
not being helpful to others in need.
High household density in noninstitutional settings also
results in antisocial behaviors and complaints of excessive
social interaction. High household density is also related to
increased negative mood and symptoms of depression and
anxiety among adults. People from high-density households
tend to have fewer friends and greater difficulty getting along
with their neighbors than do people from relatively low-
density homes. Parents tend to interact less with their young
when they live in high-density homes than when they live in
relatively low-density homes.
Children appear to be more negatively affected by high
density than are adults, partly because they have less control
over their environment than do adults. In comparison with
children from low-density homes, those from high-density
homes tend to have more behavioral problems in school, more
anxiety, greater distractibility, more conflicts, lower achievement
motivation, and poorer verbal abilities. However, caution must
be applied when interpreting these results. As discussed above,
density often accompanies other environmental conditions that
could influence children's behaviors. Noise, for example, is likely
to be greater in high-density households than in low-density
households; and noise can interfere with children's attention,
hearing, and learning abilities. Uncoupling the effects of den-
sity from those of noise is nearly impossible in naturalistic
settings. In studying the effects of crowding on children it is
also important to consider the role of other adults, particularly
parent, in the environment. Research suggests that parents
in crowded households are less responsive to their children
relative to parents in less crowded households.
Earlier in this article, it was noted that social factors could
influence whether an individual would perceive a particular
setting as crowded. The social environment also can have a
strong influence on the relation between crowding and psy-
chological distress symptoms. For example, students living
in high-density households who have frequent hassles from
roommates are more likely to be psychologically distressed
than are students living in high-density households with rela-
tively few hassles. Students living in low-density homes do not
appear to be adversely affected by social hassles. People living
in high-density households may be particularly distressed by
roommate hassles because it is more difficult to avoid or escape
from the hassles in a high-density home than in a low-density
home. In contrast to the effects of social hassles, positive social
relations can counteract the negative psychological effects of
high-density living situations. For example, college students
living in high-density households who have supportive room-
mates are less likely to be distressed than are students living
in high-density households with relatively unsupportive room-
mates or students living in low-density households. However,
as mentioned earlier, chronic household crowding can under-
mine socially supportive relations because people become
withdrawn and insensitive to social cues. Thus, although posi-
tive social relations may be beneficial to crowded individuals,
such relations may be rare in households that are chronically
overpopulated.
Effects of acute crowding on physiology and behavior
Acute crowding has been studied in scientific laboratories
and in natural settings, like trains and elevators. In the animal
studies discussed above, it was noted that many animals
living in high-density populations showed biological signs of
stress, such as enlarged adrenal glands. Elevated and sustained
642 Crowding: Effects on Health and Behavior
physiological arousal is another sign of stress that is commonly
observed in organisms under stress. In high-density laboratory
settings, arousal has been exhibited in human subjects using
many different measures. People exposed to acute crowding in
laboratories exhibit increased perspiration, skin conductance,
and blood pressure. Passengers in crowded commuter trains
exhibit increases in blood levels of adrenaline. Interestingly,
passengers who board a train when it is already near capacity
are more negatively affected by the crowded conditions than
passengers who board early, when the train is nearly empty.
Even though the latter passengers ride the train longer, they
had more choice, or control, over where they could sit on the
train than passengers who board the train when it is already
loaded with other passengers.
It should be noted that although acute exposure to high-
density settings usually increases physiological arousal, it is not
clear whether there are negative health consequences of the
increased arousal. It is possible that elevated and prolonged
arousal can wear down the body's defenses against illness and
generate illness itself, such as hypertension or ulcers. However,
there is not enough evidence at this point to state unequivo-
cally whether the arousing effects of high density have health
implications.
Impaired task performance is another common side-effect
of acute exposure to high density. Mostly, this research has been
conducted in laboratory settings that manipulate individuals'
exposure to high or low density and then measure performance
on problem-solving tasks or assess one's ability to concentrate
or persist on a task. Although high density does not seem to
interfere with performance on simple tasks, it does diminish
complex task performance. It also appears that individuals are
more easily distracted and are less persistent at completing
challenging tasks under high- versus low-density conditions.
Short-term exposure to high density produces many of the
same withdrawal behaviors observed in people chronically
exposed to high density. It is not uncommon, for example,
for people on crowded subways to read newspapers and books
as a way of avoiding interaction with others. In laboratory
settings, crowded people are more likely to leave the settings,
increase social distance, withdraw from social interactions,
increase defensive posturing, and reduce eye contact than
uncrowded people. Another way to maintain space from others
is to be aggressive and threatening. Several studies have shown
increased competition and aggression between individuals
in high-density settings. Interestingly, crowding is more likely
to evoke aggression in men than in women and when resources
in the environment are scarce rather than plentiful. The adap-
tive value of withdrawal and aggression in high-density settings
is not completely understood, but it does appear that these
social behaviors may help people to cope with crowding.
For example, social withdrawal might be a way of minimizing
physiological arousal by avoiding unwanted social interactions
and the excessive social stimulation that is common in high-
density settings.
Explaining the Negative Effects of Crowding
Several theoretical explanations of the effects of crowding on
human health and behavior have been alluded to throughout
this article. Below we discuss four of the more prominent
of these theories: behavioral constraint, control, social with-
drawal, and overload/arousal theories.
Behavioral Constraint
According to this theory, high density interferes with indi-
viduals' goal obtainment by restricting or inhibiting their
movements and behaviors. The diminished freedom makes
the high density noxious and undesirable. When people are
in high-density settings that do not thwart their goal-directed
behaviors, they tend to be less negatively affected by the high
density than when their goals are thwarted. Imagine, for
instance, two groups of individuals performing a task in a
crowded room, but one group has to complete the task while
sitting still and the other group has to complete the task
while moving around the room. It is more likely that crowding
will interfere with the task performance of the moving group
than of the still group because the crowded group will be
coping with the task and the constraints on their movements
caused by the crowding.
Behavioral constraints do not refer only to restrictions in
bodily movement. Sometimes high density can create resource
shortages, such as food shortages, which constrain behavioral
choices. That is, density can restrict access to valued resources.
In the case of food shortages, behaviors such as eating might
be inhibited. In addition, people in such situations might act
aggressively to get valued resources. Finally, it should be noted,
that high density often can have negative effects on mood
and performance because people perceive that there are behav-
ioral constraints in high-density environments. That is, simply
believing that high density can limit one's behaviors or access
to valued resources is sufficient to diminish task performance
or increase discomfort in high-density settings.
Control
Limits to behavioral freedom also can be construed as limits in
personal control over the self and the environment. Control
models of crowding hypothesize that high density is undesir-
able and harmful because it renders the environment more
unpredictable and exposes individuals to situations over
which they have little or no control. A lack of control in high-
density settings has been shown to exacerbate the negative
effects of density on humans, whereas the availability of con-
trol has been shown to reduce the negative effects of density.
One group of researchers tested the control hypothesis by
examining the effects of control on people's moods in crowded
elevators. Control in the elevator was manipulated by giving
some people access to the elevator control panel and other
people no access. Those who had panel access, or more con-
trol, in the crowded elevators felt less crowded and had more
positive moods than those without control. As with behavioral
constraint, sometimes it is enough to simply perceive, or
believe, that control is available to reduce the negative effects
of high density on performance or mood. However, if one's
expectations for control in a high-density situation do not
match the actual availability of control, then the high density
can be more disturbing than if one expected little control
in the situation. Thus, it appears that control or beliefs
Crowding: Effects on Health and Behavior 643
and expectations about control in high-density environments
influence how strongly humans are affected by crowding.
Social Withdrawal
One consequence of crowding is excessive and unwanted social
interaction and interference. Social withdrawal is a practical
behavioral response for reducing unwanted social interaction.
For example, commuters on a crowded subway will often
bury their heads in a newspaper and listen to music through
headphones rather than engage with fellow commuters. Some-
times individuals will avoid the unwanted social interaction
associated with crowding altogether by avoiding the crowded
space. For example, college students who are forced to double-
up in a small dormitory room might spend more time at the
library or gym to reduce time spent in the room. Residents
in a crowded apartment might isolate themselves by closing
doors between rooms or spending lots of time alone in a
bathroom or other private area in the apartment. Whereas
social withdrawal is effective at reducing unwanted social inter-
actions, it also can have potentially unintended and undesir-
able consequences, including reductions in social support and
companionship.
As discussed above, several studies have shown that among
people living in crowded households social withdrawal is
associated with a lower level of social support, which, in turn,
is associated with a higher level of psychological distress.
Although there is no evidence to date, it is also likely that social
withdrawal between intimates could lead to feelings of neglect
and social strains, which might have their own adverse health
effects. Additional evidence is needed to determine whether
social withdrawal and its unintended negative social conse-
quences have any implications for physical health.
Overload/Arousal
A final theory posits that high density increases pathology
because of sensory overload from excessive stimulation.
Humans have a limited capacity to process information; in
high-density settings, the information available in the environ-
ment exceeds that capacity. This process is similar to informa-
tion overload in a computer system, which can cause a
computer to make errors or shut-down operations. In humans,
overarousal is often unpleasant, can diminish complex task
performance, and could contribute to health problems.
Evidence for the overload/arousal model of crowding is
found in studies such as those discussed above that have
shown increased sympathetic arousal under high-density situa-
tions. Some scholars have suggested that heightened social with-
drawal in high-density settings is a method of reducing arousal.
At present, however, there is no strong evidence that social
withdrawal can actually lower sympathetic arousal in crowded
people. In addition, there is no evidence that the levels or
duration of arousal people experience in high-density settings
is significant enough to compromise health or contribute to
disease processes.
Conclusions
Crowding in humans is a syndrome of stress associated
with exposure to high-density households, community, or
laboratory settings. A distinction exists between the subjective
experience of crowding and the objective source of that feeling:
high-population density. Overpopulation in nonhuman ani-
mals leads to deviant social behaviors and health problems.
However, exposure to high density and overpopulation in
humans seldom results in extreme social pathologies. Unlike
lower animals, humans appear to be able to adapt to and cope
with high-density situations with a good deal of tolerance.
Humans can adapt by limiting their exposure to high density
through many means, including architectural interventions,
careful scheduling and planning of space usage, and by
engaging in distracting or withdrawal behaviors. Unfortu-
nately, some adaptations to density, such as social withdrawal,
can have unintended consequences, such as loneliness or
deterioration of interpersonal relations.
The research evidence to date suggests that people do
have undesirable psychological, social, and biological
responses to crowding. However, it is also clear that crowd-
ing is more or less aversive and detrimental to people
depending on their personal experiences with and prefer-
ences for particular levels of density. In addition, social con-
ditions, such as the presence of supportive others or
undesirable social hassles from others, can influence the
strength of the relation between density and various out-
comes, such as psychological well-being. Thus, density does
not necessarily increase social pathology in humans. Indeed,
low levels of density can be undesirable at some occasions,
such as parties, sporting events, or concerts. It appears that
density is most detrimental to human behavior and health
when individuals feel a lack of control over their own beha-
viors or the environment, or when they experience excessive
stimulation and arousal from the density.
See also: Crowd Psychology; Group Dynamics; Stress and Illness.
Further Reading
Cohen JE (2003) Human population: The next half century. Science 302:
1172-1175.
Evans GW and Lepore SJ (1992) Conceptual and analytic issues in crowding research.
Journal of Environmental Psychology 12: 163-173.
Evans GW, Lepore SJ, and Allen KM (2000) Cross-cultural differences in tolerance for
crowding: Fact or fiction? Journal of Personality and Social Psychology
79: 204-210.
Freeman H and Stansfeld SA (eds.) (2008) The Impact of the Environment on
Psychiatric Disorder, pp. 127-157. London: Routledge.
Hartig T and Lawrence RJ (eds.) (2003) The residential context of health. Journal of
Social Issues 59(3) (special issue).
Regoeczi WC (2008) Crowding in context: An examination of the differential responses
of men and women to high-density living environments. Journal of Health and
Social Behavior 49: 254-268.
Cultural Psychology
J G Miller and V Chauhan, New School for Social Research, New York, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behaviorism A school of research in psychology that
explains behavior in terms of the association between a
stimulus in the environment and a behavioral response.
Cognitive Revolution An intellectual movement in the
1950s that was oriented toward establishing meaning as a
central concept in psychology.
Cultural tools Culturally and historically based artifacts,
symbols, technologies, and forms of language which impact
on modes of thought.
Executive functioning The cognitive processes that regulate
the ability to organize thoughts and activities, prioritize
tasks, and make decisions.
Information processing approaches A perspective
on cognitive development that focuses on the
operations by which people mentally manipulate
information.
Priming The implicit memory process in which a stimulus
activates or makes more retrievable one or more memories
or associations held by an individual.
Cultural psychology refers to a perspective on psychological
research which recognizes the essential role of culture in the
formulation of basic psychological theory. The perspective
of cultural psychology captures the core psychological insight
that psychological processes always occur in socioculturally,
historically situated contexts and depends on the individual's
participation in those contexts for their emergence; that is,
without culture advanced psychological processes would not
exist, just as they would not exist without biology. While it is
well understood that people are inherently biological beings,
the insight of cultural psychology is to recognize that they are
also inherently and simultaneously cultural ones. The discus-
sion here provides an overview of the key theoretical assump-
tions of cultural psychology followed by a select overview of
major traditions of research associated with cultural psychol-
ogy and of the types of empirical findings emerging. Consider-
ation is also given to challenges facing contemporary work in
cultural psychology as well as to the contributions and promise
of research in this tradition.
Core Premises of Cultural Psychology
The perspective of cultural psychology calls into question the
dominant theoretical position that has long dominated work
in psychology, which is to downplay the importance of culture
in the formulation of basic psychological theory. This theoreti-
cal position that has tended to characterize most psychological
research treats culture as content or context and assumes that
psychological development is not only fundamentally univer-
sal but can be understood without taking into account cultural
considerations. It is assumed that the basic nature of psycho-
logical processes and structures is unaffected by the culturally
dependent content on which they operate or by the cultur-
ally structured contexts in which they occur. From this type of
perspective, cultural variation in human behavior is considered
to be of limited interest in that it is viewed as leading away
from the search for underlying regularities to a focus on out-
ward appearances. This type of stance has been associated with
programs of research in psychology that have tended to present
psychological theory in culture-free terms and to tap culturally
skewed populations, focusing predominately on American
research participants and neglecting to study the vast majority
of the world's populations.
The theoretical stances adopted in cultural psychology
differentiate it not only from this mainstream approach in
psychology but also distinguish it from earlier traditions
of cross-cultural research that were prominent in the mid-
twentieth century and that tended to focus on testing the
assumed universality of existing psychological theories. These
theoretical stances include adoption of symbolic views of cul-
ture, acknowledgment of the essential role of cultural media-
tion in the emergence of higher order psychological processes,
and recognition of the mutual constitution of culture and self.
Symbolic Views of Culture
The emphasis on meaning that occurred with the rejection
of behaviorism in the mid-twentieth century constituted an
important foundation for cultural psychology. This emphasis,
which emerged with the transition in psychology known as the
Cognitive Revolution, underscored the role of the individual in
going beyond the objective information given in making sense
of experience. Within behaviorism, the individual had been
portrayed as passively responding to objective stimuli in the
environment. The Cognitive Revolution, however, made clear
that this point of view fails to recognize the individual's role
in actively contributing meanings to experience, with the
same information potentially giving rise to different interpreta-
tions and having contrasting implications for the individual's
affective, cognitive, and behavioral reactions. Notably, there
was no straight path from this insight of the Cognitive Revolu-
tion to contemporary work in cultural psychology, as the
emphasis on meaning-making of the Cognitive Revolution was
waylaid for many years as more passive information processing
approaches came to dominant the field. However, this recogni-
tion that an act of interpretation intervenes between the stimu-
lus and the response established an important theoretical basis,
which was later drawn upon as investigators came to appreciate
the role of culture as a necessary component of meaning.
644
Cultural Psychology 645
The emergence of symbolic views of culture in work in
anthropology and linguistics formed a key foundation for
cultural psychology in making clear not only the collective
aspects of meanings but that meaning could not be reduced
to objective properties of the situation. In this view, cultural
systems are understood to do more than merely provide infor-
mation about experience or convey norms for behavior.
Rather, they are seen as creating social realities whose existence
depends, in part, on these cultural definitions and the practices
that instantiate them. This meaning creating role of culture
extends not only to recognized social institutions (e.g., mar-
riage, school) and their associated roles (e.g., bride, student)
and artifacts (e.g., wedding ring, grades), all of which depend
on socially shared definitions and activities, but also to core
concepts in psychological theory, such as categories of mind,
emotion, and self, which also depend, in part, on socially
shared concepts and practices.
The theoretical importance of recognizing the role of cul-
ture as a system of meanings and practices was that it high-
lighted the need to take culture into account in psychological
explanation. Contexts for psychological development could no
longer be understood merely in terms of objective constraints
and affordances, but required consideration of culturally
shared beliefs and values through which the contexts were
interpreted and structured.
This role of culture as a system of meanings and practices
may be illustrated through research undertaken by Tobin, Wu,
and Davidson on Japanese and US teachers' views of preschool
classrooms. US teachers were observed to be highly critical of
practices in a Japanese classroom that included a ratio of one
teacher per 30-40 students, and to much prefer the ratio found
in a US classroom that included 2 teachers per 8-10 students,
making the observation that teachers can devote more atten-
tion to individual students in the latter setting. However,
whereas Japanese teachers recognized this objective affordance
provided by the US preschool classroom, they appraised the
classroom structure negatively, as not providing children with
opportunities to function as part of a larger social group.
Cultural Mediation of Psychological Processes
In addition to recognizing the nature of culture as a system of
meanings and practices, the perspective of cultural psychology
includes the recognition that human thought and behavior is
mediated by culture and that it is only through activity in
sociocultural environments that higher order psychological
processes emerge. This insight is termed the incompleteness
thesis to capture its core assumption that human psychological
development cannot be completed without cultural experience.
The incompleteness thesis is based on the recognition that
culture is ever-present in human experience. Individuals, it is
noted, always act in culturally and historically specific envir-
onments and utilize culturally and historically specific tools.
They also carry with them in their language and understanding
systems, culturally specific assumptions through which they
interpret experience. This omnipresent enculturative structure
is understood to introduce a cultural-historical grounding to
psychological functioning with the assumption made that it is
virtually impossible for individuals to act in ways that are not
culturally mediated. Even the highly controlled context of a
scientific laboratory, it is recognized, does not fully control for
culture and is not culture-free, but rather itself constitutes a
highly particular cultural environment. While individuals are
assumed to show instinctual responses and other inborn pro-
pensities that are not culturally mediated, their environmental
experiences (even in utero), reflect a cultural patterning and it
is only by the gradual internalization of the tools of a culture
that advanced psychological processes emerge.
Importantly, this type of perspective does not assume that
all psychological processes are culturally variable. However, it
does assume that advanced psychological processes will invari-
ably reflect a cultural patterning, and thus in many more cases
than presently recognized will show cultural variability. From
the perspective of cultural psychology the acknowledgement
that a particular psychological process is culturally dependent
or culturally variable is not, however, invariably regarded as a
flaw given that it is assumed that many basic psychological
theories are likely to reflect a cultural-historical specificity.
In fact, a core part of the agenda of contemporary work in
cultural psychology is to make this cultural patterning of psy-
chological theories explicit. Work in cultural psychology, in
this way, is as much about achieving a greater understanding
of the often implicit cultural underpinnings of present psy-
chological theories as it is about gaining insight into cultural
diversity in psychological functioning. Rather than being
relegated to the role of a footnote, qualification, or exception
in the formulation of basic psychological theory, as cultural
work is currently presented in many textbooks or handbooks
of psychology, cultural evidence is recognized to be an essen-
tial basis for the formulation of psychological theory. While
cultural psychology is at times portrayed as a subfield of psy-
chology, this constitutes a misnomer. Cultural psychology is
coterminous with all of psychology, in sharing its fundamental
agenda of contributing to the formulation of psychological
theories and in explaining the processes of human develop-
ment. This recognition of the inherently cultural nature of
all fields of psychology implies that there could never be an
adequate psychology that is culture-free, just as it cannot be
assumed that there could ever be an adequate psychology that
fails to take into account the biological foundations of self.
Mutual Constitution of Culture and Self
In addition to recognizing the culturally mediated nature of
psychological processes, the perspective of cultural psychology
highlights the mutual interdependence of culture and self.
Notably, the conjoint term 'cultural psychology' was selected
by theorists to convey this insight that, just as psychological
processes need to be understood as always based in particular
cultural-historical contexts that impact on their form and pat-
terning, cultural meanings and practices depend on the exis-
tence and activities of communities of intentional agents.
While culture cannot be reduced to individual psychology
nor individual psychology understood as purely culturally pat-
terned, culture and psychology are recognized to be mutually
interdependent.
This assumption of the mutual constitution of culture and
self underscores the need to approach culture in dynamic terms
just as psychological processes themselves are understood to
develop and change. Cultural meanings and practices are not
646 Cultural Psychology
static but rather transform over time, in ways that reflect both
adaptive functional constraints, such as survival pressures, as
well as historical continuities and discontinuities, as traditions
are transformed and altered, even as they are forwarded and
preserved.
The assumption of the mutual constitution of culture and
self also highlights the compatibility, if not the fundamental
necessity, of a consideration of both biological and cultural
influences on psychological development. It is recognized that
just as individual psychological development depends on
biological processes, biological phenomena are affected by
cultural processes, such as demonstrated in the impact of tool
use on the evolution of modern humans. This fundamental
compatibility of biological and cultural aspects in constituting
the self is seen in the increasing attention being paid to neuro-
logical and other biological considerations in research in
cultural psychology.
Research Traditions and Findings
Work that shares the theoretical assumptions and goals of
cultural psychology is found in all subfields of the discipline,
with this work, in cases, being undertaken by theorists who do
not explicitly adopt the self designation of cultural psychology,
or even by investigators identified with related disciplines,
including anthropology, linguistics, and sociology. This work
incorporates all types of methodology, and is not restricted to
research that is either comparative or ethnographic in nature.
While the present discussion cannot cover all of the various
traditions of work in cultural psychology, brief consideration
will be given here to some of the major areas of work in
cultural psychology and to the types of empirical findings
emerging.
Sociocultural Approaches
One of the most longstanding traditions of research in cultural
psychology is the tradition identified originally with the
cultural-historical approach of Vygotsky and his Russian
colleagues. This tradition has been extended in recent years
by investigators such as Cole, Rogoff, and Wertsch, among
many others, and is generally referred to by the contemporary
designation of 'sociocultural' perspectives. Within this tradi-
tion, extensive consideration is given to the role of cultural
tools and other meditational means in qualitatively affecting
the form of human cognition and behavior. From this perspec-
tive, human activity is understood in a broad-based way that
takes into account historical, individual developmental, phy-
logenetic, and microgenetic levels of analysis. There is also a
shift away from an exclusive focus on psychological processes
as internal to individuals to a focus on their instantiation in
everyday social activities.
A major direction of work in this tradition is to examine
how participation in everyday cultural practices leads to
changes in thought that are qualitative in form while also
contextually dependent. For example, it has been shown that
individuals who have extensive experience in using the abacus
are more successful in solving math problems and employ
more complex and qualitatively different cognitive processing
strategies in their solutions than do individuals who have no
experience using this cultural tool, with these contrasting cog-
nitive strategies showing only limited generality to other types
of cognitive tasks. Likewise, research conducted in Brazil on
children with little education who work selling fruits on the
streets has documented their impressive and highly flexible
strategies for solving the everyday math problems that they
employ in their trade - strategies, however, that do not transfer
to their performance on school-based arithmetic tasks. Studies
by Michael Cole and his colleagues who have examined the
impact of literacy and schooling on cognitive development
have also uncovered the powerful, yet localized, impact of
exposure to cultural tools on cognition. Thus, it has been
demonstrated, for example, that while marked effects of school-
ing are observed on conventional cognitive measures, these
measures are similar in form to the types of tasks that tend
to be practiced in school contexts. It is further shown that
individuals who have had extensive schooling or who have
scored high on cognitive tests are no better than their
unschooled or low scoring counterparts in solving everyday
cognitive tasks, such as making the most successful bets at the
racetrack or the most economical purchases in the supermar-
ket. Providing insight into the processes through which experi-
ence with cultural tools and in culturally structured contexts
impacts cognition, this work challenges early global stage mod-
els that had linked modernization to global differences in
thought - a paradigm known as the hypothesis of the primitive
versus modern mind.
Work in this tradition also provides insight into the links
between social and developmental change. A dramatic exam-
ple of this type of work may be seen in the variation in every-
day practices and in associated styles of cognitive activity
observed to occur in Chiapas, Mexico, which was studied by
Patricia Greefield and her colleagues over a period of more
than two decades from 1969 to 1991. During the early period
in which the local economy was heavily agricultural, emphasis
was placed on subsistence activities, such as weaving of all the
family's clothing, and girls were taught to weave in a process
closely guided by their caregivers who were present in the
context. However, during the later period in which there had
been a shift to an economy based on commerce and money,
not only was there a reduction in the economic importance of
weaving (since clothes were now more frequently purchased),
but the activity of girls in learning to weave had changed to a
more trial and error process that was no longer characterized
by the hands-on attention and direct involvement of the
caregivers. During this historical period, there was also a mod-
ification in the weaving products, from a more detailed to a
more abstract style of visual representation, and adolescent
girls were found to show enhanced abilities in representing
novel visual patterns. This type of research notably underscores
the dynamic nature both of sociocultural environments and of
their impact on developmental outcomes.
Research in the sociocultural tradition has also led to
a rethinking of conceptions of development. For example,
in research on the teaching of reading, Michael Cole provided
a new solution to what is known as the paradox of qualitative
developmental change, that is, the problem of accounting
for how individuals can acquire more powerful cognitive
structures if they do not already possess these structures.
Cultural Psychology 647
The research demonstrated that a range of meditational means
are available in cultures that support reading, including not
only literacy activities like storytelling, but artifacts like com-
puter games. This made clear that many of the structures
entailed in the achievement of reading exist between persons
in the sociocultural context before they appear as individual
competencies. In another example, work by Barbara Rogoff has
highlighted the role of cultural practices in underlying devel-
opmental change. She provides evidence to indicate that the
marked cognitive shifts that have been documented by devel-
opmental psychologists occur over the age range of 5-7 years
old and that have classically been explained solely in cognitive
terms to result, in part, from changes in the sociocultural
settings and types of cultural supports provided to children
during this age period. It is during this period that children
commonly move more fully into the school context and gain
greater exposure to the peer group, with attendant variation
in their activities, which has implications for their emerging
cognitive competencies.
Developmental Approaches
The tradition of cultural psychology discussed here includes
approaches that draw their inspiration primarily from work
in the contemporary mainstream tradition of developmental
psychology. Theorists in this tradition focus on the impact
of culture on the path of developmental change, and on cul-
turally broadening the theoretical constructs of developmental
psychology.
Work in this tradition of cultural psychology has documen-
ted that the direction and endpoints of developmental change
are, in cases, culturally variable, and thus, that explanations
of developmental change need to take into account cultural
considerations and universal cognitive or experiential factors.
Psychological research that had been conducted primarily on
US cultural samples had uncovered an assumed universal
developmental change for young children to shift from a
focus on viewing the self and others in concrete terms, such
as by reference to their physical characteristics, to a focus by
adolescence on viewing the self in terms of traits - a develop-
mental trend that was explained as resulting from children's
increasing abilities to engage in abstract thought. However, in
comparative developmental research conducted among US
and Hindu Indian populations by Joan Miller, it was found
that not only did Hindu Indian populations not show this
developmental increase in emphasis on traits, but they showed
a significant developmental increase not found among US
populations to give increasing emphasis to contextual consid-
erations. Thus, whereas an American adult might explain
behavior in terms of the agent's personality (e.g., that an indi-
vidual was helpful), Indian adults would show a greater
tendency to explain it in terms of features of the situation
(e.g., that no one else was around, or that the individuals
involved were friends).
This same type of interest in uncovering cross-cultural
developmental variation may be seen in a recent program of
cross-cultural developmental research by Qi Wang and her
colleagues that points to the role of cultural beliefs and prac-
tices, such as everyday discourse routines, in the development
of autobiographical memory. Wang and her colleagues found
that not only is the age of earliest autobiographical memories
later among Chinese than among US child and adult popula-
tions, but the form of autobiographical memories is culturally
variable, with US respondents tending to recount earliest mem-
ories which are emotionally intense and positive and that focus
on their individual psychological reactions, and Chinese
respondents tending to recount earliest memories that are
less emotionally laden and less self-focused and that tend to
center on everyday social routines, such as a caregiver's behav-
ior in cooking the daily meal.
A related thrust of work in this tradition is to formulate
developmental theories that are more culturally inclusive.
An example of this type of stance may be seen in work by
Fred Rothbaum and his colleagues that points to the need to
broaden conceptions of attachment theory to encompass con-
cerns with amae that are salient in Japanese cultural contexts.
Amae is a culturally grounded Japanese concept that involves
positive feelings of depending on another's benevolence. This
argument for the need to culturally broaden attachment the-
ory, one of the most central theories of social development, is
based not only on the research by Rothbaum but on a growing
body of research by a range of investigators that provides
evidence of cultural variation in the incidence of secure attach-
ment as well as of uncovering cultural variation in styles of
everyday social interaction, such as the greater emphasis
found in Japanese families on nonverbal styles of parent-
child interchange that promote the development of empathy,
a dimension that is not emphasized in attachment theory.
Another example of this focus on culturally broadening
theoretical concepts in developmental psychology may be
seen in the growing body of work on culture and parenting.
For example, one of the most influential contemporary the-
ories of parenting, which was developed by the theorist Diana
Baumrind, conceptualizes parental control in terms of a cate-
gory of assumed harsh authoritarian parenting. However, as
theorists such as Ruth Chiao have observed in their research,
there exist highly directive forms of parenting among Chinese
families that combine an emphasis on standards of conduct
and preserving the integrity of the family with deep affective
concern and caring for the child. To capture this type of stance
on parenting, a stance that notably is associated with posi-
tive adaptive outcomes, such as higher levels of educational
achievement, it is crucial to conceptually broaden present psy-
chological models of parenting. Cross-cultural developmental
research has also demonstrated that, in contrast to the ten-
dency observed among European American children and ado-
lescents, Korean cultural populations show a greater tendency
to associate greater parental warmth with greater perceived
parental involvement in the child's decision making. Theorists
point out the need to recognize that in portraying control
exclusively in negative terms, present theories of parenting
fail to capture the extent to which parental involvement may
be expressed and experienced in positive terms as conveying
warmth and a welcome involvement.
Social Psychological Approaches
The tradition of work in cultural psychology under consider-
ation here shares a similar agenda to that just discussed in the
area of developmental psychology. However, in this case,
648 Cultural Psychology
the focus is on culturally broadening theory and research being
undertaken in the area of social psychology. Theorists in
this tradition center their research on highlighting the role of
culture in the formulation of basic social psychological theory,
with particular attention given to issues of cognition, motiva-
tion, and self.
In terms of cognition, work in this tradition has provided
support for claims that there is a greater emphasis on holistic
modes of cognition in Eastern cultural settings, with a greater
emphasis on analytic modes of cognition in Western cultural
contexts. The former perspective privileges a deductive logic,
which emphasizes breaking up objects into their component
elements, whereas the latter stresses a more dialectical logic,
which stresses viewing objects in relational terms. In support of
this claim, it has been observed, for example, that Chinese
adults perform less successfully than do American adults on
category learning that requires the application of formal rules,
while performing more successfully than do American adults
in detection of covariation, an activity that demands attending
closely to the environment. In another example, related cul-
tural differences have been documented in the processes
of attention, with the recall of East Asian populations for
objects less than that of Americans if the background has
been switched, suggesting that European-Americans more
fully process contextual information. This difference in attention
notably has also been observed in saccadic eye-movements, as
assessed with eye-trackers, with East Asians relative to Americans
gazing longer at background material.
In the area of motivation, cultural research is highlighting
the need to broaden psychological theory to take into account
the more social nature of internalization processes in collectiv-
ist cultural settings. For example, research has demonstrated
that Brazilian adults more fully internalize social norms than
do American adults, linking fulfilling a role related request for
help by a family member more strongly not only with social
expectations but also with satisfaction. Similar cross-cultural
differences have been documented in comparisons between
Indian and US populations with Indians associating fulfilling
role based social expectations to family and friends with choice
and US populations experiencing it in less freely chosen terms.
While this and other related research does not challenge the
existence of a concern with choice universally, it does highlight
the need to recognize that the motivational salience of choice
varies cross-culturally in ways that are not presently fully
accounted for in psychological theory.
In terms of self, recent cultural research is also identifying
cultural influences on the determinants of positive self-regard
and of psychological well-being. Research, for example, docu-
ments the existence of marked cultural variability in the ten-
dency to self enhance. For example, it has been demonstrated
that whereas Americans tend to rate their self esteem as higher
than that of others, Japanese tend to rate their self esteem as
similar in level to their peers. As another example, whereas
the self-descriptions of American adults tend to include four
to five times as many positive attributes as negative ones, those
of Japanese tend to give greater emphasis to their shortcom-
ings. It has also been documented that self esteem as well as the
experience of positive emotions correlate more strongly with
life satisfaction in individualistic as compared with collectivist
cultural populations. While work on this topic does not call
into question the importance of maintaining a positive sense
of self in all cultural populations, it does suggest that constructs
such as self-esteem, which assume a tendency toward self
enhancement, reflect a culturally specific stance and need to
be broadened to account for the cultural beliefs and practices
of different cultural populations.
Cultural Neuroscience
The perspective of cultural neuroscience is concerned with
understanding the cultural underpinnings of brain-behavior
relationships. Representing a new direction for research in
cultural psychology, this work forms part of the exponential
growth over the last two decades in interest in neuroscience in
all subfields of psychology, with the advent of functional neu-
roimaging. Cultural neuroscience provides insight into both
common and culturally variable brain-behavior pathways and
into the nature and extent of brain plasticity.
A central focus of research in cultural neuroscience is to
provide information about the role of cultural contexts in
mediating brain-behavior relationships. This work builds on
the growing body of evidence of cultural variability in basic
psychological processes and documents ways that this varia-
bility impacts the brain mechanisms recruited. Recent work in
cultural neuroscience has revealed, for example, that the neural
correlates of processes of perception, attention, and social
understanding show cultural variability that is related to the
cultural variability observed in these areas in psychological
performance. For example, congruent with recent findings
that US adult populations show greater object-focused proces-
sing than do East Asian adult populations, work in cultural
neuroscience demonstrates that stronger and more distributed
neural activation occurs during object processing among
US cultural groups than among East Asian cultural groups.
Research in this area has also revealed that brain mechanisms
are recruited in ways that reflect perceptual tasks which are
more culturally preferred being easier and thus requiring less
attention than do nonculturally preferred tasks. Thus, it is
observed that whereas East Asian adults recruit prefrontal and
parietal regions of the brain associated with attention more
for culturally nonpreferred context-independent judgments
than for culturally preferred context-dependent judgments,
the opposite pattern of neural activation is observed among
US adults. In another example, in accord with findings that
collectivist cultural populations tend to maintain outlooks that
incorporate others more fully into the self concept, research has
demonstrated that whereas the medial prefrontal cortext of the
brain is activated by judgments made about the self among US
cultural groups, it is activated by judgments made about both
the self and about one's mother among Chinese cultural groups.
Such work importantly provides evidence that these two cul-
tural groups use the same part of the brain in different ways.
Work in cultural neuroscience is also addressing key con-
ceptual questions such as the extent to which maturation of the
brain is affected by experience in contrasting cultural environ-
ments. For example, work on human aging is demonstrating
that elderly Asian populations show greater deficits in object
processing areas of the brain than do elderly Americans,
a pattern of findings that is congruent with the observed
Cultural Psychology 649
tendencies for Asian cultural groups to less fully process object
information than do US cultural groups. Such work provides
insight into processes of brain plasticity and lends support to
what is known as a use it or lose it view of human aging.
Cultural developmental neuroscience is also demonstrating
that neural functioning is not invariably the same in all sam-
ples of healthy children but reflects the influence of culturally
variable experiences. To give an example, cross-cultural devel-
opmental research has demonstrated that Asian children
tend to perform better than do North American children on
measures of executive functioning. In research conducted with
Chinese-Canadian and European-Canadian children who were
matched in terms of their performance on a reaction time test
of executive functioning, it was shown that the two groups of
children showed differential patterns of brain lateralization
compatible with the cross-cultural differences observed in
executive performance.
Challenges
As work in cultural psychology is gaining in prominence and
scope, it is also increasingly embodying a self critical perspec-
tive in which investigators work to recognize and address
certain limitations of past approaches and to identify impor-
tant directions for further work. These challenges faced by
contemporary researchers in cultural psychology center on
developing more adequate understandings of culture and the
self, and on more effectively integrating work in cultural psy-
chology with work in the larger discipline of psychology.
Enhanced Understandings of Culture
A major challenge in contemporary work in cultural psychol-
ogy is to develop more sophisticated and culturally sensitive
approaches to culture. With the exception of research from a
sociocultural perspective, much of the work in cultural psy-
chology to date has been informed by the interdependent/
independent or East-West dichotomy. This distinction gained
in prominence with a seminal review paper on culture and the
self coauthored by Hazel Markus and Shinbou Kitayama in
1991, with the distinction further promoted in the field by
the subsequent development of psychological scales to mea-
sure this dichotomy. More recently, this perspective on culture
has grown in prominence with the emergence of approaches
that utilize priming techniques to assess culture. For example,
one of the tasks that has been developed to prime an indepen-
dent as compared with interdependent cultural view of self
presents individuals with a paragraph that describes the beha-
viors of an individual who is referred to either only with the
pronoun I, in the independent case, as compared with only by
the pronoun We, in the interdependent case. As critics have
noted, however, this type of approach to culture emphasized
to date in much work in cultural psychology tends to gloss over
important distinctions between and within cultures, to give
limited attention to the impact of context on behavior, and
to embody conceptions that treat individual perspectives as
isomorphic with trends at a cultural level.
One way that theorists are working to develop more
nuanced approaches to culture is through research that is
more fully tapping within and between culture variation in
outlooks. Thus, not only are more studies being undertaken
that assess psychological variation between different individu-
alistic and collectivistic cultures, such as comparisons of North
American with Western European samples, or comparisons
that contrast collectivist populations outside of East Asia, but
increasing attention is being given to sampling cultural groups
who vary in socioeconomic status and/or ethnic background.
For example, in research on attachment that was undertaken
by Robin Harwood, it was demonstrated that ethnicity and
socioeconomic status make distinctive contributions, with the
impact of socioeconomic status showing both common and
culturally variable patterns in different ethnic groups.
Theorists are also rethinking the assumption that cultural
outlooks can be tapped by individual difference measures. It is
increasingly recognized that individual differences do not vary
in a one to one relationship with cultural differences, a stance
that it can be argued reduces one level of analysis to that of
another, as well as recognizes that culture impacts behavior in
ways that are often implicit and is not the kind of tendency that
can be tapped by explicit psychological scale measures. More
attention is also being given to assessing cultural processes,
such as by attention to discourse routines and other everyday
social practices, as well as by tapping cultural products and
their associated meanings. While there remains an appeal to
the use of priming techniques in social psychological work in
cultural psychology, given the amenability of this type of
method to experimentation, increasing concern is also being
given to the limitations of priming, with calls for greater atten-
tion to be paid to the culturally variable knowledge that is
elicited by particular priming stimuli.
Greater Impact on Psychological Theory and Research
Work in cultural psychology remains sensitive to the chal-
lenges entailed in its goal to culturally broaden theory and
research in mainstream psychology. While work in cultural
psychology has increased in quantity in recent years, it remains,
as it has been historically, in a marginal position in the disci-
pline. In most departments of psychology, little or no attention
is given to representing culture in the curriculum or in the hiring
of faculty, with culture in many departments relegated only to
the important, but still marginal, role of an approach to be
applied solely in understanding diversity and not in formulating
basic psychological theory. This downplaying of the importance
of culture is illustrated, for example, in the contrasting response
of departments of psychology to the advent of neuroscience as
compared to cultural psychology. While both of these perspec-
tives have gained in prominence in recent years, neuroscience
has quickly become fully integrated into the discipline, as
seen in departments of psychology not only hiring neuroscien-
tists but frequently restructuring their departments to incorpo-
rate sub-divisions devoted to neuroscience, while cultural
psychology has not seen a comparable growth in acceptance
or influence.
While this marginalization of cultural approaches has roots
in the idealized natural science ideal of the discipline and is not
likely to change in a dramatic way, the increasing quantity of
cultural research being undertaken as well as its increasingly
international grounding represents one way for this work to
650 Cultural Psychology
have a greater impact on the discipline. In recent years, the
population of individuals trained as psychologists is becoming
more international, with investigators of diverse cultural back-
ground more frequently engaging in productive collabora-
tions, and with the growth of psychology journals that are
based in non-Western cultural settings. Also, while there is
increasing awareness of certain pitfalls associated with the recent
wholesale embrace of neuroscience in psychology, such as its
adoption by theorists to support arguments that are determin-
istic or circular in nature, it is also increasingly recognized that
taking culture into account is essential in neuroscience work
and, thus, that culture must be a key part of any psychology of
the future that is grounded in an understanding of brain pro-
cesses. To the extent that a particular psychological process has
been shown to be culturally variable, an account of the neuro-
logical correlates of that process will be incomplete, if not mis-
leading, without taking into account this cultural variability.
Also, as a growing number of theorists are pointing out that
meanings are not given in brain processes but rather require the
consideration of culturally shared social concepts and practices.
Promise of Cultural Psychology
Even while facing challenges, this represents an exciting time
for research in cultural psychology. Work in cultural psychol-
ogy continues to pursue its central agenda of making the
implicit cultural processes that impact on psychological func-
tioning more transparent, thereby contributing to the creation
of basic psychological theory. This work also continues to be
integral to developing psychology in ways that increase its
social relevance and applicability to solving real world social
problems. In this respect, work in cultural psychology is
essential in the efforts of the field to give back the insights of
psychology to the community. Finally, work in cultural psy-
chology continues to be integral to attempts to develop a more
sophisticated understanding of what can be designated as basic
psychological mechanisms. While enhancing current under-
standing of cultural variability in psychological processes and
outcomes, cultural psychology is vital in furthering the goal of
creating a truly universal psychology.
See also: Aging and Cognition; Aging and the Brain; Attribution;
Cross-Cultural Adaptation; Individualism; Social Cognition.
Further Reading
Bruner J (1990 ) Acts of Meaning. Cambridge, MA: Harvard University Press.
Cole M (1996) Cultural Psychology: A Once and Future Discipline. Cambridge, MA:
Harvard University Press.
Greenfield PM (2009) Linking social change and developmental change: Shifting
pathways of human development. Developmental Psychology 45(2): 401-418.
Heine S (2010) Cultural psychology. In: Fiske S, Gilbert DT, and Lindzey G (eds.)
Handbook of Social Psychology, vol. 2, pp. 1423-1464. New York: Oxford
University Press.
Kitayama S and Cohen D (eds.) (2006) Handbook of Cultural Psychology, pp. 477-499.
New York: Guilford Press.
Markus HR and Kitayama S (1991) Culture and the self: Implications for cognition,
emotion, and motivation. Psychological Review 98(2): 224-253.
Matsumoto D (ed.) (2001) The Handbook of Culture and Psychology. New York: Oxford
University Press.
Miller JG (1 997) Theoretical issues in cultural psychology. In: Berry JW, Poortinga YH,
and Pandey J (eds.) Theory and Method. Handbook of Cross-Cultural Psychology,
2nd edn., vol. 1, pp. 85-128. Boston, MA: Allyn & Bacon.
Miller JG (2003) Culturally sensitive research questions and methods in social
psychology. In: Sansone C, Morf CC, and Panter AT (eds.) Sage Handbook of
Methods in Social Psychology, pp. 93-116. Thousand Oaks, CA: Sage
Publications.
Miller J and Kinsbourne M (2011) Culture and neuroscience in developmental
psychology: Contributions and challenges. Child Development Perspectives.
Murphy-Berman V and Berman J (eds.) (2003) Nebraska Symposium on Motivation:
Cross-Cultural Differences in Perspectives on the Self, vol. 49, pp. 59-99. Lincoln,
NE: University of Nebraska Press.
Rogoff B (2003) The Cultural Nature of Human Development. New York: Oxford
University Press.
Shweder RA and LeVine RA (1 984) Culture Theory: Essays on Mind, Self, and Emotion.
Cambridge: Cambridge University Press.
Sternberg R and Grigorenko E (eds.) (2004) Culture and competence: Contexts of Life
Success, pp. 11-138. Washington, DC: American Psychological Association.
Tobin JJ, Wu DYH, and Davidson DH (1989) Preschool in Three Cultures: Japan,
China, and the United States. New Haven, CT: Yale University Press.
Decision Making (Individuals)
C F Chick, S T Pardo, V F Reyna, and D A Goldman, Cornell University, Ithaca, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Affect A feeling state that is distinguished by valence
(positive vs. negative) and sometimes arousal and other
dimensions.
Cognitive reflection test A set of three questions, to which
people give a response that is assumed to be either automatic
and seemingly obvious (but incorrect) or effortful and
deliberative (and correct).
Expected utility Similar to expected value, but outcomes are
nonlinearly transformed into utilities (indices of relative
value).
Expected value The average outcome of a risky choice. For
example, a coin-toss for $200 or $0 has an expected value of
$100 (i.e., 0.5*200 + 0.5*0=100).
Framing effects Shifts in preferences for objectively
identical options when outcomes are described as gains
versus losses.
Invariance An axiom stating that individuals should show
equal preference for options that have equal consequences.
Numeracy The ability to understand and use numbers.
Preferences In psychology, an underlying construct thought
to reflect the value of options; in economics, revealed by
choices.
Risk Can be defined in many ways. One version is the
possibility of, or variance among, multiple outcomes.
System 1 and 2 Two modes of thought in standard dual-
process theories. System 1 is characterized as fast, automatic,
effortless, and intuitive, whereas System 2 is characterized as
slow, deliberate, controlled, and effortful. Standard dual-
process models posit that System 1 is responsible for many
cognitive errors and that System 2 can sometimes override
System 1 in order to avoid such errors.
Temporal discounting Devaluing of future outcomes
relative to immediate ones.
Utility In economics, a measure of relative satisfaction or
pleasure associated with an outcome.
Introduction
Decision making can be defined as the selection of a course of
action among options. Those options can range from selecting a
marriage partner to selecting a political party affiliation. Scholars
who study decision making come from many disciplines. For
example, decision making has long been of interest in econom-
ics and psychology. Early models of the decision making process
were mathematical. These mathematical models described the
decision making process as a matter of performing calculations,
involving such concepts as probabilities of outcomes and the
monetary values of outcomes, to select the best course of action.
Many modern models of decision making are variations of these
original mathematical formulations.
However, recent conceptions of decision making empha-
size psychological factors. Among these factors are relatively
stable characteristics of people, which are called individual
differences. Numeracy, or the ability to understand and use
numbers, is a popular example of an individual difference that
is thought to influence the quality of financial and health
decision making. Intelligence, or cognitive ability, is another.
In addition, emotion has received a great deal of attention
recently, especially in dual-process theories that contrast emo-
tional with deliberative decision making. Process models of
memory are also increasingly influential in research on deci-
sion making. For example, how people remember features of
options affects their preferences for the options. For these and
other topics, neuroscience has become a useful tool for reveal-
ing the mental processes that underlie preferences and choices.
There are many societies and organizations dedicated to
studying and improving decision making, such as the Society
for Judgment and Decision Making (www.sjdm.org) and the
Decision Analysis Society (http://www.informs.org/Commu-
nity/DAS). Decision making is also sometimes studied within
particular professional domains, such as business or medical
decision making. For example, the Society for Medical Deci-
sion Making (www.smdm.org) is an interdisciplinary group of
academicians (health economists, psychologists, and others)
and health professionals. These societies hold annual meetings
and publish journals.
History: Expected Value, Expected Utility,
and Prospect Theory
Psychophysical Approaches: Translating Reality into
Perceptions
From mathematics to economics
An early decision rule compared the expected values of
options. The expected value is calculated as the outcome
value multiplied by the probability of that outcome (e.g., a
0.5 probability ofwinning $10 is equivalent to winning $5 for
sure, with 1.0 probability, because 0.5*$10 = $5). In the
1700s, Daniel Bernoulli suggested that the utility or personal
value of wealth is not simply its monetary value. He suggested
thatthevalue of an additional quantity of money (say, $100) is
less for people who start out with more money. An extra $ 100
means more to a starving student making $9000 a year than it
means to a successful executive making $900 000 a year. This
idea of declining marginal utility or diminishing returns was
captured with a logarithmic function. In other words, utility of
additional money does not increase linearly with increasing
651
652 Decision Making (Individuals)
wealth, but falls short increasingly as the magnitude of wealth
increases. Thus, expected utility, not expected value, was
thought to govern decision making.
Expected utility was mainly used by economists to study
the behaviors of buyers and sellers. As with expected value, if a
decision maker makes choices consistently based on maximiz-
ing expected utility, his or her total wealth will be higher
(compared to choosing inconsistently or not choosing so as
to maximize utility). The idea that choices should be made
consistently has been formalized in terms of certain axioms or
rules of coherence. Debate continues to this day about which
axioms are the most fundamental. For example, the axiom of
transitivity holds that if someone preferred A to B and preferred
B to C, then the person must prefer A to C. Another axiom
called invariance means that if two options have the same
consequences, then individuals should show equal preferences
for those options.
For most people, the expected utility of a sure bet exceeds
the expected utility of the gamble that offers the same expected
payout; in other words, people are typically risk averse. Given
a sure option of winning $1000 versus a gamble option that
offers a 0.50 probability of winning $2000, most people will
prefer the sure option. (In the rare event that a person is risk
seeking, he or she should prefer the fair gamble of equal
expected value to the sure bet.) The more common preference
for $1000 can be explained by Bernoulli's decreasing marginal
utility of money. Alternatively, various other models attempt to
explain it. In expected utility theory, a given utility is reference
independent; that is, utility does not depend on the description
of the options. Expected utility theorists assumed that people
were rational, meaning that they maximized expected utility,
obeying axioms of internal coherence, and therefore had con-
sistent preferences for risk. Subsequent research, however, sug-
gests that decisions are based on the decision maker's initial
frame of reference.
Prospect theory
In the late 1970s, Daniel Kahneman and Amos Tversky
challenged the rational view of decision making. They intro-
duced prospect theory, which was based on psychological
research indicating that individuals are not consistently risk
seeking or risk averse. Rather, individuals interpret gains and
losses relative to a reference point, which is usually the status
quo (e.g., the amount of money that they have initially).
People tend to be risk averse for gains relative to the status
quo (money won) but risk seeking for losses relative to the
status quo (money lost). For example, imagine that you were
given $2000, but you had to then choose between two options:
losing $1000 for sure or gambling on a 0.50 probability of
losing $2000 and a 0.50 probability of losing nothing. In
response to this problem, which describes the options in
terms of losses relative to an original sum, most people prefer
to gamble (they are risk seeking). Of course, once you subtract
your losses from the $2000 you were initially given, you are
facing exactly the same options as in the previous paragraph,
where the options were described as gains to an original sum.
However, whereas most people prefer the gamble in the loss
problem, they tend to prefer the sure option in the gain prob-
lem. In light of this result, Kahneman and Tversky proposed an
alternative theory of risk in which utility is based on changes in
wealth rather than absolute states of wealth. This new theory
was called prospect theory.
The prospect theory model looked like an S curve; it was
concave in the first quadrant (representing gain) and convex in
the third quadrant (representing loss). Crucially, the slope of
the loss function was steeper than the slope of the gain func-
tion, indicating that people place a higher value per unit on
loss than they do on gain. That is, people are more upset about
losing something than they are happy about gaining the iden-
tical amount. This asymmetry between losses and gains is
called loss aversion.
Prospect theory fits some hard-to-explain data such as
the endowment effect, in which people require more money
to give up something they already have than they are willing to
pay to get the identical object. This effect is also explained in
terms of loss aversion. For example, in one study, subjects were
randomly assigned to be buyers or sellers. Sellers were given a
coffee mug valued at $6 and were asked whether they were
willing to sell their mugs at a range of prices between $0.25 and
$9.25. Buyers were not given mugs and were asked whether
they were willing to buy mugs from the sellers at the same
range of prices. The results of the study demonstrated that
sellers consistently valued the mugs higher ($7.12 on average)
than the value the buyers ($2.87) were willing to pay for the
mugs. In some cases, the sellers preferred to keep their coffee
cups, rather than to sell them. People are often unwilling to
give up what they have to get something else that they might
otherwise prefer because a loss has a greater psychological
impact than a gain.
Framing
Frames are ways of describing choices. In the example above
in which $2000 is given initially and then dollar amounts
are subtracted, the options are described in terms of losses.
In an earlier example, the same options were described in
terms of gains. Despite being objectively identical, the two
versions of the choices are not psychologically equivalent
because they result in different preferences for risk (i.e., prefer-
ring risk in the gain frame but avoiding risk in the loss frame).
Different decision making theories offer alternative reasons as
to why frames result in inconsistent preferences, even in the
same individual. According to prospect theory, the distortion
in the psychological value of gains contributes to perceiving
the gamble outcomes (which are larger) as discounted relative
to the sure outcomes (which are smaller). The same types of
distortions for losses make the losses seem smaller in the
gamble than in the sure option. More recent findings have
cast doubt on the generality of the prospect theory explanation,
although the explanation holds in some circumstances.
There are individual differences in susceptibility to framing
effects. For example, framing has been shown to interact with
numeracy (the ability to understand and use numbers). The
more people understand numbers and probabilities, the less
likely they are to shift preferences when numerically identical
options are presented. This interaction between framing and
numeracy has been obtained with what is called attribute
framing, in which the same attribute is described in different
but objectively equivalent ways (e.g., test performance is
described as 80% correct vs. as 20% wrong) as opposed
to risky choice framing as in the examples discussed earlier.
Decision Making (Individuals) 653
It suggests that people high in numeracy might notice the numer-
ical equivalence of the options by spontaneously converting
positively framed attributes into negatively framed attributes,
and vice versa, because they are good at such computations.
People who score higher on the cognitive reflection test also
show less of a framing effect. The cognitive reflection test is
a three-item math test in which intuitive answers are wrong
and need to be overridden by correct analytical processes.
Contemporary Approaches
Global Models
Global models, in contrast to dual-process models, assume
that one or more sets of processes (i.e., a single system) govern
decision making. Many resemble the traditional psychophysi-
cal approaches already discussed (e.g., expected utility theory
and prospect theory). Some recent global models, for example,
add multiple reference points in addition to the status quo.
New global models include memory theories, such as Query
Theory, which predicts a wide range of effects based on the
order in which memory is queried or interrogated. For exam-
ple, asymmetries in buying and selling prices can be predicted
by the order in which buyers and sellers think of reasons to
have the good versus reasons to have the money. Whichever set
of features is retrieved first tends to inhibit the set of features
that is retrieved second, producing preference reversals when
the order of retrieval is varied.
Standard Dual-Process Theories
Sigmund Freud proposed an early dual process theory on which
some contemporary theories are based (e.g., Seymour Epstein's
cognitive experiential self theory). According to Freud, primary
and secondary processes designated two opposed yet comple-
mentary models of psychic functioning. The primary process is
driven directly by one's basic needs and serves the pleasure
principle without inhibition. The secondary process, in con-
trast, is the intervening rational system of control and regulation
that functions in service of the reality principle. Contemporary
dual-process theories similarly posit two opposing processes,
one more visceral and the other more reflective. These processes
are often called System 1 and System 2, respectively.
System 1 and System 2
There are some judgments that are formulated upon reflection,
based on systematic and calculative operations of decision
making. Alternatively, there are decisions that are made quickly
and automatically. Dual-process theories suggest that at differ-
ent times, responses can reflect the operations of either the
fast, automatic, and intuitive System 1 or the slower, serial,
and rule-governed System 2, as they were labeled by Keith
Stanovich and Richard West. Because System 1 is faster than
System 2, intuitive impressions can outrace rational delibera-
tion, producing wrong or biased responses that would be
censored on reflection.
In contrast to System 1, System 2 is conscious and effortful.
For example, when cognitive load is high (e.g., people have a
lot of information they have to keep in mind), people may
revert to System 1 thinking because it is less effortful. System 1
is a source of errors, biases, and fallacies in decision making,
although it does not necessarily result in poor performance.
Some standard dual-process theories make slightly different
dual-process distinctions. For example, some dual-process the-
ories contrast associative (mindless associations) versus rule-
based (mindful rules, such as logic) thinking, others contrast
concrete experiential thinking with rational analytical thinking
(as in the cognitive experiential self theory of Seymour
Epstein), and still others contrast affect (or emotion) and
cognition (or rationality).
Affect versus cognition
In standard dual-process approaches, as in Freudian dualism,
when people use affect or emotion to make decisions, their
thinking is illogical, irrational, or quantitatively inaccurate. For
example, relying on affect, people will donate more money to
save one starving child than to save eight starving children. It is
suggested that portraying a single child stirs stronger emotions
than portraying a group; the former seems more personal,
whereas the latter seems like a statistic. Thus, the single child
elicits more donations. Paul Slovic has suggested that these
processes are implicated in the insensitivity that people seem
to have to genocide.
This approach has also been called the dual-process valua-
tion model. This model posits that people assess the value of a
target in different ways, namely by affect (i.e., valuation by
feeling) or by analytic cognition (i.e., valuation by calcula-
tion). In one study, participants were asked how much they
would be willing to pay for a bundle of either five or ten
Madonna CDs. Subjects in the valuation-by-calculation condi-
tions performed mathematical calculations prior to pricing the
CDs. They were willing to pay more (on average) for ten CDs
than for five CDs. Subjects in the valuation-by-feeling condi-
tion engaged in activities designed to generate emotions. The
latter group reported no difference in the amount they were
willing to pay for ten CDs versus five. In all of these
approaches, emotion is expected to flatten the discrimination
of quantity (e.g., perceptions of the numbers of CDs) so that it
is more distorted away from linearity.
Fuzzy-trace theory
Approaches to rationality differ and include Piagetian logicism
(thinking as logic), information-processing formalism (think-
ing as computation), and intuitionism (thinking as intuition,
as exemplified in fuzzy-trace theory). The core assumptions of
fuzzy-trace theory are rooted in research on memory, judg-
ment, and decision making, and they take into account the
social, cognitive, affective, and developmental factors involved
in decision making. According to fuzzy-trace theory, people
encode multiple mental representations of their experience at
the same time. These representations range in specificity from
low (gist representations) to high (verbatim representations).
A gist representation is a fuzzy impression of the general mean-
ing of information or experience, supporting intuitive think-
ing. A verbatim representation is a mental representation of
exact details, supporting precise analysis. These assumptions
are required in order to account for specific counterintuitive
and seemingly contradictory results.
Intuition in fuzzy-trace theory is defined as fuzzy, im-
pressionistic thinking using vague gist representations.
654 Decision Making (Individuals)
Thus, mindless impulsive reaction (e.g., System 1 in standard
dual-process models) is distinguished from insightful intuition
that reflects understanding (e.g., gist in fuzzy-trace theory). In
other words, there are two kinds of fast and simple ways of
thinking: a stupid kind that represents the most primitive form
of thinking (e.g., System 1) and a smart kind that represents
the highest form of thinking, insightful intuition (i.e., gist).
In the foundations of mathematics (i.e., the basic assumptions
of the field), mathematical intuition is characterized as a
similarly advanced form of thinking.
Fuzzy-trace theory draws on evidence for independent gist
(basic meaning) and verbatim-memory (exact detail) represen-
tations of information, but it differs from other dual-process
models in emphasizing that there are degrees of rationality and
that intuition is an advanced form of reasoning. Such claims
are based on empirical evidence about the development of
reasoning (e.g., evidence comparing reasoning by children
and adolescents to that of adults and reasoning of adult
novices to that of experts). Research has supported the theory
that children's cognitive processes progress from detail-
oriented computational reasoning to more intuitive proces-
sing; in other words, people process less information more
categorically and qualitatively with age. Similar predicted
trends were found among experts versus nonexperts. For exam-
ple, Valerie Reyna and colleagues found that expert cardiolo-
gists processed fewer dimensions of information than less
expert physicians when deciding whether to admit patients
with chest pain to the hospital. Crucially, experts were able to
identify the few most influential factors and focus on those in
order to make faster and more accurate decisions.
The theory also predicts parallel development of verbatim-
based analysis and gist-based intuition, which produces devel-
opmental reversals (e.g., children outperform adults) under
specific circumstances. As an example, despite increasing com-
petence in reasoning, some biases in decision making grow
with age, producing more 'irrational' violations of coherence
among adults than among adolescents and younger children.
The latter phenomena are linked to developmental increases in
gist processing with age. The first framing study in children was
conducted to test predictions of fuzzy-trace theory. As pre-
dicted, framing effects actually increased from childhood to
adulthood, as do other biases that rely on semantic gist. With
greater experience and knowledge, decision making is pre-
dicted to become more simple and straightforward, that is,
based on gist.
These developments have implications for health and well-
being, especially regarding adolescent risk taking. Fuzzy-trace
theory has been applied to understanding how risk taking
changes from childhood to adulthood. In particular, adoles-
cence is a period of vulnerability to risk taking and poor
decisions, vulnerabilities that only increase bad outcomes in
young adulthood when parental supervision lessens. Surpris-
ingly, across many domains of problem behavior, adolescents
who take risks engage in more verbatim-based analysis, but risk
avoiders use gist-based intuition.
Fuzzy-trace theory, therefore, differs from other dual-
process approaches in three respects. First, intuitions are
considered an advanced form of reasoning. Developmental
research on children's learning and on expert decision making
supports this view.
Second, in addition to inculcating the bottom line meaning
of an experience, gist representations also incorporate emo-
tion, including valence (positive/negative; good/bad), arousal
(high/low), and feeling states. Emotion can color the interpre-
tation of a stimulus (i.e., its gist), especially when the stimulus
is ambiguous. In particular, emotion influences the encoding
of stimuli (i.e., how the choice is represented), the retrieval of
values and principles, and the implementation of values and
principles (i.e., their application to option representations).
Research supports the claim that decisions about risk made
in the presence of emotion are different from those made in
its absence.
Third, fuzzy-trace theory provides specific predictions
about rationality that contradict predictions made by standard
dual-process and other theories. Most notably, fuzzy-trace the-
ory predicts framing effects, but under a theoretical justifica-
tion different from that provided by prospect theory. In
addition, some effects predicted by fuzzy-trace theory cannot
be predicted by prospect theory. According to fuzzy-trace the-
ory, invariance is achieved through gist processing because
decision making is based on the substance of information
rather than on superficial details.
Subjective understanding: Construai level theory
Drawing on fuzzy-trace theory as well as other approaches,
construai level theory posits that people can have two different
subjective interpretations (construals) of the same event:
abstract and concrete. Moreover, each type of construai affects
decisions differently. High-level construals, like gist representa-
tions, are broad, abstract generalizations that capture the cen-
tral immutable features of an event and that help generate the
event's general meaning. Low-level construals, in contrast, are
like verbatim representations in that they consist of multiple,
narrow, concrete categorizations that direct attention to the
details of an event. The concept of psychological distance is
measured by construai level, such that low-level construals are
psychologically proximal (close) and high-level construals
are psychologically distal (far).
Construai level theory predicts that, as psychological dis-
tance increases, detailed information from that event becomes
less accessible, less accurate, and less reliable, whereas the
overall understanding becomes more abstract. For example,
in one study that tested high-level versus low-level constmals,
individuals were asked to imagine a set of scenarios (e.g., a
camping trip or a friend's visit to NY) that would occur in
either the proximal or distant future. For each scenario, parti-
cipants grouped a set of related objects (e.g., a snorkel or a tent)
into as many conceptual piles as they deemed appropriate. In
line with the research hypotheses, people in the high-level
construai group (i.e., those who received the distal cue) created
fewer piles with broader, more abstract criteria than did the
low-level group (i.e., those who received the proximal cue).
Research based on construai level theory suggests that people
are often overly optimistic about choice options in the future
because such events are represented in an abstract (high-level)
manner and are not readily accessible for cognitive processing
at a concrete (low) level.
Construai level theory also makes predictions about tem-
poral discounting, which is the phenomenon in which people
devalue (i.e., discount) outcomes that will occur in the future.
Decision Making (Individuals) 655
For example, people are willing to pay the retail price for a
computer they can take home immediately, but they expect a
discount if they have to wait for it. Research on temporal
discounting reveals that, faced with the choice between receiv-
ing a small reward now and receiving a larger reward received
2 weeks from now, most people opt for the immediate reward.
However, when both options involve a delay (e.g., receiving
the reward 2 weeks from now vs. 4 weeks from now), prefer-
ences reverse in favor of the larger reward. In both examples,
the options differ by the same amount of time (2 weeks);
so, the reversal of preferences suggests a categorical difference
in the way people value (constme) rewards that are available
immediately versus those they must wait for.
Research combining temporal discounting and construal
level theory tested how high- versus low-level construals affect
patience and self-control. Results over a series of experiments
demonstrated that people who used high-level construals
showed greater optimism about their future choices, greater
physical endurance over time (indicated by their ability to
squeeze a handgrip longer), and more self-control (indicated
by their preference for larger delayed rewards over smaller
immediate rewards) than did people who used low-level
construals.
According to construal level theory, different aspects of a
reward's desirability are more salient (i.e., cognitively accessi-
ble) at different delays. Specifically, for decisions about the
present, low-level, specific, and concrete features such as prac-
ticality should be most influential. In contrast, for decisions
about the distant future, high-level, abstract, and broad
features, such as compatibility with one's ideal self, should
be most influential. Accordingly, different decisions should be
attractive at different delays. Decisions whose low-level value
outweighs their high-level value should be more attractive in
the short term. Decisions whose high-level value outweighs
their low-level value should be more attractive in the long term.
One experiment that tested how different construals affected
self-control asked participants, "Right now, if you had to
choose between an apple and a candy bar, which would you
choose?" Subjects who preferred apples were defined as having
greater self-control in relation to dieting goals. After a pretest
activity that primed participants to either a high-level or a low-
level construal condition, the study showed that people in the
high-level construal condition chose apples over candy bars
more often than did those in the low-level condition. This
suggests that high-level construals support greater self-control.
By changing the cognitive construals that research subjects
used, researchers were able to alter the nature of temptation
impulses. Constmal level theory, therefore, has implications
for reducing impulsive decision making.
Emotional Influences on Decision Making
Emotions as Content in Decision Making
Affective or emotional processes have received increasing
emphasis in models of decision making. Affect cannot be re-
duced to only valence (good-bad) and intensity. The appraisal
of affect, how it is interpreted, also depends on the situation
and the specific emotion that is elicited. Specific emotions
are associated with specific action tendencies (e.g., fear with
the action of escaping). For example, risk taking has been
shown to be influenced by affect; people experiencing anger
or happiness tend to be risk taking, whereas people experien-
cing fear or sadness tend to be risk averse. Positive emotions
result in approach (i.e., increases in the value of associated
options), but negative emotions result in avoidance (i.e.,
decreases in the value of associated options). General feelings
of arousal can affect the choices being made even when the
choices did not actually generate the feelings. For example,
men who were aroused by pictures of attractive women were
more likely to choose a sooner smaller amount of money
relative to a later larger amount of money. Economists
describe this as impatience and psychologists describe it as
impulsivity.
Emotion: Supporting or Derailing Adaptive Decisions
Emotions can impair judgments, preventing goal attainment
by increasing susceptibility to environmental temptations.
However, emotions also provide clear evolutionary advantages
to decision making, despite their ability to interfere with
higher-level thinking. For example, the intrusion of emotion
on attention, which can interrupt goal-directed behavior, is
important in supporting the ability to detect and attend to
urgent threats. Emotional stimuli are detected more quickly
than are neutral stimuli, and people can maintain their focus
better on emotional relative to neutral stimuli in the presence
of distracters.
Affect as information
Affective signals can, consciously or unconsciously, facilitate
memory of past rewards and aversive experiences, compressing
this knowledge into a quickly accessible format. This is ad-
vantageous to rational decision making because analytical
attempts to reason through all relevant past experiences
would require a prohibitive amount of time and working
memory capacity. In this way, emotional or affective signals
automate the application of knowledge from past situations
to current situations without straining cognitive resources.
Consistent with this helping role of emotion is that, per
fuzzy-trace theory, gist-based evaluation of risk leads to more
adaptive decisions and increases with both age and expertise.
Common currency
Emotional and affective reactions to different types of stimuli,
both primary (e.g., food, sex, or drugs) and secondary (e.g.,
money or gambling chips), provide a way to assess the relative
value of disparate items on a single scale of subjective value.
Affective reactions can be used to evaluate current options
against counterfactual scenarios, past experiences, and imag-
ined future rewards or punishments (although such imagined
rewards are subject to temporal discounting).
Affective priming, mood congruency, and misattribution
In affective priming, an initial stimulus induces a mood that
persists and affects the evaluation of the second stimulus in a
valence-consistent manner. Subjects who experience a negative
mood in response to an initial stimulus often misattribute the
negative mood to the second stimulus. Even after the experi-
menter alerts subjects to the alternate explanation (i.e., that the
656 Decision Making (Individuals)
first stimulus, not the second, induced the negative mood),
subjects persist in their negative evaluations of the second
stimulus. Mood congruency is similar to affective priming
except that the subject's incidental emotions, rather than a
reaction to a prior stimulus, are what color the evaluation of
a novel stimulus in a manner consistent with the subject's
preexisting mood state.
Impulsivity and the Somatic Marker Hypothesis
Physiological cues of emotional state
One emotion-based model of decision making is the somatic
marker hypothesis, according to which decision making in
normal individuals is guided by emotional states produced
by bioregulatory processes. For example, while playing a gam-
bling card game, normal subjects develop a measurable physi-
ological (skin conductance) response in anticipation of risky
outcomes. Drug addicts and people with damage to the frontal
lobe of the brain fail to create an anticipatory skin conductance
response before drawing a card from a risky deck, and some
researchers take this to suggest that their impaired emotional
function and suboptimal decision making are connected via an
altered physiological response to risk.
The Neuroscience of Decision Making
Measurement Techniques
Neuroscientific models attempt to identify the brain mechan-
isms that are associated with reasoning. Recent technology has
developed methods for correlating neural (i.e., brain) activity
with different types of thinking. These techniques include
functional magnetic resonance imaging (fMRI), positron emis-
sion tomography (PET), and electroencephalography (EEG).
Additionally, diffusion tensor imaging (DTI) and magnetic
resonance imaging (MRI) are used to infer structural differ-
ences or developmental changes in brain regions. Although
decision making is one of the most extensively investigated
areas using these new techniques, it is important to understand
the limitations of the inferences that may be drawn from such
studies. Results associating brain activity with decision mak-
ing are correlational; that is, although brain activation may
co-occur with certain behaviors, there is insufficient evidence
to infer a causal relationship. Nonetheless, techniques such as
fMRI, PET, EEG, and DTI have the potential to greatly inform
models of decision making.
Brain Regions
The integration and evaluation of information from memory,
emotional cues, and sensorimotor stimuli is thought to occur
in the lateral prefrontal cortex and parts of the anterior cingu-
late cortex. These regions, dubbed areas of executive function,
relate to critical decision making behaviors, such as inhibiting
unnecessary or goal-defeating information, planning ahead,
and contemplating risks and rewards.
Areas including the amygdala, nucleus accumbens, and
medial prefrontal cortex are implicated in the processing of
salient social information, such as faces, attractiveness, and
other social judgments. Whereas the orbitofrontal cortex is
associated with simple emotional reactions such as fear, there
is evidence of selective activation of the anterior cingulate
cortex when considering the outcome of a decision requires
assessment of another person's attitude or behavior. For exam-
ple, the anterior cingulate cortex is active during cooperation
and trust games and is associated with better memory for social
versus nonsocial information. Since most real-life decisions
occur in a social context, this is an important component of
the decision making network.
Reward and Incentive
Another component of social and emotional decision proces-
sing is how much the individual values the reward that is likely
to result from a particular decision. Research suggests that
activity in the nucleus accumbens is linked to reward sensi-
tivity. Further, dopamine, a neurotransmitter implicated in
learning and reward, is concentrated in the nucleus accumbens
and has been strongly associated with reward-related behavior.
Incentive salience
Dopamine is not thought to be essential to the intrinsic evalu-
ation of a reward. Rather, its function is thought to lie in
the connection between that reward and the action (or series
of actions) necessary to obtaining it. The association of moti-
vational (reward) value with a behavior is called incentive
salience, or value binding. The generation and execution of
a reward-oriented action plan is associated with activity in
the dorsal striatum. The processing of the magnitude of the
reward obtained from an action, particularly if it differed from
the expected reward, is associated with activity in the anterior
cingulate cortex and ventral tegmental area.
Reward-prediction-error hypothesis
Some researchers have found evidence that changes in the
pattern of dopamine release in the ventral tegmental area of
the brain reflect the difference between the expected and the
actual reward obtained for a given behavior. These phasic
surges form a record of reward prediction accuracy. Thus, if a
reward is better than expected, there is a burst of dopamine,
which indicates a positive reward-prediction error. If a reward
is worse (i.e., either less valuable or more aversive) than
expected, there is a pause in the pattern of dopamine release,
indicating a negative reward-prediction error. Baseline-level
dopamine activity indicates that the reward was as expected.
Thus, the difference between what you got and what you
thought you would get (i.e., the reward-prediction error) is
thought to be recorded in the brain by corresponding increases
or decreases in dopamine release. The reward-prediction-error
hypothesis, supported by evidence from rodents and primates,
is thought to describe the way people learn the association
between a particular action and the magnitude and desirability
of the consequence.
Risk Taking in Adolescence
Adolescents are widely assumed to make riskier decisions than
either children or adults. This phenomenon is not explained by
theories of decision making that assume that the capacity for
rational decision making increases linearly with age, prompt-
ing attempts to explain this developmental incongruity. Two
classes of explanations have emerged.
Decision Making (Individuals) 657
Developmental asymmetry
One hypothesis posits that two brain networks important in
decision making develop at different rates and that, during the
period when one is more mature than the other, adolescents
are less able to make adaptive decisions. Some researchers
attribute adolescents' proneness to risky decision making
to a developmental asymmetry between the socioemotional
system (located in limbic areas of the brain), which matures
earlier and determines reward sensitivity, and the cognitive
control system (located in the prefrontal cortex), which
matures later and governs planning and inhibition. According
to this account, over the course of development, the functional
connectivity of these two regions improves. As a result, it is
thought to be the reliance on affect over cognition in adoles-
cence and a deficit in connectivity between areas of emotional
processing and areas of cognitive control which cause this
group's proclivity for risky decision making. Naturally, brain
anatomy and connectivity are thought to be important in risky
decision making, but the details of this mechanism are far from
established.
Gist versus verbatim processing
Adolescents do not, as many believe, underestimate risks or
consider themselves to be invincible. Converging data show
that they tend to overestimate risk and often overestimate their
vulnerability to negative outcomes. Adolescents possess the
basic cognitive capacities to distinguish risky from riskless deci-
sions; their deficiency lies in overanalyzing the situation - that
is, in tending to favor calculative over gist-based decision
making when it comes to risk. In terms of fuzzy-trace theory,
adolescents tend to use verbatim processes more often than do
adults (who use gist processes) to make decisions involving risk.
As an example, consider the proposition to play Russian
roulette with a six-chamber gun for a gamble of 6 million
dollars. In Russian roulette, a gun is loaded with only one bullet
and fired to the player's head. If the chamber containing the
bullet comes up when the trigger is pulled, the player dies.
Otherwise, one of the empty chambers comes up, and the player
wins the money. Gist and verbatim thinking result in drastically
different approaches to the proposition to play Russian roulette.
A verbatim thinker will calculate the expected reward of the
risk, that is, the desired outcome weighted by the probability of
attaining that outcome. In this example, this means that there
is a five-sixths chance of winning a 6 million dollar prize,
which equals an expected reward of 5 million dollars. Given
this high expected reward, a verbatim thinker might accept the
risk and play the game. By choosing to play, the verbatim
thinker elects to take a risk that has a potentially disastrous
outcome (i.e., death) so long as there is also the chance for a
very high reward. Verbatim thinkers do factor in the high
negative value of death; however, there is a price that can be
paid that will compensate for it.
To the contrary, a gist thinker will extract the bottom-line
meaning from each option. Since the bottom line of Russian
roulette is that you could die, gist-based thinkers will refuse to
play, no matter what the potential reward is, because the
negative outcome is catastrophic. Gist thinkers are able to
tolerate a small chance of death for necessary risks (such as
crossing the street) but not for unnecessary ones (like having
unprotected sex).
According to multiple experiments, adolescents tend to
approach risky decisions with verbatim-based analysis, whereas
adults take a gist-based approach to the same risk. Studies of
real-life risk taking also are consistent with this view; adoles-
cents' perceived risks and perceived benefits predict self-reported
risk taking (e.g., substance use), in contrast to hypotheses that
such behavior is mainly impulsive and unintentional.
The better decision outcomes associated with gist-based
thinking suggest that intervention programs should foster this
type of thinking. The idea that risk taking tendencies cannot be
altered, nor outcomes improved, because of the immaturity
of brain development has been falsified by the effectiveness of
gist-based and other educational programs. As a result, inter-
vention programs might do best to shelter younger adolescents
from opportunities to engage in risky behaviors, given that
they are not fully equipped to make healthy decisions, but to
instmct older adolescents in better decision making skills.
Recent evidence supports the efficacy of this approach.
Summary
Although scholars continue to debate key aspects of the
decision making process, theories have historically shifted
from single-process and psychophysical approaches to a vari-
ety of dual-process approaches. Current dual-process models
such as fuzzy-trace and construal level theories are based on
evidence that people mentally encode multiple representations
of the same event, and cuing different representations can
result in different decisions. This observation is especially rele-
vant in the area of risk taking among adolescents; in particular,
adolescents who trade off risks for rewards seem to take more
risks than do those who take a categorical stance of avoiding
unnecessary risk. Current theories also account for the complex
role of emotion in the decision process, acknowledging that
this influence is sometimes beneficial and sometimes detti-
mental. Recent neuroscience research has produced neural
hypotheses about several phenomena, including reward val-
uation and cognitive control. As future research explores
synergies between behavioral psychology and behavioral neu-
roscience, theories from both realms should be mutually
constraining.
See a/so/ Addictions and Adolescence; Adolescence; Brain and
Behavior Relationships; Brain Chemicals: Global Projections of
Ancient Aromatic Neurotransmitters; The Brain; Cognitive Bias;
Memory; Neurotechnologies; Uncertainty.
Further Reading
Casey BJ, Getz S, and Galvan A (2008) The adolescent brain. Developmental Review
28: 62-77. doi:1 0.1 01 6/j.dr.2007.08.003.
DeMartino B, Kumaran D, Seymour B, and Dolan RJ (2006) Frames, biases, and rational
decision-making in the human brain. Science 313: 684-687. doi:10.1126/
science.1 128356.
Fujita K, Trope Y, Liberman N, and Levin-Sagi M (2006) Construal levels and
self-control. Journal ol Personality and Social Psychology 90: 351-367.
doi:1 0.1 037/0022-351 4.90.3.351 .
Kahneman D (2003) A perspective on judgment and choice: Mapping bounded rationality.
American Psychologist 58(9): 697-720. doi :1 0.1 037/0003-066X.58.9.697.
658 Decision Making (Individuals)
Kuhberger A and Tanner C (2010) Risky choice framing: Task versions and a
comparison of prospect theory and fuzzy-trace theory. Journal of Behavioral
Decision Making 23(2): 314-329. doi:1 0.1 002/bdm.656.
Reyna VF (2004) How people make decisions that involve risk: A dual-process
approach. Current Directions in Psychological Science 13: 60-66. doi:10.1111/
j.0963-721 4.2004.00275.x.
Reyna VF and Brainerd CJ (1995) Fuzzy-trace theory: An interim synthesis [The first
issue of 1995 was devoted to this invited target article on a theory developed by
Valerie Reyna and colleagues]. Learning and Individual Differences 7(1 ): 1-75.
doi:1 0.1 01 6/1041 -6080(95)90031 -4.
Reyna VF and Ellis SC (1994) Fuzzy-trace theory and framing effects: Children’s risky
decision making. Psychological Science 5(5): 275-279. doi:10.1 111/
j.1 467-9280.1 994.tb00625.x.
Reyna VF, Nelson W, Han P, and Dieckmann NF (2009) How numeracy influences risk
comprehension and medical decision making. Psychological Bulletin
135: 943-973. doi:1 0.1037/a001 7327.
Reyna VF and Rivers SE (2008) Current theories of risk and rational decision making.
Developmental Review 28(1): 1-11. doi:10.1016/j.dr.2008.01 .002.
Rivers S, Reyna VF, and Mills BA (2008) Risk taking under the influence: A fuzzy-trace
theory of emotion in adolescence. Developmental Review 28(1): 107-144.
doi:1 0.1 01 6/j.dr.2007.1 1 .002.
Steinberg L (2008) A social neuroscience perspective on adolescent risk-taking.
Developmental Review 28(1): 78-106. doi:10.1016/j.dr.2007.08.002.
Tversky A and Kahneman D (1981) The framing of decisions and the psychology of
choice. Science 211: 453-458. doi:1 0.1 1 26/science.7455683.
Relevant Websites
http://www.informs.org/Community/DAS - Decision Analysis Society.
www.sjdm.org - Society for Judgment and Decision Making.
www.smdm.org - Society for Medical Decision Making.
Defense Mechanisms
G E Vaillant Massachusetts General Hospital, Boston, MA, USA
© 2012 Elsevier Inc. All rights reserved.
The boundary between pathology and adaptation is ambigu-
ous. Whereas early nineteenth-century medical phenomenolo-
gists viewed pus, fever, pain, and coughing as evidence of
disease, late nineteenth-century pathophysiologists learned to
regard these symptoms as evidence of the body's healthy efforts
to cope with physical or infectious insult. Similarly, defense
mechanisms (often termed involuntary coping mechanisms)
produce behaviors that appear pathological but in fact reflect
the brain's homeostatic effort to cope with sudden changes in
its internal and external environment.
In 1856, Claude Bernard, a French physiologist and a
founder of experimental medicine, started us on our under-
standing of adaptation to stress when he wrote, "We shall
never have a science of medicine as long as we separate the
explanation of the pathological from the explanation of nor-
mal, vital phenomena." Put differently, it is not stress that kills
us as much as the homeostatic, often seemingly pathologic,
response to stress that permits us to survive.
In 1925, Adolph Meyer, a founder of modern American
psychiatry, asserted that there were no mental diseases; there
were only characteristic reaction patterns to stress. Meyer's
point was that although adaptive mental 'reaction patterns'
like denial, phobias, and even projections can appear to reflect
illness, they may in fact be 'normal, vital phenomena' related
to healing. Just as immune mechanisms and clotting mechan-
isms heal by distorting bodily equilibrium, defense mechanisms
heal by distorting mental processes.
Of course, the symptoms of organic brain damage usually
reflect disease and not adaptation. Manic-depressive psychosis
is due to a genetic defect; the mental devastation produced by
alcoholism is due to poisoning; and the negative symptoms of
schizophrenia reflect brain pathology and not adaptation.
Nevertheless, much of the so-called mental illness, including
some delusions and hallucinations, makes the disruption of
disordered brain function easier to bear; and defenses are often
analogous to the healthy, but red, hot, and tender swelling -
the rubor, color, dolor, and turgor - that immobilizes a frac-
ture so that it may heal.
Much of what is called mental illness in our diagnostic
nomenclature - the symptoms of many anxiety disorders,
some depressions, and most personality disorders - is the out-
ward manifestation of homeostatic struggles to adapt to life.
Admittedly, analogous to acne and autoimmune disease, such
reaction patterns often reflect rigid, poorly modulated, and
pathologic efforts at adaptation.
Norman Sartorius, the director of the Division of Mental
Health of the World Health Organization, wrote,
Research during the past two decades failed to provide evidence that
could help to create disease concepts and disease entities in psy-
chiatry . . . Other ways of thinking about health and disease, mind
and body, mental and physical, individual and social are needed . . .
I believe that in selected instances a return to the allegedly outdated
Meyerian reaction patterns and Freudian Defense mechanisms is
warranted.
Defense mechanisms are patterns of emotions, thoughts, or
behaviors that are relatively involuntary. They arise in response
to perceptions of psychic danger or conflict to unexpected
change or threat in the internal or external environment or
in response to cognitive dissonance. Defense mechanisms
obscure or diminish cognitive, emotional, and physiological
response to stress that if unmitigated would give rise to depres-
sion, anxiety, or impulsive behavior. There is increasing
evidence that choice of defensive styles makes a major contri-
bution to individual differences in response to stressful envir-
onments. Consider the differing response of individuals riding
on a rollercoaster that range from severe anxiety to stress reliev-
ing excitement.
Defense mechanisms can alter our perception of any or all
of the following: subject (self), object (other), idea, or feeling
(affect). Defenses alter the relationship between self and object
and between idea and affect. In projection, for example, 'I hate
him' becomes 'He hates me.'
The task of homeostatic mechanisms, in general, is to buffer
sudden change. The task of the defenses is to buffer sudden
change in the four 'lode stars' of mental conflict. These lode-
stars are relationships, emotions, conscience, and reality. The
use of defense mechanisms usually alters perception of both
internal and external reality, and often, as with hypnosis, the
use of such mechanisms compromises other facets of cognition
(Figure 1).
Thus, defenses can mitigate sudden unresolvable conflict
with important people, living or dead. Defenses serve to keep
affects and ideation within bearable limits during sudden
changes in relationships, such as following the unexpected
death of a loved one. People become a lodestar of conflict
when we cannot live with them and yet cannot live without
them; as people exist within us as well as without, internal
representations of important people may continue to haunt us
and to cause conflict for decades after we have ceased to live
with them and thus evoke defensive behavior. Ghosts, ancestor
worship, and demonic possession would be such examples.
Defenses can deflect or deny sudden increases in emotions
and biological instinct, such as heightened sexual awareness
and aggression during adolescence. Psychoanalysts call this
lodestar of conflict 'id,' fundamentalists call it 'sin,' cognitive
psychologists call it 'hot cognition,' and neuroanatomists point
to the hypothalamic and limbic regions of brain.
Defenses enables individuals to mitigate sudden upsurges
in conscience or guilt such as might occur for a doctor faced
with a malpractice suit. Psychoanalysts call it 'super ego,'
anthropologists call it 'taboos,' behaviorists call it 'condition-
ing,' and neuroanatomists point to the frontal lobe and the
amygdala. This lodestar is not just the result of admonitions
from our parents that we absorb before age five, but is formed
by our whole identification with our culture, and sometimes
by irreversible learning resulting from overwhelming trauma.
Finally, defenses allow individuals a period of respite to
master reality, such as the events surrounding 9/11 or changes
659
660 Defense Mechanisms
Conscience
(cultural taboos and Relationships
imperatives, super ego)
Ego (integrated central nervous system)
(denies, distorts, represses
inner and/or outer reality to reduce anxiety and depression)
Emotion Reality
(instinct, id, ‘drive’
passion, affect)
Figure 1 The four sources of intrapsychic conflict.
in self-image that cannot be immediately integrated, as might
result from an amputation, or a promotion. Such involuntary
mental mechanisms can provide a mental time out to adjust to
sudden changes in reality and self-image, which cannot be
immediately integrated. Sudden good news - the instant tran-
sition from student to physician or winning the lottery - can
evoke involuntary mental mechanisms as often as can an
unexpected accident or a diagnosis of leukemia.
Sudden change in any of the four lodestars produces de-
pression, anxiety, or impulsive disorder. Involuntary 'denial,'
'distortion,' or 'repression' of the lodestars mitigates such dys-
phoria. However, used indiscriminately, terms like repression
and denial lose the explanatory power of the conceptual
scheme - ego mechanisms of defense.
A final reason for paying attention to defenses can be illu-
strated by the internist's understanding of referred pain. A pain in
the right shoulder may reflect an inflamed gall bladder; a pain
in the left shoulder may reflect coronary thrombosis. Proper
diagnosis depends upon the internist's seeing behind a symptom
often identified as arthritic. By understanding that a function of
much of the psychopathology is to distort, displace, and deny
the conflict, we leam not to take psychological symptoms too
literally. In short, defenses provide a diagnostic template for
understanding the distress and for guiding the clinical manage-
ment of psychology's most 'unreasonable' and baffling clients.
At first, the observer may call the user of a defense stupid,
then wicked, and then mentally ill. All such responses are as
useless as scolding a child for sneezing. Yet, we can neither
condemn a tubercular patient for coughing nor ignore the
internal disorder that it portends. Similarly, we must neither
condemn the lawful eccentricities of the ego's mechanisms of
defense as sins nor dismiss them as mere chance events. As
with the cough of the consumptive, attention must be paid to
the odd behaviors of people with emotional disease.
It is important to place defense mechanisms within the
broader framework of stress and coping. Psychopathological
response to stress may be considered from two perspectives.
The first perspective emphasizes relatively nonspecific psycho-
physiological responses to stress, such as an elevated Hamil-
ton depression scale or elevations in blood epinephrine
levels. This is the perspective of Hans Selye and other phe-
nomenologists. Stress kills. The second perspective focuses on
how an individual's integrated central nervous system perceives
and copes with the stressor. In this case, the result of stress is
seen not as nonspecific depression and anxiety but as highly
differentiated synthetic behaviors that may range from arcane
delusional systems to the writing of a great novel. What does
not kill us makes us stronger. This second perspective, that of
ego psychology, is the orientation of this article.
If stress can be viewed from two vantage points, coping
responses to stress can be viewed as divided into three broad
categories. The first coping category involves eliciting help
from appropriate others, for example, by mobilizing social
supports. The second coping category involves voluntary cog-
nitive efforts like information gathering, anticipating danger,
and rehearsing responses to danger. The third coping category
entails deploying involuntary adaptive mechanisms or auto-
matic psychological processes that protect the individual from
the disorganizing effects of internal or external dangers or
stressors. These coping processes are often subsumed under
the psychoanalytic term ego mechanisms of defense.
The first class of coping mechanisms, seeking social support, is
analogous to calling 911 in response to one's own bleeding. The
second class, cognitive coping strategies, can be taught and re-
hearsed; they are analogous to consciously using a tourniquet to
stop one's own bleeding. The third class of coping mechanisms, the
involuntary defense mechanisms, are analogous to depending upon
one's own very complex and quite involuntary clotting mechanisms
in order to stop bleeding.
Freud’s Discovery of the Concept of Defense
Perhaps, Freud's most original contribution to human psychol-
ogy was his inductive postulation that unconscious 'defense
mechanisms' protect the individual from painful emotions,
ideas, and drives. In delineating the nature of ego mechanisms
of defense, Freud established not only that upsetting affects, as
well as ideas, underlay psychopathology, but also that much of
what is perceived as psychopathology reflects a potentially
healing process.
That emotions were significant to humans had been known
since ancient times, but our understanding of their modulation
through unconscious mechanisms of defense, rather than will-
power, originated with Sigmund Freud, who was trained in
both neurology and physiology.
It was difficult to develop a concept of defense, however,
until psychology could imagine that ideas gain their color,
temperature, and importance from emotion. For example, the
influential textbooks of the two greatest psychologists of the
nineteenth century, Wilhelm Wundt and William James persisted
in discussing a psychology that consisted of only cognition,
not emotion. Psychology needed to replace neutral terms like
intellect, will, moral sentiment, higher feelings, and cognition
with terms suffused with emotional color, for example, sadness,
lust, rage, and guilty fantasy. From the beginning, Freud's work
effected this change. In 1 894, Freud observed that not only could
affect be 'dislocated or transposed' from ideas (by the mechanism
that Freud would later call isolation) but also that affect could be
'reattached' to other ideas (i.e., by displacement).
Over a period of 40 years, Freud discovered most of the
defense mechanisms that we speak of today and identified
five of their important properties.
Defense Mechanisms 661
(1) Defenses are a major means of mitigating the distressing
effects of both emotion and cognitive dissonance. (2) They
are unconscious. (3) They are discrete from one another.
(4) Although often the hallmarks of major psychiatric syn-
dromes, defenses are dynamic and reversible. (5) Finally, they
can be adaptive, even creative, as well as pathological. To this
list must be added that to the user defenses are invisible, but to
the observer defenses usually appear odd.
Unlike most of the defense mechanisms, the concept of
repression was not original to Freud. J.F. Herbart had written
extensively on 'verdrangung' (Freud's term for repression also)
of ideas. Herbart's Psychology as a Science (1824) may have
influenced Freud indirectly through Freud's psychiatric men-
tor, Theodor Meynert. Schopenhauer had also perceived that
defensive forgetting had something to do with psychopathol-
ogy; but he, too, wrote of forgetting ideas and circumstances,
not feelings.
In 1905, Freud conceived of a special class of defense
mechanisms - sublimations - that could transmute conflictual
affect, not into a source of pathology, but into culture and
virtue. For example, Beethoven was often angry, depressed,
lonely, and suicidal. Nevertheless, instead of killing himself,
in his Ninth Symphony he put Schiller's 'Ode to Joy' to music.
To call such unusual behavior 'being in denial' is to sell the
ingenuity of the central nervous system short.
Humor, like sublimation, was another defense that Freud
and students of adaptation in general have perceived as having
more of an adaptive than a pathogenic quality. In Jokes and
Their Relation to the Unconscious, by hinting at a hierarchy of
defenses, Freud developed the idea of humor as a more mature
defense mechanism than wit. Freud viewed wit and practical
jokes as inextricably linked with the less adaptive, less mature,
and less empathic mechanism of displacement. Freud wrote,
"humor is a means of obtaining pleasure in spite of the dis-
tressing affects that interfere with it; it acts as a substitute for the
generation of these affects; it puts itself in their place. It arises
from an economy in the expenditure of affect."
Freud also introduced the concept of an ontogeny of
defenses. Defenses such as projection, repression, and sublima-
tion not only lay along a continuum of relative psychopathol-
ogy but also along a continuum of personality development.
Reaction formation, Freud said, reminded him of the proverb
JungeHure, alte Betschwester (a young whore makes an old nun).
With the passage of decades and the continuing myelinization
of frontal lobe's connections to the limbic system, the defense
of adolescent acting out (e.g., impulsive, temporarily comfort-
ing tantrums) could become the parent of reaction formation
(anger is bad and should be punished by spanking and jail) and
a potential grandparent of altruism (nonviolence, creative
parole and believing that conflict mediation is good).
In 1936, Freud advised the interested student that there are
"an extraordinarily large number of methods (or mechanisms,
as we say) used by our ego in the discharge of its defensive
function . . . my daughter, the child analyst, is writing a book
upon them." He was referring, of course, to Anna Freud's The
Ego and Mechanisms of Defense (1936), which to this day
remains one of the definitive texts on the subject.
Clinicians have often been reluctant to view defenses as
adaptive as well as pathological. This Encyclopedia, for exam-
ple, has a separate article for 'coping.' This difficulty has slowed
our understanding of the more mature ego mechanisms of
defense. The difficulty, of course, is that with brain maturation
comes complexity. Thus, self-deception or conflict resolution
via humor and sublimation reflects a different level of mental
complexity than via the more concrete defenses of projection
or turning-against-the-self.
Modern Conceptualizations of Defense
AH defenses involve creative synthesis. Defenses often manifest
themselves in paintings and in dreams, in phobias and in
hallucinations, in humor and in religious experience. As
such, defenses reflect the building blocks of interesting behav-
ior, of madness and of art. They help explain the unreason of
both the sinner and the saint. However, if defenses integrate,
make sense out of and master conflict, they still remain diffi-
cult to describe. Like rainbows and shooting stars, now we see
them, now we do not.
From the beginning, defenses have posed a problem for
experimental psychology. First, there is no clear line between
character (enduring traits) and defense (shorter lived response
to environmental stress), between behavior and mental mech-
anism, and between neuropathology and unconscious coping
process. Is the self-attack of major depression a result of geneti-
cally disordered brain chemistry or anger turned against the
self - or both? Second, 'defense mechanisms' can serve other
purposes; and conversely, any of the mind's functions, not just
standard defenses, can be employed in the service of defense.
Third, in any effort to produce a comprehensive list of defenses
there will be enormous semantic disagreement.
Differentiated mechanisms of defense become most clear
when one can study the psychopathology of healthy everyday life
in detail and over time. Our appreciation of the defensive nature
of mature behavior awaited studies of normal populations. Over
the last 30 years, several longitudinal studies at Berkeley (Norma
Haan) and at Harvard (George Vaillant) have clarified our under-
standing of defenses with experimental and developmental stud-
ies. By offering a tentative hierarchy and glossary of consensually
validated definitions, DSM-IV has set the stage for further prog-
ress. In the last 20 years, several empirical studies reviewed by
Cramer (1991) and Skodol and Perry (1993) finally organized
defenses into a hierarchy of relative psychopathology.
The Diagnostic and Statistical Manual of Mental Disorders,
fourth edition (DSM-IV), has included a Defensive Function-
ing Scale as a proposed diagnostic axis for further study. Table 1
arranges these defense mechanisms into seven general classes
of relative psychopathology. The so-called 'immature' defenses
(Levels II-V) in Table 1 are the ones that underlie much of
personality disorder.
There are many ways that we can ignore or distort one or
more of these four lodestars in order to mitigate intrapsychic
conflict. For semantic consistency, the Defensive Function
Scale of DSM-IV has labeled these involuntary mental mechan-
isms, defenses (Table 1). These coping mechanisms or
'defenses' can abolish impulse (e.g., by reaction formation),
or social learning (e.g., by acting out), or other people (e.g., by
schizoid fantasy), or reality (e.g., by psychotic denial). They
can abolish our conscious recognition of the subject (e.g., by
projection), or the object of a conflict (e.g., by turning against
662 Defense Mechanisms
Table 1 Defensive functioning scale (adapted from DSM-IV) (pp. 751-753)
A. Psychotic defenses
I. Level of defensive deregulation. This level is characterized by failure of defensive regulation to contain the individual’s reaction to stressors, leading to
a pronounced break with objective reality. Examples are as follows:
• Delusional projection
• Psychotic denial
• Psychotic distortion
B. Immature defenses
II. Action level. This level is characterized by defensive functioning that deals with internal or external stressors by action or withdrawal. Examples are as
follows:
• Acting out
• Apathetic withdrawal
• Passive aggression (turning against the self)
III. Major image-distorting level. This level is characterized by gross distortion or misattribution of the image of self or others. Examples are as follows:
• Autistic fantasy
• Projective identification
• Splitting of self-image or image of others
IV. Disavowal level. This level is characterized by keeping unpleasant or unacceptable stressors, impulses, ideas, affects, or responsibility out of
awareness with or without a misattribution of these to external causes. Examples are as follows:
• Denial
• Projection
• Rationalization
V. Minor image-distorting level. This level is characterized by distortions in the image of the self, body, or others that may be employed to regulate
self-esteem. Examples are as follows:
• Devaluation (hypochondriasis)
• Idealization
• Omnipotence
C. Neurotic efenses
VI. Mental inhibitions (compromise formation) level. Defensive functioning at this level keeps potentially threatening ideas, feelings, memories, wishes,
or fears out of awareness. Examples are as follows:
• Displacement
• Dissociation (empirically placed with the 1 1— V ‘immature’ defenses)
• Intellectualization
• Isolation of affect
• Reaction formation
• Repression
• Undoing
D. Mature defenses
VII. High adaptive level. This level of defensive functioning results in optimal adaptation in the handling of stressors. These defenses usually maximize
gratification and allow the conscious awareness of feelings, ideas, and their consequences. They also promote an optimum balance among
conflicting motives. Examples of defenses at this level are as follows:
• Anticipation
• Affiliation (seeking social support)
• Altruism
• Humor
• Self-assertion (cognitive coping strategy)
• Self-observation (cognitive coping strategy)
• Sublimation
• Suppression
the self), or awareness of the idea (e.g., by repression), or of the
affect associated with the idea (e.g., isolation of affect).
All classes of defenses in Table 1 are effective in 'denying' or
defusing conflict and in 'repressing' or minimizing stress; but
they differ greatly in the psychiatric diagnoses assigned to their
users and in their consequences for long-term biopsychosocial
adaptation. In Level I, the most pathological category, are
found denial and distortion of external reality. These mechan-
isms are common in young children, in our dreams, and in
psychosis. To breach them requires altering the brain by neu-
roleptics or waking the dreamer.
More common to everyday life are the relatively maladap-
tive defenses found in Levels II-V. Defenses in these categories
are common in adolescents, in immature and substance abus-
ing adults, and in individuals with personality disorders. They
often make others more uncomfortable than the user. Such
defenses are consistently and negatively correlated with global
assessment of mental health, and they profoundly distort the
affective component of interpersonal relationships. Like ciga-
rette smoking in a crowded elevator, such behavior seems
innocent to the user and deliberately irritating and provocative
to the observer. Immature defenses externalize responsibility
and allow individuals with personality disorders to refuse
help. Defenses in this category rarely respond to verbal inter-
pretation alone. The projection of the paranoid by which
unacknowledged feelings are attributed to others is well
Defense Mechanisms 663
known, and so is the capacity of schizoid individuals to relieve
loneliness by creating fantasized human relationships within
their own minds. The capacity of histrionic personalities to
disassociate themselves from painful emotion and to replace
unpleasant with pleasant affect, as if they were on stage, is
familiar; and so is the fact that passive-aggressive and maso-
chistic characters turn anger against themselves in a most
annoying and provocative manner.
The mechanisms of acting out and hypochondriasis are less
familiar. In the antisocial personality, acting out reflects a
process whereby the direct motor expression of an unconscious
wish or conflict allows the individual to remain unaware of
either the idea or the affect that the action accompanies. Thus,
acting out produces the clinical illusion that all sociopaths do
not experience the painful affects of guilt, anxiety, or depres-
sion. In fact, nothing could be further from the truth.
Hypochondriasis disguises reproach. The 'borderline's' help
rejecting complaints that clinicians make him/her worse may
conceal grief, longing, and unacceptable aggressive impulses.
By exaggerated focus upon current somatic or psychic pain that
cannot be relieved, the hypochondriac attempts to manage
past unbearable grief or abuse.
Immature defenses can be breached in two ways. First, by
confrontation - often by a group of supportive peers - or by
videotaped empathic but focused psychotherapy. Second,
immature defenses can be breached by improving intrapsychic
competence by rendering the individual less anxious and
lonely through empathy, less tired and hungry through food
and rest, or less intoxicated through abstinence.
The third class of defenses, those in Level VI, are often
associated with what DSM-IV calls Axis I anxiety dis-
orders and with the psychopathology of everyday life. These
include mechanisms like repression, intellectualization, reac-
tion formation (i.e., turning the other cheek), and displacement
(i.e., directing affect at a more neutral object). In contrast
to the 'immature' defenses, these 'intermediate' defenses are
manifested clinically by phobias, compulsions, obsessions,
somatizations, and amnesias. Also, in contrast to the immature
defenses, 'intermediate' defenses usually make the user more
uncomfortable than the observer. Such users often seek psy-
chological help and respond more consistently to psychothera-
peutic interpretation. Intermediate defenses are common in
everyone from 5 years old until death. They are neither healthy
nor unhealthy. Across studies, the intermediate (Level VI or
'neurotic') defenses show mixed positive and negative correla-
tions with outcome of only minimal statistical significance.
The mechanisms at Level VII still distort and alter feelings,
conscience, relationships, and reality; but they achieve these
alterations gracefully and flexibly. These mechanisms allow the
individual consciously to experience the affective component
of interpersonal relationships, but in a tempered empathic
fashion. Thus, the beholder regards Level VII adaptive defenses
as virtues, just as the beholder may regard common immature
defenses like the prejudice of projection and the tantrums of
acting out as sins.
Doing as one would be done by (altruism), keeping a stiff
upper lip (suppression), keeping future pain in awareness
(anticipation), the ability not to take one's self too seriously
(humor), and turning lemons into lemonade (sublimation)
are the very stuff from which positive mental health is made.
Unfortunately, like tightrope walking, mature mechanisms
cannot easily be deployed; their deployment is facilitated by
empathy, safety, and exemplars.
Deciphering a nomenclature of defenses is like translating a
foreign language. Thus, the terms that I have found useful in
my longitudinal research (given in bold in Table 1) are trans-
lated in Table 2 from label into process. As is clear from
Table 2, 1 have found it useful to condense the behaviors listed
in Levels II-V into a single level - immature defenses. Rather
than dispute my or DSM-IV choice of words, readers are asked
to substitute terms that suit them. The fictional defensive pro-
cesses described for the cases depicted in Table 2 are exten-
sively illustrated by cases from true longitudinal study in
Adaptation to Life and Wisdom of the Ego.
All classes of defenses are effective in 'repressing' conflict and
in 'denying' stress, but the individual defenses differ greatly in
the diagnoses associated with their use and perhaps in their
consequences for long-term biopsychosocial adaptation.
Writing great tragedy (i.e., sublimation), for example, is
financially rewarding, instinctually gratifying, and sometimes
life saving for the author. The 'distortion' involved in stoicism
(suppression), humor, and altruism seems as ethical and as
mentally healthy to an observer as the immature defenses seem
immoral and antisocial. Living one's life by laughter, the
Golden Mean, and the Golden Rule is not just the stuff of
morality tales but can empirically make the user healthy,
wealthy, and wise.
Less mature defenses may evolve into more mature defenses. The
adolescent's autistic fantasies may become the scholar's intellectua-
lized obsession. Mary Baker Eddy's hypochondriasis evolved first
into reaction formation against any complaint of physical illness
and then into the highly rewarding altruistic founding of the Chris-
tian Science Church. Just as some children outgrow allergies, the
second-grader's irritating sadomasochistic pleasure of putting
thumbtacks on chairs may evolve into the adolescent's enjoyment
of the displaced hostility of cartoon movies and finally into mature,
enjoyable humor where the hostile intent is invisible.
However, if defenses are the building blocks of psychopa-
thology, their division into 'levels' or categories is as arbitrary
as our division of colors. There will always be as many different
colors - and as many defenses - as the taxonomist has
imagination.
Identification of Defenses
Identification of defenses is difficult. Rarely can we identify our
own defenses, and we often fail to recognize them in others or,
still worse, 'project' our own defenses onto others. To identify a
defense, an objective observer needs to triangulate past truth
with present behavior and subjective report. First, we need to
know the biological facts. Second, we need a behavior or
symptom or a creative product (that may seem odd, sinful,
or amazing). Third, we need the subjects' autobiographical
account. In short, we must contrast a patient's symptoms (or a
healthy person's creative product) with his self-report (mental
content) and with someone else's objective report (biography, or
old charts). Llltimately, like other facets of mental health, the
reliable identification of healthy but involuntary coping
mechanisms requires longitudinal study. For example, one
664 Defense Mechanisms
Table 2 Case illustration
A woman, married at age 30, after one miscarriage tried for 7years to have children, which she wished for intensely (limbic emotion). Then, following a
cervical biopsy that showed cancer, at age 38 she is still hospitalized after a total hysterectomy (reality). She had always felt inadequate in relation to
her younger sister, who already had four children and had been the one in the family who won praise as ‘being good with kids’ (relationship). The
woman’s husband also desperately wanted children (conscience). Below are a number of possible responses to her surgery.
D. ‘Mature’ defenses
Altruism-. The woman, a month after surgery, organized a group of other women who had breast and uterine surgery to counsel and to visit patients
undergoing gynecological surgery. They tried to give information, advice, and comfort, and from their experience provide answers to questions and
fears that such new patients might have. Diagnosis: Altruist
Suppression -. She read Marcus Aurelius and Ecclesiastes in the hospital. She took great care to hide her tear-stained tissues from her husband and made
no complaint (even though the process was painful) while her sutures were removed. Knowing that baby pictures upset her, she deliberately gave away
an unread copy of her favorite magazine, which featured an article on childcare. Diagnosis: Stoic
Sublimation-. She got great pleasure from 'get well’ cards from her sister’s children, agreed to teach a Sunday School class of preschoolers, and got
a poem published in her hometown weekly on the bittersweet joys of the maiden aunt. Diagnosis: Artist
Anticipation-. Her doctor was surprised to find out how relaxed and practical she was about her postoperative course and the calm frankness with which
she could express her regret at being cheated of children. His surprise was due to the fact that she had spent her preoperative visit anxiously worrying
about possible surgical complications and weeping over the fact that she would never be able to bear children. Diagnosis: None
Humor-. She laughed so hard that tears came to her eyes and her ribs ached when she read the Playboy definition of hysterectomy: “throwing out the
baby carriage but keeping the playpen.” She explained her private mirth to a startled and curious nurse with, “The whole thing is just so damned
ironic.” Diagnosis: None
C. ‘Neurotic’ defenses
Reaction formation-. She renewed her old college interest in Planned Parenthood and passionately argued with her younger friends to limit their families.
She suddenly ‘remembered’ that she had always been afraid of the pain of childbirth and remarked to her husband how lucky she was to be spared the
burden. Diagnosis: Pollyanna
Isolation (Intellectualization): She read a lot about uterine cancer and asked the doctor a great many questions about the nature of the operation. She
concerned herself with minute details preventing postoperative infection and caring for her operative wound. She made a hobby in the hospital of
learning medical words. Diagnosis: Obsessive Compulsive
Repression-. She found herself unable to rememberthe name of the operation except that it was for ‘a little nubbin in my tumtum.’ She ‘forgot’ her first
follow-up visit to the physician. On coming home, she broke into tears when she broke an inexpensive, amphora-shaped flower vase; she had no
idea why. Diagnosis: Hysteric
Displacement-. She became very interested in growing tulips and daffodils in her hospital window. Although she never asked the doctor questions about
her own hospital course, she worried about a funny mold on the bulbs she was growing. Knowing his hobby was gardening, she repeatedly asked her
surgeon’s advice about the growth on her bulbs. Diagnosis: Anxiety Disorder or Phobia
B. ‘Immature’ defenses
Projection-. Following a slight postoperative wound infection, she wrote long, angry letters to the papers blaming the hospital for unsanitary conditions.
Blaming her doctor for not doing a Pap smear earlier, she threatened to institute malpractice proceedings. Diagnosis: Paranoid Personality
Dissociation-. Emerging from anesthesia, she felt no regret but instead enjoyed what she felt was a religious experience. Postoperatively, she told all her
friends that her pain gave herself a sense of joyous communion with sufferers everywhere. She felt an intense inward sense of good fortune that
she had been favored by God to have had her cancer discovered so soon and to have come through surgery so well. Diagnosis: Dissociative Disorder
Fantasy: She asked the nurse not to permit visitors because they made her ‘sad.’ She threw out all her flowers and instead read and reread a copy of
Parent's Magazine and looked repeatedly at The Family of Man. She would go down the corridor to the newborn nursery daydreaming about what
she would call each child if it were hers, and once a floor nurse had to ask her not to whistle lullabies so loudly. Diagnosis: Schizoid Personality
Hypochondriasis : She became worried that the cancer might have spread to her lymph nodes and belabored her visitors with accounts of the tiny lumps
in her groin and neck. When her sister came to visit, the patient angrily accused her of caring so much for her own children that she did not care if her
own sister died of cancer. Diagnosis: Borderline Personality
Passive aggression: When the intern, while inserting an IV, missed her vein, she smiled at him, told him not to worry, and said, “When you’re just a
medical student, it must be hard to get things right.” Unable to sleep, she watched her IV run dry. Later, at 4 a.m., the night nurse had to call the intern
to restart the IV. She cheerfully told him that she had not rung for the nurse because she knew how busy everyone in the hospital was. Diagnosis:
Passive Aggressive Personality
Acting out: Shortly after leaving the hospital, she was unfaithful to her husband with four different men in a month, twice picking up men in cocktail
lounges and once seducing an 1 8-year-old delivery boy. Prior to that time, she had no sexual interest in any man but her husband. Diagnosis: Sociopath
A. ‘Psychotic’ defenses
Distortion: On the third postoperative day, she announced that she was a Christian Scientist and was signing out immediately. Besides, she had to go
home because she and her husband were planning atrip to Bermuda that weekend. (In reality, their total income was $200 a week and their Blue Cross
had lapsed.) She added with a naughty laugh that they needed a vacation to do a little ‘spring planting.’ Diagnosis: Mania
Delusional projection: She felt that the hospital was being run by racists who were trying to sterilize her. She tried to telephone the FBI to report the
hospital for genocide. She refused her pain medication, claiming it was an experimental drug for thought control. Diagnosis: Paranoid Schizophrenia
Psychotic denial: She started ordering the nurses to move her upstairs to the maternity ward. She wandered about the hospital looking for her baby. By
phone, she ordered an expensive bassinet and baby clothes to be delivered to her house. She experienced no postoperative pain. Diagnosis: Mania
woman wins community praise for founding a shelter for
battered women (altruism), but she dismisses her behavior as
'a need to rent my house.' Another woman is imprisoned for
breaking her toddler daughter's arm in a tantrum (acting out),
but she dismisses her behavior as an accident. In fact, social
agency records from 30 years before reveal that both women
had been taken at age two from the care of physically abusive
alcoholic mothers.
Defense Mechanisms 665
The Q-sort method (described in Ego Mechanisms of
Defence) appears the most reliable method for assessing
defenses from a clinical narrative.
Prospective Validation of the Hierarchy
The Study of Adult Development at Harvard consisted of
three cohorts followed for a lifetime: the COLLEGE sample
(Harvard sophomores selected for mental health in 1940),
the CORE CITY sample (socioeconomically deprived, but
nondelinquent, inner city adolescents selected in 1940), and
the TERMAN sample (California 4th grade girls with high IQs
selected for longitudinal study in 1922).
For all three samples, the adaptiveness of each subject's
defenses was assessed along a 9-point scale; point 1 equaled
using only mature defenses, and point 9 equaled using only
immature defenses. Table 3 illustrates, that, analogous to clot-
ting mechanisms, choice of defense mechanisms is relatively
unaffected by parental social class, IQ, and education. How-
ever, Table 4 illustrates that prospectively choice of defense
mechanism predicts successful aging and income.
In order to cut through the unreliability of projective tests
and the subjectivity of pencil and paper tests, Menninger psy-
chologist, Lester Luborsky, devised a behavioral guide (Health-
Sickness Rating Scale or HSRS) to assess psychological func-
tioning on a scale of 0-100. Two of the architects of DSM-III
developed a revision of the HSRS called the Global Assessment
Scale (GAS). A modified version of the GAS was introduced in
DSM-III-R as the Global Assessment of Functioning (GAF or
Axis V). Table 5 illustrates how much psychopathology and/or
mental health is associated with choice of selected defense
mechanisms.
Treatment
Psychiatry needs to understand how best to facilitate the transmuta-
tion of less adaptive defenses into more adaptive defenses. The
modification of immature defenses occupies the attention of high
school teachers, criminal justice personnel, and psychotherapists
everywhere. At present, little evidence-based treatment is available;
nevertheless, there is literature information on change in defenses
over treatment and time, which indicates (mostly noncontrolled
studies) that immature defenses decrease and mature defenses
increase significantly. (Perry et al., 2008)
Perhaps, the best way to alter a person's choice of defensive
style under stress is to make his/her social milieu more predict-
able and supportive. We are all a little schizoid and paranoid
when among strangers whom we fear may treat us harshly. We
are all more adept at altruism, suppression, and playful subli-
mation when among friends and others empathic toward our
pain. The second strategy is to facilitate the intactness of the
central nervous system (e.g., sleep, nutrition, sobriety, and
removal of sources of toxicity and infection).
Toxic brain syndrome is often associated with paranoia;
intoxication leads to fantasy; both grief and anger can lead to
rage turned upon the self. We are all more mature when not
hungry, tired, and lonely.
Finally, if we breach patients' defenses, we must be sure that
we have their permission. If in the course of examination we
Table 3 Correlation of social antecedents with adaptiveness of
defenses
A ntecedent A daptiveness of defenses
College
N = 154 3
Core city
N = 189 s
Terman
H = 40
Years of education
0.13
0.10
0.33*
IQ
0.04
0.14
0.07
Parental social class
0.11
0.00
0.13
*p< 0.05, '"p< 0.001 Spearman correlation coefficient.
"Sample size is reduced. In order to control confounders, men with IQ < 86,
depression, alcohol dependence, and schizophrenia have been excluded.
Table 4 Late life consequences of adaptive defenses at age 20-47
Core city N = 137 a
College N = 154‘
1. Objective evidence
Adaptiveness of
Adaptiveness of
defenses
defenses
Income (Midlife)
0.25**
0.28***
Psychosocial adjustment
0.16
0.34***
(50-65)
Social supports
0.12
0.34***
II. Subjective evidence
Joy in living 6
0.14
0.35***
Spearman correlation coefficient (rho) was the statistic used.
*p < 0.05, ** p < 0.01 , *** p < 0.001 .
"Sample size is reduced because men who died before age 65 are excluded.
‘Measured at age 65 for the core city men and measured at age 75 for the College men.
Table 5 Percent of core city men at different levels of global
mental health using selected defenses
HSRS
Major defense
Impaired
0-65
66-70
Healthy
71-84
85-99
n = 53
n = 36
n = 143
n = 74
Projection (%)
30
17
7
0
Fantasy (%)
19
11
1
0
Hypochondriasis (%)
21
11
1
0
Passive aggression (%)
32
36
15
1
Dissociation (%)
55
36
15
1
Altruism (%)
0
3
6
35
Suppression (%)
2
14
27
59
ask patients to remove their protective covering, we must pro-
tect them with something else. Either we must offer ourselves -
a luxury rarely available to busy clinicians - or we must offer
and facilitate a substitute defense. It is only sometimes that
psychopharmacology is specific enough to do the job.
See also: Coping.
666 Defense Mechanisms
Further Reading
American Psychiatric Association (1994) Diagnostic and Statistical Manual of
Mental Disorders, 4th edn. Washington, DC: American Psychiatric Association.
Brown HN and Vaillant GE (1981) Hypochondriasis. Archives of Internal Medicine
141: 723-726.
Cramer P (1 991 ) The Development of Defence Mechanisms. New York: Springer Verlag.
Freud S (1894/1964) The neuro-psychoses of defence. The Standard Edition of the
Complete Psychological Works of Sigmund Freud, vol. 3, pp. 45-61. London:
Hogarth Press.
Freud A (1937) The Ego and the Mechanisms of Defence. London: Hogarth.
Glueck S and Glueck E (1968) Delinquents and NonDelinquents in Perspective.
Cambridge, MA: Harvard University.
Haan NA (1977) Coping and Defending. San Francisco: Jossey Bass.
Luborsky L (1962) Clinicians Judgments of Mental Health. Archives of General
Psychiatry 7: 407-417.
Perry JA, Beck SM, Constantinides P, and Foley JE (2008) Studying change in
defensive functioning in psychotherapy, using defense mechanism rating scales
four hypotheses, four cases. In: Levy RA and Ablon JS (eds.) Handbook of
Evidence-Based Psychodynamic Psychotherapy, pp. 121-153, Chapter 6.
New York, NY: Humana.
Sartorius N, Jablensky A, and Regier DA (eds.) (1990) Sources and Traditions of
Classification in Psychiatry. Toronto: Hogrefe and Huber.
Skodol AE and Perry JC (1993) Should an axis for defence mechanisms be included in
DSM-IV. Comprehensive Psychiatry 34: 108-119.
Strachey J (ed.) (1926/1964) Inhibitions, Symptoms, and Anxiety, standard edn.,
vol. 20. London: Hogarth Press.
Vaillant GE (1977) Adaptation to Life. Boston: Little Brown.
Vaillant GE (1979) Natural history of male psychological health: Effects of mental
health on physical health. New England Journal of Medicine
301: 1249-1254.
Vaillant GE (1992) Ego Mechanisms of Defence. Washington, DC: American Psychiatric
Press.
Vaillant GE (1993) The Wisdom of Ego. Cambridge, MA: Harvard University.
Vaillant GE (2000) Adaptive mental mechanisms: Their role in a positive psychology.
American Psychologist 55: 89-98.
Delusions
P R Corlett, Yale University, New Haven, CT, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Habit Habitual behavior is rigid, inflexible, and insensitive
to the current value of the ends that it achieves. Delusions
may be habits of thought.
Heuristic A 'rule of thumb' or educated guess; used to come
to an optimal solution as rapidly as possible.
Inference The process of interpolating the next logical step
in an intuited pattern. The conclusion drawn is also called
an inference.
Judgment The evaluation of evidence in the making of a
decision.
Perception The process of understanding sensory
information; distinguished from sensation by the
observation that it is substantially constructive, that is,
what we have previously perceived influences subsequent
processing of inputs.
Prediction error The mismatch between what is expected in
a given situation and what is subsequently experienced.
Prediction errors relate to the very essence of learning; by
minimizing prediction errors, organisms learn to better
predict their environment.
Salience The capacity for stimuli, thoughts, and percepts to
grab attention and drive action.
Uncertainty Related to prediction error - this quantity is
minimized through learning. It drives the allocation of
attention toward explanatory cues. In formal models, it is
related to free energy or entropy. Motivational factors like
need for closure are engaged when uncertainty is high.
Introduction
Delusions occur in the context of numerous pathological states,
ranging from head injuries to schizophrenia, Alzheimer's, and
Parkinson's diseases, to rare autoimmune encephalopathies.
This biological heterogeneity is mirrored in the psychological
and philosophical realms; a menagerie of state- and trait-related
processes have been invoked to explain delusions. Worse still,
delusions have myriad contents, ranging from the mundane
('my partner is cheating on me') to the exotic ('I am the right
foot of God'). Hence, delusional beliefs represent a challenge:
Can they be understood? Are they aberrations of belief? Are they
the rational products of aberrant perception or empty speech
acts? To meet this challenge, perhaps a definition is a good place
to start. As we shall see, this may be easier said than done.
Delusions Defined?
We all know what delusions are, but can we define them?
Clinicians adopt a normative approach to diagnosing delu-
sions, and, since delusions are judged against societal norms,
such an approach allows scientists and laypeople to recognize
delusions without necessarily understanding or being able to
define them.
According to the Diagnostic and Statistical Manual (IV, Test
Revision), delusions are
false beliefs based on incorrect inference about external reality that
persist despite the evidence to the contrary and these beliefs are not
ordinarily accepted by other members of the person's culture or
subculture.
There are numerous problems with this definition: for exam-
ple, delusions may not be beliefs and, even if they are, what
are beliefs? Perhaps, a pragmatic approach might help: what are
beliefs for? One function they might serve is in representing
and organizing information and thus coordinating adaptive
responses, that is, beliefs might be a special kind of memory.
This definition of belief entails a close relationship between
beliefs and actions. To this end, some object to delusions being
beliefs because individuals with delusions do not act consis-
tently with their beliefs; a patient with Cotard delusion believes
that she is dead (note that in describing delusions, one cannot
help but use terms like belief! ), she will nevertheless eat and
move (unlike typical dead people); a patient with Capgras
delusion who is convinced that his/her spouse has been
replaced by an imposter will share the house with the apparent
imposter and will not search for the 'real' spouse. Of course,
some patients act on their delusions with violent and tragic
consequences. Those who favor the doxastic (belief-based)
account of delusions point to the inconsistency with which
even healthy individuals act upon their beliefs and the intrac-
table computational power that would be required to act with
complete consistency on all of our beliefs all of the time.
The consistency argument also extends to fixity. Delusions
are defined as tenacious, fixed in the face of contradictory
evidence. Yet, patients' assent waxes and wanes, so fixity or
strength is not pathognomic of delusions (despite being an
important feature and one that engenders a great deal of dis-
tress for the sufferer, and their family and friends).
Next, delusions do not have to be about external reality, for
example, patients can be deluded about their internal states; an
individual may believe, based on his/her internal sensations,
that he/she has a nuclear reactor inside the stomach. Further-
more, emotional factors may guide delusion formation and
maintenance and, while these may be responses to external
events, they nevertheless involve internal sensations that may
or may not be veridical.
Finally, delusions do not have to be based on incorrect
inference; a patient who is convinced that his/her spouse is
being unfaithful (often called Othello syndrome) may be cor-
rect. However, although the belief may be correct, the key
667
668 Delusions
point is that the evidence on which it is predicted is false.
This is particularly difficult to discern, as patients often cannot
describe the data that support their contentions.
So, is there hope for a scientist who studies delusions?
How can one explore the cognitive and neural mechanisms
of something that cannot be defined? The answer involves
making certain assumptions that may amount to delusions in
themselves! One approach that has proved fruitful in recent
years involves cognitive neuropsychiatry; that is, we attempt
to understand a mental symptom in terms of a dysfunction to
one or many cognitive modules; delusions may be understood
as aberrations of normal belief function.
The remainder of this article is concerned with the variety
of cognitive neuropsychiatric models of delusion, and their
strengths and weaknesses. We conclude by suggesting that a
unifying theory may be possible but that further empirical and
theoretical work is essential to make this a reality.
Perception
One category of models holds that delusions are a normal and
rational response to an abnormal percept or percepts. That is,
the delusion represents a belief that is formed to explain away
an unusual experience. Maher conceives of this perceptual
kernel as a 'surprise' that demands explanation and that the
particular explanation would be generated by all of us, were we
to experience the same surprise. This model has found favor
with cognitive neuroscientists and has garnered support from
functional imaging studies of belief formation and paradigms
that involve the administration of psychotomimetic drugs like
ketamine, amphetamine, and cannabis to healthy volunteers.
These treatments engender alterations in visual and auditory
perception as well as affective responses and, while they also
impair cognition and decision-making, there is at least some
evidence relating the degree of perceptual aberration they
induce to the degree of delusion-like ideation. These studies
suggest that all of us have the capacity to experience delusions
given the correct perturbation; however, these studies have not
demonstrated the primacy of perceptual aberration. Further-
more, top-down cognitive manipulations can also engender
delusion-like ideation, for example, hypnosis. Finally, the
delusion-like ideas that subjects express are not identical with
clinical delusions (although some subjects do lose insight and
report considerable conviction). Perhaps it is best to consider
these manipulations modeling the very early phases of delu-
sion formation, the delusional mood, in which the world feels
changed, strange, and puzzling, pregnant with new meaning,
though these experiences have not yet crystallized into a
delusion.
One influential neurobiological model of delusions explains
this process in terms of the brain systems that ascribe impor-
tance or salience to stimuli, thoughts, and percepts. Salience
represents the capacity for an experience or idea to grab atten-
tion and drive goal-directed action. In a drive for consilience,
theorists sought to relate a candidate neurotransmitter, dopa-
mine (implicated in the deluding process, because when its
activity is blocked in psychotic patients, delusions resolve), to
the phenomenology of having delusions. While it is an admira-
ble effort that garnered some empirical support, the heuristic
explanation of 'too much dopamine, leads to too much salience
leads to delusions' has been supplanted by more mechanistic
explanations that are capable of explaining more of the empiri-
cal data.
Surprising experiences that demand explanations are also
salient, and they lie at the heart of a category of delusion
models that are grounded in formal animal learning theory.
Building upon preclinical data from laboratory animals learn-
ing to predict their environment and respond appropriately,
these models identify the mismatch between expectation and
experience or prediction error as crucial for learning and atten-
tional allocation. These accounts were applied to human
causal belief formation in the 1980s with demonstrations
that human subjects acquiring cause-effect beliefs showed
evidence for the same biases and patterns of responding as
experimental animals confronted with contingencies between
arbitrary stimuli (lights and tones) and the delivery of salient
events (food, water, electric shocks, etc.).
Central to these behavioral effects is prediction error -
subjects learn most when their expectancies are violated, and
without prediction error, no learning occurs. On the other
hand, if there is too much prediction error that is inappropri-
ately timed relative to environmental cues and context, then an
organism may learn inappropriately. Delusions form when
prediction errors are signaled inappropriately, driving atten-
tion toward irrelevant stimuli, forging maladaptive beliefs.
There is empirical support for this model from pharma-
cological model psychoses and from patients with first-episode
psychosis; in both cases, aberrant prediction error signals cor-
related with the severity of delusions.
More recently, this model has been extended in an attempt
to explain the fixity of delusional beliefs. Briefly, if delusions
form as schemas to explain aberrant experiences, then, once
formed, they are deployed to explain future experiences. This
process involves activating the delusion representation and
adventitiously reinforcing it such that, even when the logical
predictions of the delusion are not borne out, the fact that the
delusion was reevoked and ruminated upon dominates over
any new learning that the delusion is untrue. There are preclin-
ical data that suggest that prediction errors also govern this
reactivation and reconsolidation process, and some clinical
data to suggest that the phenomenology of delusions conforms
with a reconsolidation-like mechanism for delusion mainte-
nance. For example, confronting patients with evidence that
contradicts their beliefs induces belief 'elasticity'; they will
engage in quite impressive explanatory feats to encompass or
explain away the contradictory data, often incorporating the
information or the confronting individual into the delusion.
Additionally, as delusions resolve, patients describe a
'double-awareness' in which they endorse the delusion but
also reject it, a duality of belief and disbelief. In preclinical
animal learning theory, dual representations exist following
extinction of prior learning (hence the spontaneous renewal
of fear responses after extinction), as well in animals over-
trained on instrumental contingencies. Here, there are at least
two representations competing for expression in behavior: a
flexible goal-directed representation that is sensitive to the
value of the outcome of an instmmental response and a more
rigid response tendency controlled by contextual cues and
insensitive to outcome value (a habit). It is possible that the
Delusions 669
fixity of delusional beliefs is explicable if we conceptualize
them as instrumental habits that compete with and overwhelm
an impaired cognitive system that is attempting to impose a
veridical view of the world. Neurobiologically, boosting dopa-
mine function in the prefrontal cortex can switch an animal
from habitual responding back to more flexible, goal-directed
behavior. Perhaps delusions may be ameliorated by increasing
the sufferers' capacity for top-down executive control. In the
next section, we discuss such high-level executive cognitive
factors in more detail and how they may contribute to the
psychopathology of delusions.
Cognition
Perceptual models of delusion formation suggest that delusions
represent a rational inferential response to an unexpected per-
cept. While there are empirical data supporting the notion of
peripheral perceptual dysfunctions that conform to the thematic
content of a patient's delusions, and even nondeluded indivi-
duals are suboptimal when it comes to probabilistic inference,
there are theories of delusions that appeal to inferential dis-
ruptions generating these aberrant beliefs and mediating their
maintenance. In this section, we discuss those models.
Reasoning involves the purposeful manipulation of relevant
details from previously acquired stored information. Psychotic
patients report that the logical sequence of their ideas was
replaced by sequences of merely associated thoughts, and the
reasoning process therefore became increasingly concrete. Psy-
chotic reasoning involves 'a lack of the inhibition of peripheral
ideas necessary for effective abstraction.' Some patients report
that almost every thought immediately and automatically sug-
gested an enormous number of associated ideas. Some enjoy this
enhanced ability to associate, feeling endowed with enhanced
mental reactions and a more sophisticated problem-solving abil-
ity; indeed, delusions of grandeur often develop from the appar-
ent insights that accompany this hyperactivity of association.
Chapman reported that most patients who experienced the
changes outlined previously reported that, for a time, every-
thing around them looked fascinating and that objects stood
out vividly against the background. They regarded everything
as significant and found themselves attending to many irrele-
vant aspects of the environment. There was a general tendency
for their interest to be turned to ruminating about psychology,
philosophy, art, and literature. Those patients who experienced
perceptual and attentional aberrations also experienced hyper-
activity of associations. Matussek's patient (R) illustrates this
concept clearly: "Out of these perceptions came the absolute
awareness that my ability to see connections had been multi-
plied many times over. " These experiences are appraised posi-
tively at first; however, as they persist, patients become
increasingly anxious. Attempting to account for their bizarre
experiences, they arrive at irrational conclusions. Paranoid
ideas and delusions of various kinds develop in response to
the various alterations of sensory and attentional experience.
Heuristics
Human judges are not always rational. People tend to jump to
conclusions, employing shortcuts and heuristics, some of
which are relevant to the present discussion. For example,
people tend to assume that the features of a causal event should
resemble the features of its outcome (the representativeness
heuristic). Unpleasant effects should have unpleasant causes.
Furthermore, peoples' causal judgments tend to be greatly
influenced by their a priori theories about causation: If some-
one has the idea that many unpleasant events in the outside
world reflect the activities of an international terrorist con-
spiracy, those same terrorists may be held responsible for
unpleasant internal events as well. Thus, in seeking an expla-
nation for their distressing anomalous experiences, the psy-
chotic individual experiencing them may be especially likely
to focus on members of out-groups, individuals, and classes of
people who are generally feared in his or her culture. It seems
possible to appeal to an associative mechanism to explain this
heuristic, a particular personal bias may be mediated by asso-
ciations; the increased salience of a particular out-group may
increase the propensity to form associations between that
group and events in the environment.
The simulation heuristic posits that the basis for judgment
is the ease with which a plausible scenario can be constructed
mentally. Judgments of causality are affected by the ease with
which the person can imagine a path from a presumed cause to
a known effect. When unpredicted events occur, the simulation
process traces causal links back to prior causes. Consider a
psychotic patient searching the environment for a likely cause
for anomalous experiences. Salient objects and events - a honk
or a wave from a passing driver, perhaps a member of a
minority group standing on a street corner - will inevitably
draw attention, and be given special weight as a likely cause of
their troublesome internal events. If there is nothing perceptu-
ally salient, events may be retrieved from memory - a curse
uttered in anger by a coworker or a sin unconfessed to a priest.
If no suitable cause is generated through perception or mem-
ory, the simulation process may be invoked. The person may
imagine possible causes and grasp the first one that comes to
mind as the most likely explanation.
It is plausible that the simulation heuristic may be mediated
by associative mechanisms, namely, the retrieval of associative
chains such that the individual can mentally trace the asso-
ciations from outcome to cause. A computationally intensive
Bayesian tree-search mechanism located in the prefrontal
cortex outlined by Daw and colleagues may underpin this
heuristic. This mechanism may be invoked to account for
the apparent relatedness of stimuli and events or the aber-
rant attentional salience of previously irrelevant background
stimuli.
While the heuristics described are involved in the initial
generation of a causal explanation, anchoring and adjustment
might be involved in the maintenance of delusional beliefs.
Many judgments begin as hypotheses - tentative conclusions
that can be revised on the basis of newly acquired evidence.
However, it has long been appreciated that final judgments are
inordinately influenced by first impressions: the initial judg-
ment serves as an anchor for the final one, and there is very
little subsequent adjustment. The anchoring and adjustment
heuristic reflects a general tendency to rely on initial or partial
judgments, giving too little weight to newly acquired informa-
tion. By virtue of its use, judgments of causality tend not to
accommodate new information that should instigate revision.
670 Delusions
Instead, knowledge gained subsequent to the initial judgment
may be distorted so as to fit the original causal theory; subjects
adopt suboptimal verificationist strategies, seeking and paying
special attention to information that is consistent with their
hypothesis. As many researchers will attest, when confronted
with evidence that counters a cherished belief, individuals
often react by challenging the evidence. Once an explanation
for odd perceptual and attentional phenomena is arrived at,
the patient experiences relief from anxiety. The experience of
insight relief diminishes the person's subsequent motivation to
question his or her original conclusions and increases resis-
tance to contrary information. This theme is represented in
Miller's associative learning-based account of psychosis. He
argues that arriving at a causal explanation that accounts for
aberrant experiences is so rewarding/relieving that it is accom-
panied by a surge of dopamine. Dopamine also impacts upon
the consolidation of memories and as such, a delusional con-
clusion may be 'stamped-in' to long-term memory by dopa-
mine, rendering it relatively impervious to disconfirmatory
evidence.
Jumping to Delusions?
A prominent cognitive theory of delusion formation involves
the 'jumping to conclusions bias.' Deluded individuals tend to
make decisions hastily, on the basis of little evidence. Empiri-
cal evidence for this account comes from investigations of
patient performance on probabilistic reasoning tasks; typically,
participants are presented with two jars holding colored beads
in different proportions. The jars are removed from view and
subjects are presented with beads, drawn one at a time, from a
jar and patients are asked to predict from which jar the beads
are coming. Deluded individuals tend to make a decision after
only one bead. However, rather than maintaining their hasty
conclusion, patients also tend to overadjust their judgments in
the light of potentially disconfirmatory evidence.
A liberal acceptance bias might account for apparent jump-
ing to conclusions. When only two mutually exclusive options
are available (like the beads task), patients rapidly accept that
the beads are coming from a particular jar, but they do not
decide that they are to the exclusion of other possibilities. This
account allows for patients' overadjustment following contra-
dictory evidence, since although they have strongly accepted
one conclusion (the beads are from one jar), they do not
exclude the alternative conclusion (that the beads are coming
from the other jar). When given more than two alternatives
(e.g., in a thematic apperception task, where participants are
shown pictures and asked to rate the plausibility of particular
interpretations), psychotic patients entertain a broader range
of possible interpretations (rating multiple alternatives as
excellent or good interpretations of a particular scenario),
whereas healthy participants are more cautious and effectively
narrow down the set of possible alternatives. Hence, psychotic
individuals might entertain a broader range of possible ex-
planations for their experiences and an absurd idea may
prevail. This broadening of plausible explanations may be a
manifestation of Miller's inappropriate relatedness of entities,
but at a higher, representational level, leading to the entertain-
ment of implausible or absurd accounts for a particular set of
circumstances.
However, jumping to conclusion biases are not specific to
deluded patients and they do not account for the fixity of
delusional beliefs (since patients overadjust following discon-
firmatory evidence). It is possible that patients' performance on
probabilistic reasoning tasks may reflect the bizarre and unlikely
nature of delusional conclusions rather than their fixity.
Self-Serving Bias
Since anomalous perceptual and attentional experiences may
be unpleasant, it is important to consider the biases that distort
causal judgments about negatively valenced events. For exam-
ple, when humans make causal attributions, they tend to suffer
from a benefactance bias, such that they internalize the cause
of positive events and externally attribute negatively valenced
events. Hence, psychotic individuals seeking an explanation
for their unpleasant anomalous experiences will most often
look to the environment outside them.
In an fMRI study of the self-serving bias in healthy indivi-
duals, subjects silently read sentences describing positively and
negatively valenced social events, then imagined the event
happening to them, and finally decided the cause of the
event, whether internal (Was it something about you?) or
external (Was it something about your friend?/Was it some-
thing about the situation or circumstances?). Self-serving
biased attributions (internal attribution of positive and exter-
nal attribution of negative events) were associated with striatal
activation, previously implicated in the motivational control of
behavior as well as in the mediation of hallucinations and
delusions.
The cognitive biases outlined are not specific to individuals
with delusions, being present in individuals with schizophre-
nia but not related specifically to delusion severity (like the
jumping to conclusions bias) or being present more generally
in the population and being engaged by the need to explain an
odd percept or experience in an individual developing delu-
sion (like self-serving and simulation). What then is the evi-
dence for an inferential deficit in individuals with delusions
and how might that deficit relate specifically to their belief
pathology?
Belief Evaluation Deficits
Perhaps the strongest evidence in favor of an inferential deficit
related specifically to delusions comes from the elegant work
emanating from the Macquarie Center for Cognitive Science,
specifically, the Belief Formation program, under the direction
of Max Coltheart. This work focuses on neuropsychological
delusions, delusions that arise as a result of some sort of head
injury or neurological insult, although attempts have been
made to extend it to the delusions that occur in the context of
schizophrenia. They advocate a two-factor approach in which
perceptual and reasoning abnormalities combine. The latter
derive from observations that neurological patients with delu-
sions often have two sites of damage: ( 1 ) a lesion in a percep-
tual region (such as the fusiform face area) and (2) an
additional lesion in 'belief evaluation' regions, possibly in the
right frontal cortex. The first damage engenders odd percepts
and the second generates bizarre explanations. The second
factor is necessary because there are individuals with damage
Delusions 671
to factor 1, engendering aberrant percepts, who nevertheless
remain without delusions; for example, subjects who have
suffered haptic deafferentiation and therefore do not perceive
sensory feedback from the actions they perform. While the
haptically deafferented subjects have problems performing
actions, they do not feel that their actions are under the control
of another agent. Two-factor theorists equate the inappropriate
prediction error signals reported in the dorsolateral prefrontal
cortex with the aberrant belief evaluation process or factor 2.
However, it is important to emphasize the potential con-
sequences of prefrontal cortical damage alone (their factor 2)
as well as peripheral perceptual dysfunction (their factor 1);
there are patients who suffer from delusion-like spontaneous
confabulations following damage to the ventromedial and
lateral prefrontal cortex and at least one patient in whom
peripheral sensations are perturbed (following damage to the
brachial plexus) who has somatic delusions in the absence of
any apparent structural damage, and by extension, any deficit
in factor 2.
Top-Down Theories
In addition to the probabilistic reasoning dysfunctions out-
lined earlier (about which the empirical data are equivocal),
there are theories of delusion that invoke top-down factors
alone. According to such theories, deluded individuals confuse
belief with imagination. Individuals susceptible to delusions
do tend to confuse the items they perceived with the events
they imagined. Such theories are also popular because they
allow for the inconsistency observed in patients between their
beliefs and actions; if a delusion is aberrant imagination, then
the deluded individual need not act upon it. However, these
theories do not allow for the role of perceptual dysfunctions in
the formation of some delusions outlined previously.
Is there any way to reconcile these groups of explanations?
Perhaps, a hybrid model in which bottom-up experiential- and
top-down expectancy-based processes can both contribute to
delusion formation and maintenance. Such a model involves
taking a Bayesian approach to perception, cognition, and brain
function. In the next section, we discuss such models.
Bayesian Models
Thomas Bayes was an English mathematician and minister
whose theorem on probabilities was published posthumously.
The theorem has had an incredible impact on science, in
particular, cognitive and neural science, in recent years. In the
paper, Bayes' stance on probability is as follows:
The probability of any event is the ratio between the value at which
an expectation depending on the happening of the event ought to be
computed, and the value of the thing expected upon its happening
Why should a cognitive scientist care about such an arcane
mathematical theory? Bayes theorem might help us, as scien-
tists, to reason about new data and evaluate our a priori
hypotheses: according to Bayes rule, the probability that a
hypothesis is true given the observed data (the posterior prob-
ability) is proportional to the likelihood (the probability of
those data given that the hypothesis is true) multiplied by the
prior probability (the probability that the hypothesis was cor-
rect before the data were seen). For simplicity, the 'prior prob-
ability' is often abbreviated as the prior.
More formally, we can express the relationship between a
hypothesis and some new data as
(H\ = P(D/H)P(H)
V dJ P(D )
where H is the hypothesis and D is the data. P(H ) is the prior
probability of H : the probability that H is correct before the
data D was seen. P(D/H) is the probability of seeing the data D
given that the hypothesis H is true. P(D/H) is the likelihood.
P(D) is the probability of observing the data D and P(H/D) is
the posterior probability.
Hence, surprising data might be discredited in the face of
overwhelming prior experience, or it might demand explana-
tion and an updating of our priors. Herman von Helmholtz
was a physician and physicist who constructed prescient the-
ories of visual perception that appear steeped in Bayes theo-
rem; Helmholtz postulated that perception was a process of
unconscious inference; that the brain entertains hypotheses
about possible inputs and compares them with incoming sen-
sations. Through experience, particular candidate hypotheses
can come to be favored over others, based on their minimizing
the uncertainty or free energy associated with a given situation.
A compelling demonstration of this phenomenon involves
observing a rotating hollow face mask: the binocular depth
inversion illusion. Since our overwhelming experience of faces
is that they are convex, the backside of the mask is perceived as
convex when it rotates into view, even though the sensory
information incident on the retina indicates a concave object.
An important distinction, within Bayesian models, is
between prediction errors per se and the precision or uncer-
tainty about those errors. These processes may have separable
neurobiological bases and cognitively, these processes may
mediate inference and learning, respectively. While inference
and learning are intimately connected, they are dissociable to
some extent. Inferences involve short-term decisions, like
unconscious perceptual inductions, where one perceptual
hypothesis is favored over another, for example, the Necker
Cube. On the other hand, learning represents the set of prior
expectancies brought to bear on current processing; the set of
learned predictions that drives healthy individuals to perceive a
hollow mask as a convex face.
While Bayesian models are often considered rational and
optimal, they have nevertheless been deployed to explain irra-
tional processes such as the spread of panic and rumor within a
crowd (which occurs rapidly in salient situations with few
explanatory priors). Delusions may involve a disruption to the
normal Bayesian predictive mechanisms of the brain such that
predictable and irrelevant events mismatch with expectancies
and their salience demands new learning and explanation; a
delusion represents an explanatory mechanism, an attempt to
impose order on a disordered perceptual and cognitive world.
Delusions and the Self
Delusions often involve dismptions to the representation of
self. Wegner and others hypothesize that a sense of self-agency
and free will may be learned through experience; having an
672 Delusions
intention to act very frequently precedes the action itself and
this contiguity binds intentions with actions through associa-
tive learning. Subjects judge the time between performing an
action and producing an outcome as shorter when the action
was intentional, a process of action-outcome binding. Patients
with delusions show a hyperbinding effect, an exaggerated
binding between their actions and the outcomes they produce,
consistent with a disturbed agency account of paranoia. This
process of learned intentionality has been modeled using
Bayesian mechanisms; in essence, the task of inferring causal
agency involves conditioning the evidence (whether the out-
come occurred?) over the priors (was there an intention to act
and would the outcome be consistent with the outcome per-
formed?). Inappropriate engagement of this inference mecha-
nism could account for excessive and inappropriate agency
underpinning, for example, beliefs in telekinesis or telepathy,
but what about delusions of passivity or external control?
Producing movements over which we feel a sense of agency
also involves predictive learning and prediction error; an inter-
nal 'forward' predictive model of motor commands or efference
copy is used to predict the sensory consequences of movements,
and these predictions are compared with the actual sensory
feedback during movement execution. Impairment in such a
predictive system would result in a failure to attenuate the
sensory consequences of self-produced actions, making them
appear indistinguishable from externally generated sensations
and engendering the inference that one's own movements were
externally caused. This theory provides an elegant explanation
for why we cannot tickle ourselves, since we cancel the predicted
sensory consequences of the action. However, patients experi-
encing passivity phenomena rate self-generated stimulation
as ticklish.
Delusions and Others
Observing, imagining, or in any way representing an action
excites the motor program used to execute the same action.
Mirror neurons discharge not only during action execution but
also during action observation. Implicit in their description is
the idea that information is passed by forward connections
from low-level representations of the movement kinematics
to high-level representations of the intentions subtending the
action. Formally, this is a recognition model that operates by
the inversion of a generative model. A generative model will
produce an estimate of the visual consequences of an executed
action given the causes or goals of that action. By inverting the
model, it is possible to infer the cause or goal of an action,
given the visual input.
Again, bottom-up or top-down biases in this inference pro-
cess would lead to gross misrepresentations of other's inten-
tions. Those biases may arise due to aberrant prediction error
signals. Such signals would drive learning of maladaptive social
expectations. These would manifest phenomenologically as
intense feelings of social uncertainty and ultimately, paranoia.
Stress, Need for Closure, and Self-Deception
The impact of our desires and preferences upon our ordinary,
everyday beliefs is well documented. Motivational approaches
view delusions as extreme instances of self-deception, as
defensive efforts to relieve pain and distress. For example, a
motivational explanation of the Fregoli delusion, the belief
that strangers one encounters are deceased family members
in disguise, would point to the Freudian psychodynamic pro-
cess of wish fulfillment, gratifying their wish that their dead
relatives are still present. Of course, not all Fregoli sufferers
believe that they are encountering disguised and deceased
relatives so this is not a general explanation; however, it high-
lights the potential contributions that motivational processes
might make to delusion formation.
Motivational models are perhaps more appropriate in the
domain of paranoid and persecutory beliefs. Here, persecutory
beliefs represent a defense against low self-esteem, hence, those
with persecutory delusions should have high overt self-esteem
and relatively low unconscious, covert self-esteem. Empirical
data from an Implicit Association reaction time task suggest
that this is the case.
One constmct that may provide consilience between Moti-
vational and Bayesian models of delusions is the need for
closure - a motivational construct, associated with a preference
for certainty and predictability. Patients with persecutory delu-
sions have a higher need for closure. To ground this psycho-
dynamic process in Bayesian terms, the need for closure may
represent a drive to minimize prediction error; in this case, a
mismatch between desired and actual self-representation.
Encapsulation, Scaffolding, and Delusion Contents
Early life experiences significantly shape the adult's under-
standing of and reactions to the world. Via the process of
scaffolding, early sensorimotor experiences serve as a founda-
tion for the later development of more abstract concepts. Scaf-
folding refers to the process through which humans readily
integrate incoming information with extant knowledge struc-
tures - a passive natural process through which new concepts
are formed, especially in early childhood. Features of abstract
or less-understood concepts are mapped onto existing and
well-understood concepts such that the structure of the devel-
opmentally earlier, primary concept is retained in the newly
constructed concept. This structure imbues the newer concept
with meaning. This interchange mirrors the updating of priors
by new experience in Bayesian learning. Indeed, category for-
mation and the association of categories may proceed by mini-
mizing prediction error.
Delusion proneness may manifest in development as a
different set of learned priors predicated on aberrant sensori-
motor interactions with the world. These priors may underpin
an alternate set of what Wittgenstein called bedrock beliefs in
delusion-prone individuals. Bedrock beliefs (or priors) are a
framework of foundational and indubitable propositional cer-
tainties ('I am Phil Corlett' ; 'The world is very old'; 'My parents
are human') which globally constrain other empirical beliefs
and provide the terms for their rational interrogation.
The alteration of particular bedrock beliefs in isolation might
allow for the encapsulation of delusions. That is, although
delusions permeate many domains of cognition, deluded indi-
viduals tend not to be deluded about all information they come
across. Rather they have delusions but retain normal beliefs too
and can recognize the absurdity of other's delusions. Previously,
we have speculated that homeobox genes that pattern cortical
Delusions 673
development could predispose individuals toward particular
delusional content and not others. Psychological scaffolding is
a potential cognitive mechanism for this encapsulation process.
This speculation has much in common with Gerald Edelman's
neural Darwinism and the idea that particular representational
categories are sculpted through interaction with the world.
Under the influence of aberrant prediction error, present during
critical developmental periods, individuals would form aberrant
bedrock beliefs or priors on which subsequent delusions are
scaffolded.
Conclusion
In considering delusions, we had to make a number of simplify-
ing assumptions. Taking a reductionist cognitive neuroscientific
approach, we highlighted the importance of dysfunctions in
predictive learning for delusion formation and maintenance.
These same processes may account for the perceptual, affective,
and social disruptions that attend delusions. In deluded indivi-
duals, the ability to use learned information to constrain current
experience is impaired, resulting in aberrations of sensory and
affective perception as well as cognition. Delusions arise as an
explanation for these odd happenings and they engage new
learning. They bring such relief that they are stamped into
memory and become a new explanatory scheme for the sufferer,
that is, delusions are elastic; they encompass new experiences
and maintain a certain consistency of the world for the patient.
In terms of the Bayesian model we outlined, delusions become
the sufferer's new priors, and they are used to predict and
explain future experiences. We believe that the same prediction
error-driven learning mechanisms can account for the fixity of
delusional beliefs, since, now, when subsequent physiological
noise elicits a reactivation of the delusion, it is reinforced and
reconsolidated more strongly.
However, Brendan Maher, Emeritus Professor of the Psy-
chology of Personality at Harvard, astutely aligned delusions
with scientific theories, suggesting that scientists, like indivi-
duals with delusions, were extremely resistant to giving up
their preferred theories even in the face of damningly negative
evidence. While the account outlined is a preferred (some
might say cherished!) explanation of delusions, we hope to
engender discussion, debate, and investigation. As Maher says
of science and psychosis: "Puzzles demand an explanation; the
search for an explanation begins and continues until one has
been devised." We hope this article provides some closure on
the puzzle, but it is clear that we have much more explanatory
work ahead of us.
See also : Alzheimer’s Disease; Behavioral Pharmacology; Classical
Conditioning; Free Will; Human Mating; Memory, Neural Substrates;
The Mirror Mechanism; Motivation; Operant Conditioning; Reasoning;
Schizophrenia; Surprise; Tickling.
Further Reading
Coltheart M, Langdon R, and McKay R (2010) Delusional belief. Annual Review of
Psychology 62: 271-298.
Corlett PR, Krystal JH, Taylor JR, and Fletcher PC (2009) Why do delusions persist?
Frontiers in Human Neuroscience 3: 12.
Fletcher PC and Frith CD (2009) Perceiving is believing: A Bayesian approach to
explaining the positive symptoms of schizophrenia. Nature Reviews Neuroscience
10: 48-58.
Frith CD, Blakemore S, and Wolpert DM (2000) Explaining the symptoms of
schizophrenia: Abnormalities in the awareness of action. Brain Research Brain
Research Reviews 31: 357-363.
Gray JA, Feldon J, Rawlins JNP, Hemsley D, and Smith AD (1991) The
Neuropsychology of Schizophrenia. Behavioral and Brain Sciences 14: 1-84.
Helmholtz H von (1878/1971) The facts of perception. In: Kahl R (ed.) Selected Writings
of Herman von Helmholtz. Connecticut: Weslyan University Press.
Kahneman D, Slovic P, and Tversky A (1982) Judgment Under Uncertainty: Heuristics
and Biases. New York: Cambridge University Press.
Kapur S (2003) Psychosis as a state of aberrant salience: A framework linking biology,
phenomenology, and pharmacology in schizophrenia. American Journal of
Psychiatry 160: 13-23.
Maher BA (1988) Delusions as Normal Theories. New York: Wiley.
Matussek P (1954) Studies in delusional perception. In: Cutting JC and Shepherd M
(eds.) The Clinical Roots of the Schizophrenia Concept: Translations of Seminal
European Contributions on Schizophrenia, pp. 89-103. Cambridge: Cambridge
University Press.
Miller R (1976) Schizophrenic psychology, associative learning and the role of forebrain
dopamine. Medical Hypotheses 2: 203-21 1 .
Sass LA (1994) Paradoxes of Delusion: Wittgenstein, Schreber, and the Schizophrenic
Mind. Ithaca, NY: Cornell University Press.
Schneider K (1959) Clinical Psychopathology. New York: Grune & Stratton.
Schultz W and Dickinson A (2000) Neuronal coding of prediction errors. Annual Review
of Neuroscience 23: 473-500.
Zimbardo PG, Andersen SM, and Kabat LG (1981) Induced hearing deficit generates
experimental paranoia. Science 212: 1529-1531.
Dependent Personality (Disorder)
T A Morgan, University of Iowa, Iowa City, IA, USA
LA Clark, University of Notre Dame, Notre Dame, IN, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Agreeableness Personality trait defined by interest in
maintaining social harmony. One of five general trait
domains in the five-factor model of personality, and
theoretically related to dependency. Low levels generally
characterize individuals with personality disorder.
Autonomy Personality trait defined by low interpersonal
orientation (separation and individuation from others) and
high self-governance (e.g., making own decisions without
others' advice). Some researchers hypothesize autonomy
and dependency to be opposite ends of the same dimension.
Dependency (trait) Personality trait defined by the central
motivation to locate and maintain affiliative relationships
with specific individuals. Multiple models of dependency
exist, but to varying degrees all emphasize lack of
self-confidence coupled with overreliance on others.
Some researchers hypothesize autonomy and dependency to
be opposite ends of the same dimension. Trait dependency
overlaps significantly with dependent personality disorder.
Dependent personality disorder (DPD) Personality
disorder is defined in the DSM and ICD as pervasive patterns
of inner experiences and behavior that are maladaptive,
differ from the norm, and cause significant personal or
interpersonal distress. In the DSM, Dependent,
Obsessive-compulsive, and Avoidant PDs comprise the
'Anxious' cluster of personality pathology. Like trait
dependency, DPD is characterized by a view of the self as
incompetent/incapable, coupled with neediness and
overreliance on others.
Diagnostic and Statistical Manual of Mental Disorders
(DSM) Research-based manual published by the American
Psychiatric Association (APA) that provides definition,
description, and standardized criterion sets for the diagnosis
of mental disorders. The current edition is the DSM-IV; 10
personality disorders are detailed on 'Axis II' of this manual.
Modifications for the next edition (DSM-5) are currently
under way.
Factor A component dimension of a larger structural model
that is statistically distinct. A factor is defined by an
interrelated set of variables such as items on a trait
dimensional measure, or diagnostic criteria such as those
from dependent personality disorder.
Factor analysis A set of statistical procedures used to
identify the component dimensional structure of sets of
putatively interrelated variables (exploratory factor analysis)
and/or to test the replicability of such a structure
(confirmatory factor analysis).
Five-factor model (FFM) A dominant model of normal
personality consisting of five general trait domains (factors):
Neuroticism, Extraversion, Openness (to experience),
Agreeableness, and Conscientiousness. The five factors have
been shown to be relatively stable across time, culture, and
gender, and nonhuman organisms, and provide a broad
basis for conceptualizing normal and abnormal personality.
International Statistical Classification of Diseases and
Related Flealth Problems (ICD) A manual published by
the World Health Organization (WHO) that provides
definition, description, and standardized codes
for diagnosing all medical conditions, including
mental disorders. The current edition is the ICD- 10.
Personality disorders in the ICD roughly correspond to
those found in the DSM. Modifications for the next
edition (ICD-1 1) are currently under way.
Neuroticism The tendency to experience negative emotions
such as fear, anxiety, anger, guilt, sadness, and irritability.
One of five, general trait domains in the five-factor model,
and moderately related to dependency in most research.
High levels generally characterize personality disorders.
Personality disorder (PD) A diagnosis of a mental disorder
currently defined as a pervasive pattern of maladaptive inner
experiences (e.g., thoughts, feelings, motivations) and
interpersonal behaviors that deviates from the norm
(see 'DSM' and 'ICD- 10').
Personality trait A defined set of interrelated thoughts,
feelings, and behaviors that ( 1 ) describe an important
quality of individual differences (i.e., can be used to
differentiate one organism from another); (2) are stable over
time; (3) are consistent across multiple environments; and
(4) can be meaningfully linked to functional outcomes and/
or behavior.
Dependent personality is defined diagnostically in the Diag-
nostic and Statistical Manual, 4th Edition (DSM-IV), as
dependent personality disorder (DPD): "A pervasive and
excessive need to be taken care of that leads to submissive
and clinging behavior and fears of separation" (American
Psychiatric Association, APA). Specific criteria include the
following:
1. Difficulty making everyday decisions without excessive
reassurance;
2. Dependence on others to take responsibility for important
parts of one's life;
3. Difficulty expressing disagreement with others due to fear
of losing approval or support;
4. Difficulty beginning and completing tasks/projects due to
lack of self-confidence;
5. Excessive seeking of support from others;
6. Discomfort when alone due to fear of not being able to care
for oneself;
674
Dependent Personality (Disorder) 675
7. Quick seeking of a replacement source of care and support
when one close relationship ends; and
8. Preoccupation with fear of being left alone and inability
to care for oneself.
Per the DSM-IV, an individual may be diagnosed with DPD
if five of these eight criteria are met. DSM diagnostic criteria are
similar to those defined by the World Health Organization in
the International Statistical Classification of Diseases and
Related Health Problems (ICD-10), which is used internation-
ally to diagnose medical conditions. The ICD-10 requires only
three or more of the following criteria for diagnosis:
1. Dependence on others to make one's important life
decisions;
2. Subordination of one's needs to those of others on whom
one is dependent (i.e., compliance);
3. Inability to make demands of others on whom one is
dependent;
4. Feeling of discomfort when alone due to fears of not being
able to care for oneself;
5. Preoccupation with fears of abandonment; and
6. Limited capacity for decision making without excessive
advice/reassurance.
Both manuals also require that individuals meet general
descriptions of personality disorder, which emphasize evi-
dence of significant psychological distress, maladaptive behav-
ior, relative stability of the disorder over time, and broad
deficits affecting multiple domains.
A review of these criteria suggests the following basic com-
ponents of DPD common to both diagnostic systems: First,
individuals with DPD present with a self-image of inadequacy,
helplessness, and inability to cope, and they view others as
much more capable, able to cope, and dominant. Juxtaposed,
these views of self and others lead to overvaluing significant
relationships and prioritizing the needs of others to maintain
the dependent relationships. Thus, DPD necessarily reflects a
combination of personal and interpersonal deficits. Theoreti-
cally, this may lead to any number of behaviors - even seem-
ingly 'independent' behaviors such as decision making - in
service of maintaining affiliative relationships. However, the
exact behaviors related to dependency have been somewhat
less studied, and it is unclear how frequently behavior would
deviate from the clinical pictures described in the DSM-IV and
ICD-10.
In various forms, dependent personality has been included
in all modern formal diagnostic systems. Historically, 'depen-
dent personality' is linked to early Freudian theory, specifically
to frustration in the oral stage during which the conflict of
dependence-independence was expected to be resolved. Since
that time, all versions of the DSM and ICD have referenced
dependency and dependent PD (in the DSM): first as a subtype
of passive-aggressive PD, later as a part of asthenic PD, and,
finally, as a separate diagnostic category. Material from the
DSM-5 website - which provides the most up-to-date propo-
sals for changes to its psychodiagnostic models - also proposes
the inclusion of dependency, although as component traits
(submissiveness, anxiousness, separation insecurity) that may
characterize personality disorder, rather than as a separate
diagnostic category. The proposed change is representative of
a general shift in diagnostic methods from a categorical to a
dimensional model for personality disorder, and is not specific
to DPD.
Antecedent Validators for Dependent Personality
Disorder
Very little research has been done establishing any clear ante-
cedent to DPD per se, and data for biological or genetic
antecedents are particularly sparse. Some evidence has sug-
gested that dependent traits are consistent within families,
but these results emphasized patterns of interactions and beha-
viors and did not identify DPD or define dependency specifi-
cally. Notably, dependency within families overlapped with
avoidant PD traits, with which it shares both low self-esteem
and need for social contact. Some researchers suggest that -
despite this overlap - the fundamental distinction between
avoidant and dependent PD is that, although both desire
close relationships, those with avoidant PD nevertheless with-
draw from contact. However, among the relatives of the DPD
patients studied, withdrawal from contact was found in virtu-
ally the same frequency as the dependent characteristic of
passively allowing others to assume responsibility for major
life areas. On the other hand, the least frequently observed of
all the avoidant and dependent criteria was the unwillingness
to form relationships unless such relationships guaranteed
acceptance, providing some measure of support for this
avoidant-dependent distinction.
Nevertheless, in the aforementioned research, immediate
family members of DPD-only patients were more likely to
report high dependency as compared to control and 'near
neighbor' samples. These data suggest that DPD patients may
have some family-history liabilities. However, much more
research is necessary to establish the aggregation of DPD/
dependency within families. We did not find any twin or
adoption studies examining the heritability of DPD, nor any
examining biological or genetic material directly.
Similarly, little research is available on demographic or
cultural factors that influence or characterize DPD, except for
gender, with self-reported DPD reportedly more common in
women than in men. Although the DSM-IV also makes this
distinction, some evidence suggests that gender differences
may be due to effects of gender stereotyping and/or sex-role
identification rather than true differences. For instance, self-
report and other rating scales are more likely to yield gender
differences than direct behavioral observations of dependent
behaviors, perhaps because it is more socially acceptable for
women than men to self-report dependency. This hypothesis
has yet to be fully explicated. Similarly, when gender is not
specified in written patient descriptions that include depen-
dency traits and/or dependency-relevant behaviors, diagnosti-
cians are reportedly more likely to believe the patient is female.
This suggests there may be some bias in attention and attribu-
tion of patients' dependent behavior by clinicians.
Moreover, some researchers have not found evidence for
gender bias in DPD using patient samples. Such studies also
report on age, and find that as compared to other personality
disorders DPD is more likely to occur in older patients (>40
years). It is unclear to what extent this is due to naturally
676 Dependent Personality (Disorder)
occurring reliance on others over time (e.g., strengthening fam-
ily relations and decreased physical independence). Although
numerous articles discuss the importance of culture in under-
standing and interpreting dependent behavior, especially in the
psychological anthropology literature, this research is largely
qualitative or theoretic-analytic. Very little formal research cur-
rently exists that examines pathological dependency, as it is
conceptualized in DPD.
Much of the research in the epigenesis of DPD has focused
on environmental risk factors. For example, research shows that
male patients diagnosed with (vs. without) DPD are more likely
to report poor home functioning and lower socioeconomic
status (SES). The latter (low SES) has been replicated multiple
times, and may be mediated by differing parenting styles
reported in such families. Overprotective and authoritarian par-
enting styles may be particular risk factors for later development
of DPD. For example, retrospective studies of family environ-
ments of females with histrionic and dependent PDs (relative to
other psychopathologies and normal controls) have been con-
ducted. These studies show that individuals with DPD reported
high-control and low-independence environments compared to
both groups. Another study showed similarly that high control
and low expressiveness uniquely characterized the early family
environment of females with DPD.
Overall, it is clear that little research exists establishing
antecedents to DPD. What research there is often uses correla-
tional models, which are insufficient to establish the direction
of causality. Although some environmental risk factors have
been implicated - most notably authoritarian parenting and
control - it remains unclear to what extent such factors interact
with or are the result of actions perpetrated by the individual,
or are based on biased retrospective recall of individuals
with DPD.
Predictive Validators and Treatment for Dependent
Personality Disorder
There is no specific or established recommended treatment
protocol for DPD. It is widely accepted that clients with per-
sonality disorder generally present with unique challenges to
therapy, including increased distress and severity of symptoms.
With respect to DPD, for example, patients with depression
and comorbid DPD are more likely to report suicidal thoughts
and attempts than are patients with depression alone. Patients
with DPD also report significant functional impairment in
their day-to-day lives. Such impairments include difficulty
with (or lack of) decision making, task completion, and emo-
tion regulation, and dependency often interferes with the abil-
ity to hold a job, complete education, or maintain healthy
relationships with significant others.
Research on treatment generally falls into two categories:
research protocols/reviews and individual-case or small-
sample studies. Both sources suggest that DPD may be treated
using cognitive-behavioral, interpersonal, or psychodynamic
models of therapy. However, little research has been published
using sufficient sample sizes to comment conclusively about
specific treatments. In a recent review of treatment for cluster of
personality disorders that includes DPD, results were based
on only 15 studies of treatment effects for DPD with these
therapies. Results from this study showed little difference in
treatment outcomes for specific PDs: all benefited the most
from therapy that included components of social skills train-
ing, and few showed complete remission of symptoms. It has
also been suggested that patient dependency may have an
initial positive effect on therapeutic outcome (e.g., due to
patients' efforts to please the therapist), but it is likely to inhibit
progress over time because the patients' behavior is maintained
through interaction with the therapist rather than by 'real-
world' environmental outcomes.
In exception to the above, one large-sample study on per-
sonality trait change and personality predictors of change in
depression severity (over a 20-session cognitive-therapy proto-
col) has been published. These researchers showed that trait
dependency decreased significantly over the 20 sessions. How-
ever, higher initial trait dependency predicted less reduction in
depressive symptoms for these clients. Overall, studies such as
this typically do not report on DPD per se; instead, they report
on traits theoretically related to DPD. Thus, treatment options
and response to treatment have yet to be established for DPD.
Concurrent Validators for Dependent
Personality Disorder
There are currently no published data on biological markers
(e.g., molecular genetics, neural substrates) of DPD. In con-
trast, there is a fair body of research that addresses relations
between DPD and other diagnoses, particularly depression. In
fact, in the 1980s, researchers were suggesting that dependency
and depression were so highly related as to be interchangeable
for some clients. However, many such studies used depen-
dency scales that are embedded within depression measures
(e.g., dependency/self-criticism in the Depressive Experiences
Questionnaire) rather than stand-alone measures of DPD or
trait dependency. As such, it remains unclear to what extent
DPD and dependency uniquely relate to depression.
In any case, later research suggests that DPD is more broadly
related to psychopathology than just depression. Although
some researchers replicated a strong relation between depres-
sion and dependency, literature reviews more typically reveal
only moderate overlap between them; on the other hand, they
find notable overlap of DPD with other psychological disorders,
including agoraphobia, phobias in general and social phobia
in particular, alcoholism, eating disorders, smoking, drug and
alcohol abuse, suicidal ideation and attempts, and somatization
disorder. DPD is also highly comorbid (overlapping or co-
occurring) with other personality disorders, including avoidant
and borderline PDs. Such studies strongly indicate that depen-
dency can and does play a broad role in clinical diagnoses.
Research on this issue supports the role of DPD in the
development and course of clinical disorders. When multiple
DSM-IV disorders are assessed with stmctured interviews, DPD
is typically associated with a broad range of disorders of many
different types. Moreover, DPD is not specifically related to any
particular disorders above and beyond their associations with
other personality disorders. Taken together, these results sug-
gest that DPD may reflect maladaptive traits that are important
to virtually all personality pathology, and that it likely taps an
even more general, cross-cutting psychological distress.
Dependent Personality (Disorder) 677
The DSM-IV states specifically that individuals diagnosed
with DPD are likely to be also diagnosed with comorbid anxi-
ety disorders, an association that was found early on and has
continued to be supported by subsequent research. Significant
relations have been found between DPD and generalized
anxiety disorder, separation anxiety disorder, social phobia,
obsessive-compulsive disorder, and panic disorder. It must be
noted, however, that the meaning of DPD's overlap with anxi-
ety disorders has been questioned because of methodological
and conceptual concerns, most notably that dependency on
others is a frequent concomitant of anxiety. As such, diagnos-
able DPD in these studies may reflect functional impair-
ment due to anxiety rather than individual differences in
dependency.
As noted previously, DPD also shares diagnostic similarity
and comorbidity with other personality disorders, most nota-
bly borderline (BPD) and avoidant PDs (AVPD). Interestingly,
relations with these disorders may be attributable to correla-
tions between dependency facets (discussed further below) as
found in trait measures and DPD criteria. That is, BPD appears
to share the 'emotional neediness' component of dependency
and even in 'remitted' BPD patients, DPD is more often
observed than any other personality disorder (although this
relation is not exclusive, as AVPD and self-defeating PD are
also observed frequently in these samples). In contrast, AVPD
appears to tap the 'perceived incompetence' facet of depen-
dency. For example, in a factor analysis of all DSM-IV PD
criteria, AVPD and DPD criteria form a single factor with
minimal cross-loadings, which was termed 'neurotic avoidant.'
Replications of this research show that AVPD correlates with
DPD criteria that addressed perceived incompetence, but not
with those pertaining to emotional neediness and attachment.
These data underscore the fact that dependency may be con-
ceptualized best as a two-facet dimension. It thus has been
suggested that DPD can be differentiated from other PDs
when the full dimensionality of dependency is assessed.
Dependent Personality Disorder and Trait Dependency
Dependent PD is one of only a few among the DSM-IV PD
diagnoses that are both conceptually defined and empirically
represented by a single personality trait, dependency, and its
concomitant behaviors. In early analyses, multiple researchers
found DPD to be among the more internally consistent of
the DSM-III and DSM-III-R PDs, making DPD one of only
three DSM-III and DSM-III-R PDs (along with antisocial and
borderline PDs) that had sufficiently acceptable psychometric
properties, including discriminant validity, to be considered
measures of coherent dimensions that were distinct from other
personality disorders. More recently, examinations of all the
DSM-IV PD criteria using both exploratory and confirmatory
factor analyses have been conducted. These studies have found
that DPD was best represented by a single latent trait dimen-
sion in both types of analyses.
Thus, current research indicates that DSM-III, -III-R,
and -DPD criteria, all essentially reflect the characteristics
of a single dimension. Later analyses of the DPD criteria
alone, however, have indicated that this dimension can be
subdivided into two correlated factors. These factors have
been found by others as well, and have been termed variously
as (1) attachment/abandonment, emotional neediness, active-
emotional, insecure attachment, and separation insecurity; and
(2) dependency/incompetence, self-perceived incompetence,
passive-submissive, and submissive.
Importantly, DPD as defined in the DSM and trait depen-
dency as defined in the personality literature bear more than
definitional similarity. Studies that use both DPD and trait
dependency measures consistently find similar results for
these variables. In a meta-analytic review of relations between
measures of personality disorder and the dominant model of
personality, that is, the five-factor model (FFM), comparable
results were found for studies that used self-report and inter-
view measures of DPD, as well as trait dependency ques-
tionnaires. These data have been replicated multiple times,
including in a meta-analysis of dependency and DPD measures
in relation to the FFM. Follow-up research indicates that both
trait dependency and SCID-II DPD scores accounted for simi-
lar variance in depression in an undergraduate sample.
Taken together, these results suggest that (1) DPD can be
well modeled using a single trait dimension (perhaps with two
subfacets) and (2) current measures of DPD and trait depen-
dency are virtually interchangeable. However, it is noteworthy
that, although dependency is well established as a risk factor
for a range of psychological disorders, there is a paucity of
research on this topic with DPD itself. Indeed, at the time of
writing of this article, Psyclnfo search of 'dependent personal-
ity disorder' in any field (e.g., abstract, keywords, title, etc.)
yielded only 223 hits. This is in contrast to similar searches for
other personality disorders, such as obsessive-compulsive,
schizotypal, borderline, or antisocial PDs (574, 1248, 4765,
and 4247 hits, respectively).
Trait Dependency, Temperament, and
Normal Personality
Although most PD research suggests that personality pathology
is best modeled using traits, it nonetheless can be argued that
current trait models do not fully capture all information rele-
vant to currently recognized personality disorders. This may be
because models such as the FFM that are designed to reflect
normal population variation fail to assess traits specific to
personality pathology. Dependency is particularly problematic
in this regard. Few measures of normal-range personality con-
tain scales to assess dependency. Moreover, there is currently
no official consensus regarding a working definition of depen-
dent personality.
Of further concern is the inconsistency in the empirical
relations between dependency and theoretically related traits.
Specifically, it has been hypothesized that dependency, which
is characterized by a need for approval and prioritization of
others' desires, should relate reliably to the FFM trait of agree-
ableness - the tendency to be motivated by social harmony
and minimizing conflict. Moreover, the placement of DPD in
the 'anxious' cluster of personality disorders and the fact that
the primary affective state associated with dependency is
worry/anxiety, suggests a significant overlap with FFM neuroti-
cism or negative affectivity (N/NA) - the tendency to experi-
ence negative emotions such as anxiety, sadness, and anger.
678 Dependent Personality (Disorder)
This is also reflected by inclusion of 'anxiousness' in the trait
description of DPD proposed for the next edition of the DSM
(DSM-5). However, only the latter theoretical relation (i.e.,
that with N/NA) has received empirical support, with one
exception, described below.
In patients, moderately high neuroticism has been found to
characterize those diagnosed with DPD. This is unsurprising,
in that DPD criteria contain clear references to neuroticism
correlates (e.g., discomfort, difficulty, and fear). Moreover,
when comparing structures of PD and FFM traits from 33
previous datasets, researchers found DPD to have the highest
loading on the largest factor, which was identified as akin to
neuroticism. Similarly, DPD typically has a high (second or
third highest) correlation with neuroticism in meta-analyses
of PD-FFM relations. However, these data do not distinguish
DPD from the other disorders that are also associated with
high neuroticism; moreover, correlations between DPD and
neuroticism tend to be moderate. As discussed later, lower
order (aka facet-level) traits appear to be needed to diagnose
DPD distinctly from other disorders.
In contrast, DPD neither loads (in factor analyses) nor
correlates with agreeableness in most of the published research
on this topic. Indeed, correlations with agreeableness generally
fall in the near-zero range and, in some cases, stronger relations
have been found between measures of DPD and the FFM when
all agreeableness facets were removed, and neuroticism and
conscientiousness facets were added. However, one recent
study demonstrated that a 'maladaptive agreeableness' scale -
created by altering agreeableness items so that they measure a
maladaptive version of same content - correlated more highly,
although still only moderately, with a wider range of depen-
dency measures than did the original scale. Thus, somewhat
different results might have been found in the other studies if
'maladaptive agreeableness' had been used. Alternatively, dif-
ferent results may obtain simply because the revised scale
capitalizes on its conceptual and empirical similarity to
neuroticism.
Models of Dependency
Four Factors
As noted previously, when modeled as a single trait, depen-
dency is not well represented in trait models such as the FFM.
However, current research suggests that dependency is best
modeled as multiple factors. For example, current definitions
hold that dependency has at least four principal interrelated
components: motivation - strong desires to obtain and main-
tain supportive relationships with others; cognition - a percep-
tion of oneself as ineffective and incapable, in contrast to
a perception of others as effective and capable; affect - anxiety
and worry in the absence of support from others; and
behavior - interpersonal yielding, and help and approval seek-
ing. Theoretically, this description of dependency suggests
an FFM. However, dependency items generally contain multi-
ple domains, and to date there are no four-factor measures
of dependency. Although the APA notes that self-reported
dependency is higher in women than in men, there is no
hypothesized, gender-based difference in trait content or
structure.
Three Factors
In contrast to the four factors described above, early research
conceptualized and assessed dependency as having three
facets: low self-confidence and emotional reliance on others
versus assertion of autonomy, which was thought specifically
to assess the opposite or low end of the dependency dimen-
sion. The Interpersonal Dependency Inventory (IDI) is based
on this model. The IDI autonomy scale's items suggest a path-
ological rejection of and inability to connect with others, and it
is one of only a few attempts to capture the low end of depen-
dency. However, recent research in student samples (some of
which utilize the IDI) suggests that autonomy and dependency
capture distinct trait variance, and are best modeled separately
rather than as opposites.
Yet another three-factor model of dependency draws from
interpersonal theory and the interpersonal circumplex (IPC),
which conceptualizes interpersonal behavior as reflecting two
underlying dimensions of agency and communion. This
model also draws from the FFM, of which extraversion (the
tendency to be sociable and to experience positive emotions)
and agreeableness dimensions represent an alternative rotation
of the IPC dimensions. The IPC model reflects three vectors
of dependent personality: love dependency, characterized by
warmth, nurturance, and a need for social interaction;
exploitable dependency, characterized by being unassuming and
by fear of judgment or evaluation; and submissive dependency,
characterized by unassertiveness and the tendency to yield to
others. The resultant measure - the Three-Vector Dependency
Inventory (3VDI) - contains items to measure three types of
dependency following this model.
It is important to note that, unlike other conceptualizations
of dependency, these vectors all hypothetically stem from the
same motivation - obtaining and maintaining personal rela-
tionships - as manifested in different behavioral and psycho-
logical strategies. The authors presented evidence that each
dependency facet was associated with a relatively distinct
behavioral phenotype (e.g., attachment style and behavior,
parental representations, loneliness [in men only], location
in the IPC). Further, data show that the FFM traits of openness
and conscientiousness related meaningfully and differentially
to the three dependency factors. Only high neuroticism typifies
all three types of dependency, again suggesting that increased
neuroticism is the FFM domain that most clearly and consis-
tently marks pathological dependency.
Finally, a later study sampled more widely from the depen-
dency domain, and proposed a model consisting of three
single-dimension dependency-relevant constructs reflecting
the four-factor multicomponent view of dependency. Thus,
the Relationship Profile Test (RPT) models dependency as
consisting of three types: destructive overdependence (patho-
logically high dependency), healthy dependency, and dysfunc-
tional detachment. Contrary to frameworks that conceptualize
dependency and detachment as opposing traits, the two dys-
functional types were found to be orthogonal rather than
opposed. The RPT has shown moderate levels of test-retest
reliability over 85 weeks and 3 years, as well as expected rela-
tions to measures of interpersonal relationships. Interestingly,
it is also the only current measure to include a formal scale
of 'healthy' dependency, or dependency as manifest in
Dependent Personality (Disorder) 679
normative interpersonal relationships (e.g., as in a marriage or
in a family).
Two Factors
The concept of dependency as a multifaceted structure also is
reflected in a considerable body of research on depression. Much
of this research has used the term 'sociotropy' rather than 'depen-
dency,' but the constructs appear to be nearly identical in that it
emphasizes nurturance and an excessive need to please others.
Factor analysis of the sociotropy items of the Sociotropy-
Autonomy Scale (SAS) yields three distinct subscales: concern
about disapproval, attachment, and ingratiation. Analyses of the
(nonoverlapping) set of autonomy items yielded three different
subscales. In contrast, later research presents a personality-based
model of depression that relies heavily on two correlated depen-
dency facets: dependency, or excessive reliance on interpersonal
support; and self-criticism, or excessive emphasis on achieve-
ment and fear of failure. These bear superficial similarity to
emotional reliance and low self-confidence, the two dependency
scales of the IDI. Measures of depression typically do not empha-
size autonomy as (negatively) related to dependency.
Several two-factor models for dependency have also been
posited in the personality literature. First - and as noted in the
RPT - one can conceptualize dependency at both adaptive
(connectedness) and maladaptive (neediness) levels, a distinc-
tion also reflected by 'love dependence' and 'submissive depen-
dence.' Another measure, the Dimensional Assessment of
Personality Pathology (DAPP) also posited two components
of trait dependency as separate scales in a multidimensional
measure of personality pathology: submissiveness, which also
encompasses low self-esteem and need for advice/approval;
and insecure attachment, consisting of need for affection, secu-
rity, and proximity to an attachment figure. The scales show
high internal consistency and strong dependability (i.e., short-
term stability) over a 3-week test-retest interval, and have been
shown to relate meaningfully to other measures of dependency
such as that of the Schedule for Nonadaptive and Adaptive
Personality (SNAP/SNAP-2).
A two-factor structure for dependency was also found in
factor analyses of the diagnostic criterion for DPD both with
and without criteria from other PDs. When factored with cri-
teria from other PDs, the DPD criteria did not cohere, but
split between those reflecting incompetence or lack of self-
confidence (DSM criteria 2, 3, and 4), which formed a factor
with most of the avoidant PD criteria, and those reflecting
emotional neediness and fear of being alone (DSM criteria 6,
7, and 8), which formed a factor with three criteria from
borderline PD. These results were later confirmed by other
researchers, who also conducted exploratory analyses using
PD criteria from the DSM-IV. Again, most criteria from depen-
dent and avoidant PDs grouped to form a single factor, which
the authors named 'neurotic/avoidant.' Notably, this factor
had the strongest negative correlation with daily functioning,
suggesting that these types of dependent characteristics are
quite detrimental to overall functioning.
Taken together, these results indicate that DPD as a diagno-
sis has a clear two-factor structure ('perceived incompetence'
and 'attachment/abandonment') reflecting the distinction
posited by the DAPP, but with two of the eight diagnostic
criteria - Criterion 1 (difficulty making decisions) and
Criterion 5 (going to excessive lengths to ingratiate oneself to
others) - not highly related to either factor. Similar findings
emerged in a subsequent confirmatory factor analysis of DPD
criteria alone, where a two-factor model of 'perceived incom-
petence' and 'dysfunctional attachment' provided the best fit
for DPD. In these data, Criterion 5 again related only weakly to
other criteria, and Criterion 3, difficulty expressing disagree-
ment, correlated more strongly with avoidant than dependent
PD. This suggests that the criteria for DPD could be best revised
by eliminating Criteria 3 and 5, thus reducing the disorder's
bifactorial nature. An alternative conclusion would be that
the disorder is best represented by two trait dimensions.
Similar results have been found in student samples, suggest-
ing that the structure of dependent personality (and DPD) is
consistent across normal and abnormal populations. One such
study combined and assessed all previously described measure-
ment models, using measures consisting of one, two, or three
dependency factors. Factor analysis at the scale level yielded
three interpretable factors. The first factor consisted of both
pathological dependency (e.g., SNAP-2 dependency scale) as
well as lack of self-confidence or willingness to assert oneself in
the presence of others (e.g., IDI low self-confidence), and was
termed 'passive-submissive dependency' ('P-submissive'). The
second factor primarily comprised measures tapping emo-
tional neediness in personal relationships (e.g., 3VDI love
dependency, DAPP insecure attachment), and was termed
'active-emotional dependency' (A-emotional). The final factor
consisted of only two scales that measured theoretically low-
end dependency, and was named 'detachment/ autonomy' after
the two scales' names. All but one scale loaded strongly on only
one factor and the factors were moderately correlated, support-
ing the notion that dependency is at least bifactorial and not
simply a single unified construct.
It is noteworthy that dependency scales purporting to
reflect DPD as conceptualized in the DSM-IV (e.g., SNAP-2-
dependency) all loaded solely or primarily on P-submissive
(the first two scales also having moderate negative loadings
on autonomy/detachment). This suggests that current self-
report DPD diagnostic measures assess low self-confidence
(P-submissive) more than emotional neediness (A-emotional),
despite the fact that some of the diagnostic criteria for DPD
appear to be highly affective/needy in nature. In fact, three (of
its eight) DSM diagnostic criteria specifically mention neuroti-
cism in some form. These results are in contrast to those
reported previously, where DPD criteria formed two dimen-
sions, which may be due to differences in the study methods:
In the study finding two factors, the DPD data were derived in
the context of a structured interview for all DSM-IV PDs, and
were analyzed at the criterion level, thus increasing the possi-
bility that a given set of PD criteria could load differentially on
distinct factors, whereas in the student sample, the data were
analyzed at the scale level, which may have favored emergence
of the most dominant component of each scale.
One factor
Despite these results, DPD as a diagnostic category has
remained a unitary construct, and measures assessing DPD as
a disorder necessarily reflect this conceptualization. That is,
680 Dependent Personality (Disorder)
ultimately a diagnosis is either made or not made, and the
categorical diagnosis does not reflect varying levels of depen-
dency as a trait, nor is the possible bifactorial nature of
dependency reflected in the diagnosis (i.e., the diagnosis
alone does not indicate the balance of 'perceived incompe-
tence' vs. 'dysfunctional attachment'). One such measure, the
Dependent Personality Questionnaire (DPQ), was developed
specifically to reflect DPD as in the DSM-IV and ICD-10.
Although the authors also took into account research defining
trait dependency, and the measure yields dimensional scores,
the DPQ consists of eight items essentially reflecting the eight
diagnostic criteria for DPD, and the criteria are used to provide
a diagnosis. The DPQ was normed on a patient sample and
showed high levels of specificity and sensitivity in this popula-
tion, but its reliability over time has yet to be established.
Similarly, the DPD scale of the Wisconsin Personality Dis-
orders Inventory (WISPI) is a self-report questionnaire ratio-
nally derived from the DSM-III-R PD criteria, and is intended
to diagnose DPD from an interpersonal perspective, that is,
based on interpersonal and ttait theory. Nonetheless, the
WISPI-DPD scale ultimately yields information about the pres-
ence or absence of dependency as a disorder, and not as trait
dependency. The scale shows a high level of internal consis-
tency in patient populations, has demonstrated acceptable
levels of specificity and sensitivity, and relates meaningfully
to established measures of dependent personality.
Notably, several recent measures of trait dependency also
treat dependency as a unitary dimension. For example, the
Schedule for Nonadaptive and Adaptive Personality, Version
2 (SNAP-2) contains a single unidimensional dependency
scale, primarily reflecting decision making and (low) self-
confidence, and includes few items assessing pathological attach-
ment. This is similar to the way in which dependency is assessed
by measures drawing suictly from the ICD-10 and DSM-IV. For
this measure, internal consistency (alpha) in two psychiauic
patient samples was very high, and retest reliability was accept-
able across multiple time frames (e.g., lweek, 1-3 months,
6 months) in a wide variety of patient and student samples.
Detachment/Autonomy
Although not technically part of dependent personality,
detachment/autonomy appears regularly in the dependency
literature (as detailed above) and thus deserves some mention
in this article. Initially, detachment/autonomy and depen-
dency may appear to exist along a single conceptual dimen-
sion. Flowever, as indicated earlier, evidence suggests that - at
least as concerns measures of detachment/autonomy that have
been developed in the context of assessing dependency - they
are uncorrelated phenomena with distinct variance. In studies
of all personality disorder criteria, the criteria pertaining to
aloof detachment and dependency do not form a single factor,
and do not appear to represent a single dimension.
Notably, the scales reported in such studies represent only a
portion of the larger body of research on trait detachment/
autonomy, much of which exists separate from that on depen-
dent personality. In general, results from studies on detach-
ment suggest that detachment itself is a multifaceted construct
that includes a dimension termed self-governance, which
correlated strongly with FFM extraversion and conscientious-
ness. Thus, it may be better to conceptualize the opposite end
of pathological dependence simply as healthy levels of depen-
dency rather than as pathological independence. Flowever,
more research is needed to confirm this conclusion.
Dependency and the DSM
DPD relates significantly to a wide range of psychopathology
and, not surprisingly, trait dependency appears to have similar
relations with these disorders. Specifically, in addition to being
associated with DPD and other personality pathology, trait
dependency has been linked also with mood disorders, sui-
cidal ideation and attempts, anxiety disorders, eating disorders,
and adjustment disorder.
Reconceptualizing DSM-V PDs dimensionally - more spe-
cifically, multidimensionally - would likely facilitate explora-
tion of the role of personality in these disorders, without the
difficulties entailed in examining comorbid diagnoses. As indi-
cated earlier, many researchers have suggested using the FFM
for assessing personality in DSM-V. Although this has the draw-
back that the FFM does not assess dependency per se, use of the
'maladaptive agreeableness' dimension would ameliorate this
issue, at least to some extent. Dependency perhaps could more
easily be evaluated as a facet or correlate of neuroticism, as is
suggested in the proposed changes for the DSM-5.
Interestingly, some researchers have proposed inclusion
of more specific FFM components to address the incomplete
relation of FFM traits to current DSM-IV PD diagnoses.
These components cast a broader net than standard FFM
measures to encompass domains of functioning and the
environment and to include basic tendencies, characteristic
adaptations, self-concept, objective biography, and external
influences. Inclusion of such components may provide an
alternative framework in which to conceptualize personality
disorder in general and traits such as dependency in particular.
Some evidence supporting a more detailed approach to
maladaptive-range personality traits has already been demon-
sttated. For example, the two factors of dependency (depen-
dency/incompetence and attachment/abandonment) have
been found to relate differentially to maladaptive cognitive
schemas in patients. Similarly, more significant correlations
are found between dependency and the FFM when more spe-
cific (facet) scales are used. The field would benefit from fur-
ther research exploring relations of the identified two-factor
structure of trait dependency with both the 'standard' FFM and
an expanded FFM with coverage of the broader domains men-
tioned above. This is particularly tme given that dependency is
primarily viewed as a pathological trait, and thus is situated in
and reflects current psychopathology research more than
normal-range personality research. Integrative research that
bridges across the adaptive and maladaptive ranges of this
personality traits would be especially beneficial.
See also: Big Five Model and Personality Disorders; Personality,
Structure; Psychopathology: Diagnosis, Assessment, and
Classification.
Dependent Personality (Disorder) 681
Further Reading
American Psychiatric Association (1 994) Diagnostic and statistical manual of mental
disorders: DSM-IV, 4th ed. Washington, DC: American Psychiatric Association.
American Psychiatric Association (2010) Proposed draft revisions to DSM disorders
and criteria. Retrieved from http://www.dsm5.org.
Bornstein RF (1993) The Dependent Personality. New York: Guilford.
Clark LA (2007) Assessment and diagnosis of personality disorder: Perennial issues
and emerging conceptualization. Annual Review of Psychology 58: 227-258.
Grant BF, et al. (2004) Prevalence, correlates, and disability of personality disorders in
the United States: Results from the national epidemiologic survey on alcohol and
related conditions. Journal of Clinical Psychiatry 65: 948-958.
Gude T, Hoffart A, Hedly L, and Ro 0 (2004) The dimensionality of dependent
personality disorder. Journal of Personality Disorders 18: 604-610.
Livesley WJ, Shroeder ML, and Jackson DN (1990) Dependent personality disorder.
Journal of Personality Disorders 4: 131-140.
Morgan TA and Clark LA (2010) Passive-submissive and active-emotional trait
dependency: Evidence for a two-factor model. Journal of Personality 78: 1 325-1 352.
O’Connor BP (2002) The search for dimensional structure differences between
normality and abnormality: A statistical review of published data on personality and
psychopathology. Journal of Personality and Social Psychology 83: 962-982.
Saulsman LM and Page AC (2004) The five-factor model and personality disorder
empirical literature: A meta-analytic review. Clinical Psychology Review
23: 1055-1085.
Simon W (2009) Follow-up psychotherapy outcome of patients with dependent,
avoidant, and obsessive-compulsive personality disorders: A meta-analytic review.
International Journal of Psychiatry in Clinical Practice 13: 153-165.
Strack S and Lorr M (1997) Invited essay: The challenge of differentiating normal and
disordered personality. Journal of Personality Disorders 11: 105-122.
Trull TJ and Durrett C (2005) Categorical and dimensional models of personality
disorder. Annual Review of Clinical Psychology 1 : 355-380.
Widiger TA, Simonsen E, Sirovatka PJ, and Regier DA (eds.) (2007) Dimensional
Models of Personality Disorders. Arlington, VA: American Psychiatric Association.
World Health Organization (2007) International Statistical Classification of Diseases and
Related Health Problems: 10th revision for 2007. Retrieved from http://apps.who.
int/classifications/apps/icd/icd10online/.
Depression
R Ingram, University of Kansas, Lawrence, KS, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Anhedonia A lack of interest in daily activities, or the ability
to gain pleasure from these activities.
Bipolar disorder An affective disorder that is divided
into Bipolar I disorder, which is characterized by the
occurrence of mania, and Bipolar II disorder which is
characterized by hypomania.
Cardinal features Symptoms that must be present to qualify
for a diagnosis of depression. The cardinal features of
depression are either sad mood or a lack of interest
in daily activities.
Epidemiology Information about the prevalence of
disorders in a population. In the case of disorders such as
depression, a prevalence rate refers to the number of people
who have the disorder during a particular time period (e.g.,
the percentage of people in a given location diagnosed with
major depressive disorder within a 1-year time period).
Socioeconomic status differences are quite small.
Hypomania Episodes characterized by the same symptoms
as manic episodes, but that are shorter and associated with
less impairment.
Mania A state that frequently includes increased energy,
inflated self-esteem, a decreased need for sleep, and
involvement in potentially harmful activities.
Negative affectivity The tendency to be chronically
distressed and to chronically view oneself negatively.
Negative affectivity may be a precursor to both depression
and anxiety.
Nosology The diagnosis of disorders into discrete entities
with the assumption that each disorder that is nosologically
classified will have a different cause, course, prognosis,
and treatment response.
Unipolar disorder An affective disorder that is
characterized by depression without any evidence for the
occurrence of manic episodes.
Depression
Depression has been recognized for as long as history has been
recorded. Indeed, in the fourth century B.C., Hippocrates
described depression and located its cause in an excess of
black bile. Descriptions of what appear to be depression can
also be found in the Bible, while other historical references
show that around 120 AD, Araetus of Cappadocia described
sadness, suicidal tendencies, and apathy as symptoms of
depression. More 'modern' accounts show that Freud's descrip-
tions of melancholia resemble contemporary accounts of
depression, and that he recognized that the emotional and
psychological factors play a role in this state.
Today, depression means different things for different peo-
ple. To many members of the general public, depression can
mean feelings of unhappiness or sadness, for example, "I'm
feeling a little depressed today." For mental health profes-
sionals, on the other hand, depression is a disease or a psychi-
atric disorder characterized by depressed mood that interferes
with living, and that is associated with symptoms such as loss
of appetite, lack of concentration, and suicidal thoughts. In
1987, Kendall, Hollon, Beck, Hammen, and Ingram summed
up the various meanings of depression in this way:
The professional use of the term depression has several levels of
reference: symptom, syndrome, nosologic disorder. Depression
itself can be a symptom - for example, being sad. As a syndrome,
depression is a constellation of signs and symptoms that cluster
together (e.g., sadness, negative self-concept, sleep and appetite
disturbances). The syndrome of depression is itself a psychological
dysfunction but can also be present, in secondary ways, in other
diagnosed disorders. Finally, for depression to be a nosologic cate-
gory, careful diagnostic procedures are required during which other
potential diagnostic categories are excluded. The presumption, of
course, is that a discrete nosologic entity will ultimately prove to be
etiologically distinct from other discrete entities, with associated
differences likely in course, prognosis, and treatment response.
(Beck, 1967, p. 290)
The focus of this article is the nosologically defined disorder of
depression, in particular, the idea that depression, as it is used
by mental health professionals, is an affective or mood disor-
der that exceeds normal sadness and mood fluctuations. Even
more specifically, the focus will be on unipolar mood disorder
or major depressive disorder, even though the other classifica-
tions of mood disorder will be noted.
Definition of Depression
Symptoms of Depression
Professional definitions of depression in the United States
begin with the fourth edition, text revision, of the Diagnostic
and Statistical Manual of Mental Disorders (DSM-IV-TR). The
DSM-IV is the official diagnostic classification system of the
American Psychiatric Association and provides the criteria that
are used to diagnose depression. The DSM-IV-TR will soon be
replaced by the fifth edition of the DSM, but it is unlikely that
any fundamental changes in the definition of depression will
take place.
The 'cardinal features' of depression are either sad mood or
a lack of interest in daily activities or the ability to gain pleasure
from these activities (known scientifically as anhedonia). Car-
dinal features are those features that must be present before the
disorder can be diagnosed. Beyond these cardinal features,
symptoms of depression fall into several categories that are
682
Depression 683
Table 1 Symptoms of depression according to the Diagnostic and
Statistical Manual of Mental Disorders, 4th edn. Text Revision
1. Depressed mood for most part of the day
2. Diminished interest or pleasure in activities
3. Significant weight loss or gain, decrease or increase in appetite
4. Insomnia or hypersomnia
5. Psychomotor retardation or agitation
6. Fatigue or loss of energy
7. Feelings of worthlessness or excessive or inappropriate guilt
8. Diminished ability to concentrate or think
9. Recurrent thoughts of death, suicidal ideation, a specific plan, or a
suicide attempt
grouped together based on their similarities. As would be
expected, mood symptoms represent predominant features of
depression and are reflected in sad mood. Motivational symp-
toms include behaviors that are related to goal-directedness.
Depressed people often suffer a deficit in this area and some
may find it difficult to do even the smallest task. Somatic
symptoms refer to the physical changes that may accompany
depression and include changes in sleeping patterns, appetite
disturbances, and loss of sexual interest. Cognitive symptoms
reflect the ability to concentrate and make decisions.
Symptoms of depression may vary according to an indivi-
dual's age and culture. Children who are depressed, for instance,
may express symptoms of irritability rather than sadness or they
may fail to make expected weight gains rather than lose weight.
On the other end of the age continuum, older adults are more
likely than younger adults to experience symptoms such as
loss of appetite, loss of interest, and thoughts of death. Cultural
differences also exist in the report of depressive symptoms.
One study, for example, found that depressed Jewish patients
reported more somatic symptoms and less guilt than did non-
Jewish patients. Another study that examined depressive symp-
tomatology in American, Korean, Philippine, and Taiwanese
college students found that Taiwanese students reported the
lowest numbers of somatic symptoms and the highest numbers
of affective symptoms. Hence, one's age and culture seems to
affect not whether depression is experienced, but rather how it
is expressed.
DSM-IV-TR lists nine 'officially recognized' symptoms of
depression (see Table 1), although it is important to note
that this list does not include all possible symptoms. Pessi-
mism, for example, is often seen in depressed individuals, but
is not included in the list. In order to receive a diagnosis of
depression, the individual must evidence at least five of the
nine symptoms, and one of these symptoms must be either
depressed mood or anhendonia. Moreover, these symptoms
must be experienced for at least a 2-week period before a
diagnosis can be assigned. Given the five symptom cutoff for
a diagnosis, it is important to note that even though there are
nine recognized symptoms, not all depressed people will expe-
rience all nine of these symptoms.
The Relationship Between Depression and Anxiety:
Comorbidity
Comorbidity refers to the occurrence of more than one disor-
der at the same time. Although researchers and clinicians
generally acknowledge depression as a distinct disorder, it
does overlap with a variety of other difficulties such as sub-
stance abuse. However, much research on this overlap has
focused on the relationship between anxiety and depression.
This is not surprising, given the high rates of comorbidity
found in studies of the two disorder types. One possible expla-
nation provided for such overlap lies in the concept of 'nega-
tive affectivity.' Originally proposed in 1984, Watson and
Clark described individuals with high levels of negative affec-
tivity as having a tendency "to be distressed and upset and have
a negative view of self, whereas those low on the dimension are
relatively content and secure and satisfied with themselves"
(p. 465). Other characteristics of high negative affectivity
include nervousness, tension, worry, anger, scorn, revulsion,
guilt, self-dissatisfaction, rejectedness, and sadness.
Both anxiety and depression seem to consist of high nega-
tive affectivity (general distress). There are, however, impor-
tant differences between depression and anxiety. While
both depression and anxiety are characterized by high levels
of negative affect, only depression is related to low levels of
positive affect. Thus, depressed individuals tend to display
both high negative affect and low positive affect, whereas
anxious individuals display high negative affect and may or
may not have low positive affect. Anxious people are also
untiringly characterized by physiological hyperarousal. These
distinctions have been characterized as a tripartite model that
describes those features that are common to both depression
and anxiety as well as those features that are unique to each
disorder.
Classification of Mood Disorders
Normal Depression
Depression can be classified on a continuum ranging from
normal depression to major depressive disorder. Normal
depression refers to the mood swings that every individual
experiences and is usually expressed as sadness. Normal
depression is generally viewed as a symptom and thus, beyond
sad mood, few if any of the other depression symptoms are
present.
Subclinical Depression
Subclinical depression is a more severe form of depression that
includes not only sad mood but also some of the other symp-
toms of depression. There are differences in terminology, so
subclinical depression is similar to 'minor depression' which is
an exploratory category of possible depression diagnoses (dis-
cussed later in this article). Subclinical depression usually
refers to an elevation of depressive symptoms that usually do
not warrant treatment but that may nevertheless interfere with
an individual's ability to function effectively.
Earlier the DSM-IV-TR was noted. Although there are a
number of mood disorders included in DSM-IV-TR (e.g.,
'mood disorder not otherwise specified'), the manual lists five
major types of mood disorders. These include (1) major
depressive disorder, (2) dysthymic disorder, (3) bipolar I
disorder, (4) bipolar II disorder, and (5) cyclothymic
disorder.
684 Depression
Major Depressive Disorder
As noted, for a diagnosis of major depressive disorder, DSM-IV
specifies that at least five symptoms must occur for a period of at
least 2 weeks. Major depressive disorder can be further classified
according to severity (i.e., mild, moderate, severe without psy-
chotic features, severe with psychotic features), course (e.g.,
single episode versus recurrent episodes), and presentation
(e.g., with catatonic features, with melancholic features). Psy-
chotic features of depression include such experiences as delu-
sions (i.e., false beliefs) and hallucinations (i.e., sensory
experiences which have no basis in reality). Catatonic features
of depression involve psychomotor disturbances such as exces-
sive movement or stupor. Melancholic features include the
inability to experience pleasure even when good things happen
and a lack of interest in previously pleasurable activities. No
matter what the specific characteristics of a given individual's
disturbance, major depressive disorder is, by definition,
extremely distressing to the sufferer and is associated with sig-
nificant impairment in important areas of the person's life (e.g.,
at work, home, or school).
Dysthymic Disorder
Dysthymic disorder is characterized by a chronic depressed
mood which lasts at least 2 years in adults and at least 1 year
in children and adolescents. According to DSM-IV-TR, this
depressed mood is accompanied by at least two of the follow-
ing six depressive symptoms: (1) poor appetite or overeating,
(2) insomnia or hypersomnia, (3) low energy or fatigue,
(4) low self-esteem, (5) poor concentration or difficulty
making decisions, and (6) feelings of hopelessness. Because
fewer depressive symptoms are required to make a diagnosis,
dysthymic disorder is often considered a milder form of depres-
sion than major depressive disorder. However, it can be just as
upsetting to the sufferer and can cause just as much impairment.
In addition, when dysthymic disorder occurs along with major
depression, the individual is considered to be suffering from
a 'double depression.' The co-occurrence of major depressive
disorder and dysthymia is not uncommon.
Bipolar I Disorder
The hallmark characteristic of bipolar I disorder is mania.
According to DSM-IV-TR, a manic episode is characterized by
elevated, expansive, or irritable mood that is persistent and
distinctly different from normal elevated or irritable moods.
This period is accompanied by at least three of the following
seven possible symptoms: (1) inflated self-esteem, (2) a
decreased need for sleep, (3) unusual talkativeness, (4) the
feeling that one's thoughts are racing, (5) increased distractibil-
ity, (6) increased activity, and (7) involvement in pleasurable
but potentially harmful activities (e.g., sexual indiscretions).
Bipolar I disorder is typically recurrent; according to DSM-
IV-TR, additional episodes occur in more than 90% of indivi-
duals who have had a single manic episode. The manic
episodes of those with bipolar I disorder are often intermixed
with periods of depression. Like those with major depressive
disorder, people with bipolar I disorder may exhibit psychotic,
catatonic, and melancholic features as part of either their
mania or their depression.
Bipolar II Disorder
Bipolar II disorder is characterized by periods of hypomania
intermixed with periods of depression. Hypomanic episodes
are characterized by the same symptoms as manic episodes.
However, hypomanic episodes may be shorter (e.g., 4 days in
duration) and are associated with less impairment. Whereas
manic episodes may include psychotic features, interrupt daily
functioning, and require hospitalization, hypomanic episodes
typically do not. The depression experienced as part of bipolar
II disorder, however, can be just as severe as that experienced in
major depressive disorder and bipolar I disorder.
Cyclothymic Disorder
Cyclothymic Disorder is characterized by hypomanic periods
intermixed with depressive periods that are not as severe as
those experienced in major depressive disorder, bipolar I
disorder, and bipolar II disorder. In cyclothymia, the periods
of mood disturbance may alternate rapidly, with little respite
from affective difficulties. For a diagnosis of cyclothymia,
these periods of shifting moods must be problematic for at
least 2 years in adults and at least lyear in children and
adolescents.
Exploratory Categories of Depressive Disorders
In addition to the five official diagnoses, DSM-IV-TR has
denoted four classifications for further study that include
depression as a significant component. It remains to be seen
whether, and in what form, these syndromes will be included
in DSM-V. Even though they are currently exploratory, they are
described here because these may represent serious problems.
They are (1) premenstrual dysphoric disorder, (2) minor
depressive disorder (MDD), (3) recurrent brief depressive dis-
order (RBDD), and (4) mixed anxiety-depressive disorder.
Premenstrual dysphoric disorder
Premenstrual dysphoric disorder is characterized by several
hallmark symptoms of depression (e.g., decreased interest in
usual activities, depressed mood, difficulty sleeping, or sleep-
ing too much) in addition to symptoms such as affective
labiality, feelings of being overwhelmed or out of control,
and food cravings. In order to meet the criteria that have
been proposed for this diagnosis, such symptoms must have
occurred during the late luteal phase of most of a woman's
menstrual cycles in the past year. As a number of authors have
pointed out, such a classification has potentially serious social,
political, and legal ramifications for women. For example,
some have argued that if this classification is adopted as an
official diagnosis, then women might be stigmatized as more
unstable than or inferior to men. Arguments such as these keep
the classification of premenstrual dysphoric disorder a topic of
considerable debate.
Minor depressive disorder
MDD is characterized by fewer depressive symptoms than are
seen in major depressive disorder. The level of impairment
is also less than that associated with major depressive disorder.
To meet the proposed criteria for MDD, a person must
Depression 685
demonstrate either a depressed mood or loss of interest and
two additional symptoms of a major depressive episode. If this
classification were included in future DSM editions as a disor-
der, it would constitute a residual category to be used only after
the other mood disorders have been ruled out.
Recurrent brief depressive disorder
The principal difference between RBDD and major depressive
disorder is one of duration. RBDD is characterized by periods
of depression that meet all of the criteria for a major depressive
episode except for duration requirement. While in major
depressive episodes, symptoms must last at least 2 weeks, in
recurrent brief depressive episodes, symptoms must last at least
2 weeks but less than 14 days. In addition, these brief episodes
must occur at least once a month for 12 months to meet criteria
for the classification of RBDD. RBDD is quite similar to major
depressive disorder in its age of onset and family incidence
rates, thus raising questions as to whether this should be
considered a distinct disorder.
Mixed anxiety-depressive disorder
The impetus behind a mixed anxious-depressed category lies
in the finding that there are many people suffering from symp-
toms of anxiety and depression who do not meet criteria for
any DSM anxiety or mood disorder, but who are nonetheless
significantly impaired by their difficulties. The classification of
mixed anxiety-depressive disorder (MADD) is characterized by
a dysphoric mood for at least 1 month in addition to at least
four additional symptoms that primarily reflect anxiety (e.g.,
mind going blank, worry, hypervigilance). The primary argu-
ment in favor of adopting this proposed disorder is that it
would cover the large number of people who have significant
impairment linked to depression and anxiety but who do not
fall into any currently existing diagnostic category. The primary
argument against this classification is that people suffering
from both depression and anxiety could in fact be categorized
into already existing disorders with the use of more precise
assessment methods.
Epidemiology
Epidemiology refers to information about the incidence and
prevalence of disorders in a population, while a prevalence rate
refers to the number of people who have a given disorder
during a particular time period (e.g., the percentage of people
in a given location diagnosed with major depressive disorder
within a 1-year period of time). An incidence rate refers to the
number of new cases of a disorder which occur during a given
time period (e.g., the number of people diagnosed with dys-
thymic disorder during April 2010). Because the distribution of
a disorder can be examined to determine whether it correlates
with other factors or not, epidemiological information can be
important for understanding some of the possible causes and
correlates of depression.
To assess the prevalence, three major epidemiological stud-
ies have been conducted to determine the rates of psychiatric
disorders, including depression, in the United States. The first,
the National Institute of Mental Health Epidemiologic Catch-
ment Area (ECA) study, interviewed more than 20 000
community and institutionalized adults in five sites, including
New Haven, Connecticut; Baltimore, Maryland; St. Louis,
Missouri; Durham, North Carolina; and Los Angeles, California.
The second major study to examine the rates of psychiatric
disorders was the National Comorbidity Survey; the third epi-
demiological study was the NCS-Replication which surveyed
approximately 9000 individuals over the age of 18.
Prevalence
Prevalence in the United States
The ECA study found the lifetime prevalence of major depres-
sion (i.e., the number of people experiencing major depression
during any point in life) to be 4.9% and the lifetime prevalence
of dysthymia to be 3.2%. Alternatively, the national comorbid-
ity survey (NCS) reported much higher prevalence rates:
approximately 15% for lifetime major depression, while the
NCS-R found a rate of 16.6%. Other studies have reported rates
as high as 25%.
Discrepancies between studies are probably accounted for
by different assessments, different definitions, and different
samples. For example, differences between the ECA and the
NCS studies can be found assessment instruments used, slightly
different diagnostic criteria, and different age ranges studied
(i.e., the ECA sample was 18 years of age or older, whereas
the NCS sample ranged in age from 15 to 54years). Given the
various estimates across studies, Kessler has suggested that
the actual prevalence rate of major depressive disorders is
around 17%.
International prevalence rates
International lifetime prevalence rates vary widely across a
variety of different studies (e.g., from a low of 3.3% in Seoul
to a high of 15.1% among New Zealand residents aged 25-46).
Such differences may indeed reflect true international differ-
ences in the occurrence of depression, but other factors such as
cultural differences in the sensitivity of the instruments used to
assess depression, and different sample ages may also account
for this range. Some data on point prevalence (the incidence of
a disorder at a particular point in time) are available from the
WHO's international consortium on psychiatric epidemiology
(ICPE). The ICPE seeks to consolidate data from international
epidemiological surveys and finds that point prevalence rates
are between 2% and 4% for various samples of adults across
countries.
Gender, Age, and Culture
Gender
There are consistent and substantial gender differences in
depression rates; roughly twice as many women as men suffer
from major depression, and women tend to report more severe
depression than men. Furthermore, the gender difference
occurs in different and across demographic and cultural
groups. These differences tend to be observed around the age
of 1 2 or 1 3 years (around the time of puberty) . Prior to this age,
depression is uncommon, although when depression sets in,
girls and boys are equally affected. Moreover, in adults, recur-
rence rates tend not to differ, so that even though women are
more likely to experience an initial onset, once an onset has
686 Depression
occurred, men and women experience generally equivalent
rates of depression. Additionally, there do not seem to be
differences in the length of depressive episodes between men
and women.
A variety of explanations of these gender differences have
been proposed, although research has yet to confirm the causal
validity of these possibilities. Biological views focus on hor-
monal explanations, genetic factors, and biological stress reac-
tivity. Hormonal explanations, for example, note that many
women experience mood disorders at times of peak hormonal
change, such as puberty, the premenstrual phase of the men-
strual cycle, and the postpartum period. Some psychological
theories focus on greater interpersonal orientation; research
suggests that women tend to be socialized and to be more
likely than men to look excessively to others for reassurance.
Another psychological variable, rumination (a persistent and
obsessive thought about a certain topic) has been suggested
to propel women into a cycle of negative affect that eventuates
into a depressive state. Social explanations have also been
proposed and focus on the greater likelihood that women
experience from traumatic events (abuse) or chronic negative
events (such as poverty or sexual harassment). Each of these is
plausible, and it seems likely that gender differences will even-
tually be best explained by a model that integrates biological,
psychological, and social explanations.
Age
Depression has an average age of onset in early adulthood,
typically in the early to mid-20s, although some individuals
experience depression at a later age. However, depression rates
vary with age and gender. Not surprisingly, depression is more
common in younger than older adults, with rates being highest
for individuals from 18 to 45. Men have the highest rates
among those aged 18-29, with a large decline in incidence
for those aged 45 and older. For women, the incidence of
major depression is highest for ages 30 through 44 and with
declines in incidence not dropping substantially until age 65.
For both genders, rates of first onsets are considerably lower for
individuals over 65 years old. There are also differences in
symptom patterns in different age groups, particularly at the
extremes of these groups. Children who are depressed, for
instance, tend to express symptoms of irritability rather than
sadness and they may also fail to make expected weight gains
rather than lose weight, which is more frequently seen in
adulthood. On the other end of the age continuum, older
adults are more likely to experience loss of appetite, loss of
interest, and thoughts of death.
Culture
Despite the numerous studies on culture and depression that
have appeared in the literature, Chentsova-Dutton and Tsai
note that "it is sobering to realize how little we know about
depression across cultures" (p. 378). It is thus not possible
to do justice to all the issues involved in understanding depres-
sion across cultures, but we do address several different issues.
It seems clear that although humans are more alike than dif-
ferent, those differences that are cultural in nature play an
important role in depression. Culture can be broken down
within countries, so that in the United States, for example,
Latino, African-American, Asian-American, Native American,
and Caucasian groups may share a national identity but also
share experiences that are common to these cultural and ethnic
identities.
One important issue when considering depression and
culture is the nature of the depression idea itself and how it
might vary across cultures. That is, in some cultures, what
might reflect depression is not expressed in ways that are in
line with current diagnostic definitions. Hence, in some non-
Western cultures, there may be a dominance of the somatic
aspects of depression and a relative absence of the psychologi-
cal components of the disorder. Physiological and psychologi-
cal responses to stress may also differ for different groups.
Clearly, cultural variables are important in understanding
depression, but at this point, theory and research is far away
from specifying these variables in their conceptual, empirical,
and practical complexity.
Social and Environmental Correlates of Depression
These correlates refer to factors such as employment status,
socioeconomic status, and marital status. Consider findings
from the ECA study which found that people who were sepa-
rated or divorced had higher 1-year prevalence rates of major
depression (6.3%) than those who were currently married
(2.1%). The 1-year prevalence rate of major depression was
also higher for the unemployed versus those who were
employed (3.4% vs. 2.2%). Additionally, the ECA study
found higher rates of major depression among white-collar
workers and those with at least 12 years of education.
Although all of these variables are important, consider
marital status. According to the EPA study: "When current
marital status, regardless of past history is considered . . . sepa-
rated and divorced men and women have the highest rates"
of depression. Why is this the case? There are several possible
explanations, each of which may be true to at least some degree.
One possibility is that marriage provides a protective effect
against the stresses and strains that can precipitate depression;
having a confidant may help people deal more effectively with
the problems in their lives. Another possibility is that people
who are prone to depression may lack the stability to success-
fully navigate the vicissitudes of courtship and then marriage.
In this case, marriage may not have a causal relationship
with depression but may instead simply reflect a correlation
between two variables.
Theories of Depression
Stress and Depression
A variety of different theories of the causes of depression have
been proposed. Before examining these theories, it is impor-
tant to consider the role of stress in depression. Indeed, the link
between stress and depression has been recognized for centu-
ries. Research examining this link consistently finds a strong
relationship between the experience of stressful life events and
the onset of depression. Even though the link between stress
and depression is well documented, the precise nature of this
link is complex. As noted by Monroe and Hadjiyannakis,
although most models emphasize the role of stress in the
etiology of depression, it is also important to ask if stress
Depression 687
plays a role in the course of depression (e.g., degree and dura-
tion of symptoms). Another question concerns the nature
of the stress that is associated with depression; sometimes,
depression is associated with a major life event or several
major events; sometimes, it is associated with chronic minor
life events. Stress can consist of events that independently
'happen' to people, but data have also shown that depressed
individuals may play a role in generating stress that prolongs
depression and that may predispose to future depression.
However, not all individuals who experience significant
stress develop depression, nor do all individuals who develop
depression experience significant stress. Diathesis-stress ideas
can account for these facts. Diathesis refers to a predisposition
to an illness, and although not all models of depression
emphasize a diathesis, most contemporary theories do make
use of the idea to some degree. Stress and diathesis processes
may be closely intertwined in that depression, and depression
vulnerability, may affect stress responsivity; that is, psycholog-
ical or biological predispositions to depression may affect how
stress is perceived. Individuals who might have a predisposi-
tion to depression might therefore also have a predisposition
to either perceive or experience stress in an exaggerated way.
Furthermore, some approaches suggest an inverse relationship
between stress and diatheses, such that as the degree or severity
of the diathesis increases, less stress is necessary to initiate
depression. Alternatively, both severe stress and a strong diath-
esis may be necessary to provoke depression.
Psychological Theories of Depression
Cognitive approaches
Cognitive theories of depression are among the most widely
studied theories in the etiology of depression. One of the most
influential of these theories was proposed by Aaron Beck in
1967. Beck's model argues that depression results from the
activation of depressive self-schemas. These schemas refer to
organized mental structures that, in the case of depression,
are negatively toned representations of self-referent knowledge.
Moreover, schemas guide appraisals and interact with informa-
tion to influence selective attention, memory, and cognition.
Although all persons evidence schemas, the schemas of
depressed individuals are dysfunctional because they lead to
negative perspectives about oneself, the world, and the future,
or what Beck has termed, the negative cognitive triad.
An important aspect of Beck's model is that depressive
schemas lay dormant until activated by relevant stimuli:
"Whether he will ever become depressed depends on whether
the necessary conditions are present at a given time to activate
the depressive constellation." Thus, stressful life events are
necessary to activate negative schemas, and once activated,
schemas provide access to a complex system of negative per-
sonal themes that give rise to a corresponding pattern of nega-
tive information processing that eventuates in depression.
Learned helplessness and hopelessness theory
The helplessness theory of depression represents a cognitive
theory that evolved from an earlier emphasis on learned help-
lessness in depression. This work began with Seligman who
observed that animals which were unable to control negative
events often developed behavior that resembled depressive
symptoms. Based on these observations, Seligman developed
a theory of depression that focused on depressed individuals'
expectations that they were helpless to control aversive out-
comes. Even though much of the research on learned helpless-
ness was supportive of the basic tenants of the theory (e.g., that
depressed people tended to display more features of helpless-
ness than nondepressed people), other research highlighted
substantial shortcomings. In response to these shortcomings,
the theory was later reformulated as an attrihutional theory
which focused on how attributions about the causes of events
were linked to depression. In 1989, Abramson, Metalsky, and
Alloy further refined this theory into the hopelessness theory of
depression to suggest that hopelessness depression represented
a specific subtype of depression caused by the expectation that
highly desired outcomes will not occur, or that highly aversive
outcomes will occur, and the perception that no response can
change the likelihood of these outcomes. Both Beck's cognitive
model and the hopelessness model have received considerable
empirical support.
Interpersonal approaches
Interpersonal approaches to depression focus on the interplay
between a depressed person and his or her relations with
others. The underlying theoretical idea is that depression is
maintained by a vicious cycle that is caused by disruptions in
interpersonal interactions. For instance, many depressed indi-
viduals understandably seek out social support from others,
and initially, others are supportive. However, such support
does not alleviate the negative feelings for long, or at all, and
further support is sought. Rather than culling additional sup-
port, this intensified support seeking has the paradoxical effect
of pushing away those who have been supportive. That is, as
individuals begin to feel that their support capacity has been
exhausted, they pull back from the depressed person, leading
to an even further intensification of social support seeking, and
the further distancing of potentially supportive people.
Biologically Based Approaches to Depression
Although related in important ways, genetic approaches to
depression, biological approaches, and neuroscience perspec-
tives and data are examined here individually. More broadly,
both biological function and genetics are intricately tied to
brain structure and functioning. To facilitate understanding,
we discuss these variables separately, but it is important to keep
in mind that these, and other, variables are closely intertwined.
Additionally, given their central emphasis in much of the
literature, norepinephrine and serotonin are highlighted here,
but it is important to note that other biological variables (e.g.,
neuropeptides and cortisol) may also play a role in depression.
Genetic approaches
Genetic models propose that some of the variance in the onset
of depression, at least in some cases of depression, is linked to
genetic factors. Genetic models are generally informed by fam-
ily studies, twin studies, and adoption studies. Family studies
are based on the observation that depression runs in families.
However, because families share both genes and environ-
ments, such studies cannot separate genetic factors from envi-
ronmental factors. Twin studies, on the other hand, compare
688 Depression
concordance rates for monozygotic twins, who are genetically
identical, to dizygotic twins whose genes are similar but not
identical. Research shows that concordance rates for twins
are, depending on the study, approximately twice as high
concordance rates for dizygotic twins, thus suggesting a genetic
component to depression. Adoption studies are a rarely used
but powerful method for examining genetic effects. Adopted
children are genetically quite similar to their biological par-
ents, and hence evidence for a genetic role in depression is
found when adopted children with a biological family history
of depression are also more likely to experience a mood
disorder.
Taken together, these types of studies have provided some
estimates of the degree to which depression is influenced by
genes. In general, the heritability of milder forms of depression
ranges anywhere from 20% to 45%, with estimates for more
severe cases of depression likely to touch as high as 70%.
Heritability for recurrent or chronic cases of depression may
also be higher than those for single episodes. Therefore,
depending on the severity and type of depression, data suggest
that heritability of depression is anywhere from moderate
to high.
Neurotransmitters
Norepinephrine and serotonin
Data have shown that both catecholamine norepinephrine
(NE) and indoleamine serotonin (5-HT) are involved in
depression, and research shows that several possible areas of
pathophysiological processes in depression are linked to 5-HT
and NE. For example, low levels of a metabolite for NE have
been found for some patients, while low levels of a me-
tabolite for 5-HT have been found in some subgroups of
depressed individuals. One important source of data for the
role of 5-HT in depression comes from tryptophan depletion
studies. Tryptophan is a precursor to 5-HT and, when
depleted, diminishes the availability of 5-HT. Studies have
shown that depressed individuals who respond well to selec-
tive serotonin reuptake inhibiters (SSRIs) are more likely
to experience a return of symptoms with tryptophan deple-
tion. However, tryptophan depletion for depressed patients
not taking SSRIs does not appear to cause a worsening of
symptoms.
Research clearly shows evidence of biological dysregulation
in depression, but several qualifications are important. For
example, although a variety of biological processes are dis-
turbed in depression, these processes do not characterize all
cases of depression, nor are they necessarily specific to depres-
sion. Some biological processes appear to be stable in some
depressions, such as decreased 5-HT, but other processes
appear to be more characteristic of the depressed state and
tend to normalize in recovery. Moreover, there is evidence
that the subgroup of individuals who show evidence of
biological disturbances tend to be older and more severely
depressed. Moreover, rather than a cause of depression, it
may be that disrupted neurotransmitters serve as risk factors
for depression rather than as primary causes. Hence, one way
to think about disrupted biological function is within a
diathesis-stress framework, in that these variables may be
necessary for the onset of depression but are not sufficient
to bring about the disorder until they are triggered by stress.
Neuroscience approaches
Neuroscience focuses on the neural processes that are linked to
cognition and emotion. Although neuroscience uses a variety
of methods, the use of imaging techniques such as functional
magnetic resonance imaging (fMRI) has become increasingly
common. fMRI assesses the role and function of brain struc-
tures and regions during various tasks by creating images
caused by the oxygen in the blood flow to brain areas that are
active during these tasks. As such, fMRI can help to map the
brain circuitry that may be involved in depression.
Neuroscience approaches have tended to focus most atten-
tion on the prefrontal cortex, the anterior cingulate cortex, the
hippocampus, and the amygdala. Even though these anatomi-
cal structures serve numerous functions and are interconnected
in complex ways, several functions that may be particularly
involved in depression are highlighted. For example, depressed
individuals have been found to evidence left-side activation
deficits prefrontal; such deficits may reflect the disruption of
goal-directed activities that frequently characterize depression.
Some studies have also found that the children of depressed
mothers exhibit a similar left-side hypoactivation, suggesting
the possibility of a genetic link to this brain abnormality. Like
the prefrontal cortex, decreases in the activation of the anterior
cingulate cortex have also been found in some depressed indi-
viduals and are thought to be linked to social withdrawal and
anhedonia that is sometimes seen in depression. Other data
have shown reduced hippocampal volume in depressed
patients, suggesting the possibility of atrophy because the hip-
pocampus contains high levels of cortisol receptors; cortisol,
which is neurotoxic, leads to cell loss. This idea is in line with
some data showing that hippocampal atrophy is correlated
with the duration of depression.
The amygdala, which is another brain region that has been
linked to depression, has been shown to be hyperactivated
in depressed individuals, and has been hypothesized by some
investigators to be responsible for the maintenance of sad
mood states. Moreover, not only is the amygdala linked to
the preservation of sad mood, amygdala hyperactivation may
result from the failure of the prefrontal cortex to inhibit this
region. Hence, failure of the prefrontal cortex to inhibit the
amygdala may precipitate the maintenance of emotional infor-
mation processing in depression that perpetuates sad mood
and interferes with adaptive functioning.
Certainly, other areas of the brain are involved in depres-
sion, but the prefrontal cortex, anterior cingulate cortex, hip-
pocampus, and amygdale circuit appear to play a key role in at
least some forms of depression. It is important to note, how-
ever, that much is still unknown about the neural circuitry of
depression. For example, it is unclear if deficits in these areas
are linked to the cause of depression, might co-occur with
depression, or might arise as a result of depression. Moreover,
because these brain regions are densely connected, it is difficult
to determine whether one region is primarily linked to depres-
sion, while other regions are more secondary. Additionally,
some researchers hold out hope that patterns of brain abnor-
mality may themselves lead to different subtypes of depression.
See also: Anxiety Disorders; Cognitive Behavior Therapy; Personality
Disorders; Psychopathology: Diagnosis, Assessment, and
Classification; Suicide.
Depression 689
Further Reading
Abramson LY, Metalsky Gl, and Alloy LB (1989) Hopelessness depression:
A theory-based subtype of depression. Psychological Review 96: 358-372.
Beck AT (1967) Depression: Causes and Treatment. Philadelphia, PA: University of
Pennsylvania Press.
Chentsova-Dutten YE and Tsai JL (2009) Understanding depression across cultures.
In: Gotlib IH and Hammen CL (eds.) Handbook of Depression, 2nd edn.,
pp. 363-385. New York: Guilford Press.
Dobson K and Dozois D (2008) Risk Factors for Depression. New York: Elsevier
Science.
Eberhart NK and Hammen CL (2009) interpersonal predictors of stress generation.
Personality and Social Psychology Bulletin 35: 544-556.
Gotlib I and Hammen C (2008) Handbook of Depression, 2nd edn. New York:
Guilford Press.
Ingram RE (2009) International Encyclopedia of Depression. New York: Springer.
Kendall PC, Hollon SD, Beck AT, Hammen CL, and Ingram RE (1987) Issues and
recommendations regarding use of the beck depression inventory. Cognitive
Therapy and Research 11: 289-299.
Monroe SM and Reid MW (2009) Life stress and major depression. Current Directions
in Psychological Science 18: 68-72.
Seligman MEP (1975) Helplessness: On Depression, Development, and Death.
San Francisco, CA: Freeman.
Sheline Yl, Gad MJ, and Kraemer HC (2003) Untreated depression and hippocampal
volume loss. The American Journal of Psychiatry 160: 1516-1518.
Stein DJ, Kupfer DJ, and Schatzberg AF (2006) Textbook of Mood Disorders.
Washington, DC: American psychiatric Publishing.
Watson DW and Clark LA (1984) Negative affectivity: The disposition to experience
aversive emotional states. Psychological Bulletin 96: 465-490.
Relevant Websites
www.beckinstitute.org - Beck Institute of Cognitive Therapy.
www.depression.com - GlaxoSmithKline.
www.nimh.nih.gov/health/publications/depression - National Institute of Mental Health.
www.depression-screening.org - Self-assessment tool for depression.
www.webmd.com/depression - WebMD.
Depth Perception
M R Watson and J T Enns, University of British Columbia, Vancouver, BC, Canada
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cue The signal or information in a two-dimensional image
that can be used to infer or reconstruct the spatial relations
in a three-dimensional scene.
Depth Refers to the distance between a viewer and an object
as well to the distance between two objects that extend away
from a viewer. This is the missing dimension when depicting
a three-dimensional scene in a two-dimensional image,
whether the image be a picture or the pattern of light
projected onto the retina at the back of the eye.
Fovea An area of the retina in the center of each eye in
humans (as well as in the eyes of many other sighted
animals) containing the most densely packed array of
light-sensitive neurons. High-resolution vision is only
possible when light from an object falls onto the fovea.
Heuristic A method for solving a problem that begins by
first making certain assumptions (e.g., a gradation of
shading in an image that runs vertically from light to dark
corresponds to a surface with convex curvature, but only on
the assumption that light is shining from above the surface
and the viewer).
Image A two-dimensional projection of a three-
dimensional scene from a specific viewpoint.
Retina(e) The array of light-sensitive neurons at the
back of the eye. Light rays passing through the cornea
and lens of the eye are focused on the retina and these
retinal neurons convert the light into electrochemical
signals.
Retinal image The array of light projected onto the retina
through the cornea and lens of the eye.
Introduction
Depth perception is the ability of humans and other sighted
animals to see objects as having volume (as opposed to seeing
flat silhouettes) and to see the relative position of objects in a
three-dimensional environment (as opposed to in a two-
dimensional picture). This ability is crucial for everyday action.
Pushing an elevator button requires information about the
button's location relative to our eyes, which acquire the infor-
mation, and relative to our hand, which must make contact
with the button. Picking up a coffee mug requires knowledge of
the volume of the handle relative to the cup and to our hand.
Thus, depth perception enables interaction with our immedi-
ate surroundings. It also grants us information about objects
and surfaces that are more distant from us, enabling us to
consider future actions. In this article, we review how scientists
from several interrelated fields currently understand the prob-
lem of depth perception, concentrating on how it is accom-
plished by humans.
Depth perception is complex and still not completely
understood. It begins with the two-dimensional patterns of
light that are projected onto the retina, the array of light-
sensitive neurons at the back of each eye. When considered in
this way - as the problem of recovering the third dimension
from a two-dimensional image - depth perception is what
scientists refer to as an ill-formed problem. That is, there are
infinitely many arrangements of three-dimensional objects that
could lead to the two-dimensional array of light that is projected
onto the retina. To break this impasse, the visual system uses a
wide range of heuristics that are applied to the depth cues that it
considers. None of these problem-solving techniques is perfect,
but used in combination, they are able to hone in on the likely
three-dimensional source of the retinal images.
This article divides depth cues into two broad categories:
those cues that are only useful in determining depth for objects
relatively near the eyes and those that are reliable over a wide
range of distance, both near and far from the eyes. This is
an important distinction because humans have evolved a
specialized set of mechanisms for near distance vision, capita-
lizing on the fact that some of the biological apparatus for
seeing can alter the nature of the image before it reaches the
retina. This includes changing the shape of the lens at the front
of the eye to selectively focus the retinal image (accommoda-
tion), changing the relative gaze direction of the two eyes to
select one of many possible depth planes for sharp focus
(vergence), and integrating different images from the two
eyes to form a composite binocular image in the brain (stereo-
vision). It is probably not an accident that these structural cues
are useful when objects are a short distance from our bodies,
enabling us to use these cues when acting rapidly to intercept
or avoid objects. Among the cues that are more generally
useful, for both near and far vision, we have further distin-
guished between those that are based on changes in the retinal
image that occur over time (depth from motion) and the
so-called pictorial or monocular cues (here referred to as static
cues to depth).
Specialized Mechanisms for Near Distance
The depth cues that are restricted in their range of distance to
within about 3 m (10 ft) from the viewer are known as struc-
tural cues, because they involve alterations to the images on the
retina by physical changes occurring in a single eye or in the
anatomical relations between the two eyes.
Accommodation
The cornea and the lens at the front of each eye are both
curved, so as to focus rays of light on the retinal surface at the
690
Depth Perception 691
back of the eye. In order to form a sharp image at a given
distance, the curvature of the eye's lens is adjusted through
contraction of cilial muscles that surround the lens, with
greater curvature for near objects and reduced curvature for
distant ones. This process is called accommodation.
In healthy humans between the ages of 2 months and
45 years, the eye's lens is maximally curved when focusing on
objects about 20 cm away (8 in.), and is flattest when focusing
on objects at a distance of 3 m or more (10 ft). Neural feedback
from the cilial muscles signals the degree of curvature of the
lens, which is one indication of the current focal distance of
the eye, and hence of the distance of the object the eye is
currently focused on, which enables accommodation to be
used as a depth cue.
The lenses of newborns are curved to focus on objects about
20 cm (8 in.) away and cannot accommodate for objects at
other distances. However, by about 2 months, full accommo-
dation is possible. The material of the lens begins to become
more rigid by the age of 20 years and by the time most people
enter their mid-40s, the accommodative range of the lens has
been sharply reduced. This condition is called presbyopia and
is the cause of the widespread use of reading glasses and
graduated spectacles in people over the age of 45.
There is some controversy over the utility of this cue in
everyday situations because changes in accommodation are
rather slow and the neural signal from the cilial muscles is
limited in its resolution. A more likely way in which accom-
modation contributes to the determination of distance is
through the amount of blur that is caused in the image when
objects are out of focus. Indeed, observers in controlled experi-
ments are able to judge the relative distance of two spots of
light in a completely dark room using only one eye, presum-
ably because when one of the spots of light is in focus, the
other spot is slightly blurred.
Vergence
The retina contains a small region of tightly packed receptors at
the center known as the fovea. In order to see an object at high
resolution, the eye needs to be pointed directly at the object
such that light reflected from the object falls on the fovea. This
is known as fixating the object. Humans typically make three to
five fixations every second, moving their eyes to a new loca-
tions in order to sample up-to-date, high-quality information
about the scene they are viewing. Because the left and right eyes
are at different locations in the head, separated by 5-6 cm in an
adult, each eye must move differently to fixate the same object.
When the eyes move from fixating a distant object to a
closer one, they rotate slightly toward each other (conver-
gence). This increases the angle of convergence between the
two eyes, measured as the angle formed at the meeting of two
imaginary lines projecting from the pupil of each eye. When
the eyes move from fixating a close object to a more distant
one, they do the opposite, turning away from each other
(divergence). The overall degree of vergence (referring to both
convergence and divergence) can be estimated by feedback
from the extraocular muscles that control the position of the
two eyes, and as such, vergence is a potential cue for determin-
ing the absolute distance between the object currently being
fixated and the viewer.
The utility of vergence as a general cue to depth is limited,
however, by several factors. First, the eyes are already maxi-
mally diverged for objects more than 6 m from the viewer,
making it potentially useful only as a near distance cue. Sec-
ond, fairly common eye conditions such as strabismus and
amblyopia render vergence ineffective. Third, the signal from
the extraocular muscles is quite coarse. Yet, like all the other
depth cues, when vergence information is combined with that
from other cues, the precision of depth perception improves in
a synergistic way. Notably, vergence and accommodation
in concert yield more accurate judgments of the distance
from the viewer to a single point of light than either cue does
alone. Vergence can even be seen working in concert with other
depth cues when it is technically unnecessary. For example,
research has documented that viewers change their vergence
angle when merely viewing pictures. That is, even though all
objects in a picture are the same actual distance from the
viewer, viewers display greater convergence when looking
at objects depicted as closer in the scene than when viewing
objects depicted as farther away.
Stereovision
Because our eyes are 5-6 cm apart, the images projected onto
each retina are slightly different. This difference in images is a
cue to depth called binocular disparity, which enables the
experience of depth through the process of stereovision. This
process combines corresponding features in each retinal image
into a single representation that includes information about
distance from the viewer.
We are not normally aware that our eyes contain different
images of the same scene, but this can be easily demonstrated.
Hold the index finger of each hand in an upright position
directly in front of your nose, with one finger about 20 cm
away (9 in.) and the other finger about 40 cm away (18 in.).
Now focus your eyes on the more distant finger and take turns
closing and opening each eye. As you do this, the nearer finger
will seem to jump from one side of the farther finger to the
other. If you now open both eyes together you should see that
there are actually two images of the nearer finger. This is binoc-
ular disparity, which enables an accurate perception of depth.
The greater the horizontal distance between the corresponding
images of the same object in the two eyes (the two images of
the closer finger in this demonstration), the greater will be its
perceived distance from the object that is currently at the center
of the fovea in both eyes (the farther finger).
The positions of an object in the two retinal images are
systematically related to the distance of that object from the
object that is currently at the center of the two images in each
eye. In comparison to the rays of light that project from the
fixated object to the center of each retina, light from an
object that is closer to the viewer will fall slightly to the right
of center in the left eye, and to the left of center in the right eye
(this is called crossed disparity). Light from an object that is
farther away from the fixated object will do the opposite,
falling slightly to the left of center in the left eye, and to the
right of center in the right eye (uncrossed disparity). For any
object that is fixated, there is an imaginary region of space
encircling the viewer at the same distance, called Panum's
area. Objects at this distance have no binocular disparity,
692 Depth Perception
meaning that the rays of light projecting from them fall an
identical distance from the center of the retina in each eye. As
such, these objects also appear to be at the same distance from
the viewer as the object currently fixated. Objects outside this
region will appear to be nearer or farther, depending on whether
they produce crossed disparity (for nearer objects) or uncrossed
disparity (for farther objects) in the two eyes. Moreover, the size
of the disparity corresponds to an object's relative distance from
the fixated object. The process of stereovision, therefore, allows
the brain to infer the relative distance of objects on the basis
of both the sign (crossed or uncrossed) and the magnitude
(size) of the image disparities in the two eyes.
Stereovision can be exploited to create illusions of three-
dimensionality, such as seen in Victorian-era stereoscopes, the
popular twentieth century Viewmaster series of children's toys,
and the glasses worn by audience members at modern three-
dimensional films. Though the pictures used in such devices
always include depth cues other than binocular disparity, such
as occlusion, relative size, and shading (see section on Static
Image Cues), it is possible to create a compelling illusion of
depth using only changes in disparity, which means that
stereovision is a more powerful depth cue that the other struc-
tural cues. Bela Julesz invented random dot stereograms at Bell
Laboratories in the 1960s to demonstrate this. More recently,
the concepts used in making random dot stereograms have
been employed to generate the fascinating images known pop-
ularly as autostereograms or Magic Eye™ images.
As the name implies, a random dot stereogram appears
initially as nothing but a group of dots in a chaotic pattern.
However, some of the dots have actually been horizontally
displaced relative to one another, such that verging the eyes
either in front of or behind the depth of the picture allows
an illusion of depth to pop out. When the eyes are focused
to the correct distance, each eye's image of the dots is roughly
the same, yet some of the corresponding dots in each image are
displaced relative to each other. This binocular disparity gen-
erates the experience that a subset of the dot pattern has
popped into the foreground relative to other regions of the
dot pattern that now appear to be in the background.
In addition to demonstrating that stereovision can function
independently of other depth cues, random dot stereograms
also point to the complexity of the brain's stereovision
mechanisms. This is because in order to perceive depth in the
pattern of random dots, the brain must somehow know in
advance which dots in one retinal image correspond to the
same dots in the other retinal image. This is known as the
correspondence problem, and like many problems in human
vision, it is paradoxically both an ill-formed problem and yet
one that the brain seems to solve effortlessly. The fact that it is
ill-formed means that in the absence of any information other
than that contained in the dot patterns, there are an infinite
number of possible ways to align any two retinal images. The
fact that the brain solves the problem without effort is inter-
preted to mean that the brain must be using a priori assump-
tions about regularities in the environment to solve the problem.
A major challenge for vision researchers is to determine what
those a priori assumptions are. What is already clear is that the
process of stereovision comes to a conclusion more rapidly and
more reliably when it is informed by other depth cues, including
the monocular cues to depth reviewed later in this entry.
Human infants do not appear to possess functional stereo-
vision at birth, but it develops quite quickly. By the time
infants are 6 months of age, most will display stereovision
at essentially adult levels. Like the other physiological cues
(accommodation and vergence), stereovision is only effectively
useful within distances of about 3 m (10 ft) from the viewer.
Also, for some of the same reasons mentioned in the discus-
sion of vergence (e.g., conditions of strabismus, amblyopia),
between 5 and 10% of the general population does not have
usable stereovision because of imbalances in the nature and
quality of the information contained in the two eyes.
General Mechanisms for Near and Far Distance
Most of the cues for human depth perception are useful across
a wide range of distances. Among these are the cues that derive
from motion of objects in the scene and from motion of the
viewer, as well as the cues that derive from features in an image
seen only by one eye. Here we refer to these single-image cues
as static cues, in order to contrast them with the cues available
from analyzing motion, but readers should note that they are
also called monocular or pictorial cues, since they do not
require two eyes and are often used by visual artists.
Depth from Motion
The world is often in motion, giving rise to several rich sources of
depth information. For example, when a stationary viewer sees an
object in motion, there are systematic changes in what is visible
over time. At the leading edge of an object in motion, features of
the background that were previously visible will suddenly vanish
as the moving object occludes their view (called surface deletion),
while at the same time, other features of the background that were
previously invisible will suddenly appear (surface accretion). The
features of the moving object will remain constantly visible dur-
ing this time, unless the object is rotating, in which case the
systematic deletion and accretion of its features can be used as
cues to its three-dimensional shape. Surface deletion and accre-
tion from motion are therefore powerful cues for segregating
objects from backgrounds and for determining object volume.
These two cues alone are likely the most powerful comributors to
the compelling nature of motion pictures, which are not able to
benefit from any of the structural cues to depth we have already
reviewed, simply because all the information in a motion picture
is displayed at the same actual distance from the viewer.
When a viewer moves relative to stationary objects in a
scene, and the viewer's eye remains fixated on one object,
there are a number of cues to help indicate the relative distance
of the various objects in the scene from the viewer. These are
collectively referred to as the depth cues from motion parallax.
Relative to the object that is at the point of fixation, objects
nearer to the observer will move across the retina in the same
direction as the observer, while objects further away than the
fixated object will move in the opposite direction. Furthermore,
the closer an object is to the observer, the faster will its move-
ment be across the retinal image. These changes in the direction
and speed of the images of objects across the retina are very
effective guides to the relative positions of objects, although less
effective for judging the absolute distance of objects.
Depth Perception 693
Motion parallax is not only useful when a viewer is
actively and intentionally moving, such as when walking or
selectively moving one's head and eyes, but can also be used
to accurately judge the relative distance of objects when the
viewer is moving passively, as when sitting in a train or in a car
watching the world move relative to oneself. However, there
are important differences between motion parallax caused by
active and passive motion. Most importantly, passive viewers
do not generate any muscular feedback from their voluntary
actions to refine their passive reception of motion, so the
judgments of distance made in such cases are less accurate
than when motion is self-generated.
When an object rotates with respect to both its background
and a stationary viewer, there is a systematic pattern in the
motion of its various parts across the retinal image. These
motion patterns contain rich cues for determining the volume
of the object. For example, when a car turns toward us, the
regions of the retinal image associated with its various parts
will move at different speeds, with, for example, the headlight
farthest away from us moving at a faster rate than the closer
headlight. These differential rates of motion provide the brain
with rich information about the relative distance between
various object features, thereby allowing the perception of
three-dimensional shape to emerge from an analysis of motion
patterns alone. This object-relative motion cue was called kinetic
depth by Hans Wallach, who first systematically studied the
perception of three-dimensional shape that occurs when viewing
the shadows of moving objects on a screen, and more recently
has been referred to as structure-from-motion by researchers
studying the computational complexity of the problem.
When an object moves toward or away from an observer,
the size of the region it occupies on the retinal image also
increases and decreases proportionately, which is a depth cue
from motion known as looming. A particularly important
aspect of this depth cue is how symmetrical the changes on
each side of the image of the object are. Specifically, a looming
image that is expanding symmetrically and rapidly is consistent
with an object that will soon hit the viewer on the head.
In contrast, a looming image that is expanding asymmetrically
is consistent with an object that will move past the viewer
without a collision. Human vision is extremely sensitive to
looming cues, likely for good evolutionary reasons, and these
processes occur largely without awareness.
Research on the development of sensitivity to the entire
class of motion cues to depth indicates that these are among
the most primary. Newborn infants show sensitivity to the
surface accretion and deletion cues of relative motion, to motion
parallax, to structure-from-motion, and to looming. Indeed,
the newborn startle response to a looming image includes eye
blinking, cardiac slowing, and neck and limb stiffening, as
though humans are innately programmed to avoid collisions
with other objects.
Static Image Cues
Many of the static cues to depth available in static images can
be illustrated by considering the beach scene shown in
Figure 1.
Researchers have noted that one of the first simplifying
assumptions humans make when viewing such a scene is to
Figure 1 This beach scene from Santa Monica California depicts many
of the static cues to depth, including edge intersections, attached and
cast shadows, familiar size texture gradients, proximity to the horizon,
and aerial perspective. Specific examples of each cue depicted are given
in the text.
ignore unlikely possible arrangements of the world. One such
unlikely possibility in Figure 1 is that the beach umbrellas are
really very different sizes, and have been carefully cut out and
positioned into a mosaic pattern, such that they are all at the
same distance from the viewer. The visual system ignores this
possibility in favor of the more likely interpretation that the
umbrellas are all more or less the same size and shape, and
that they are positioned at various distances, so that the nearer
ones occupy a larger retinal size and occlude our view of
parts of the umbrellas that are farther away. This is known
as the generic viewpoint assumption, because we assume that
the image would not change drastically if our viewpoint on the
image were to change slightly. Note, in contrast, that even a
small change in viewpoint on this scene would quickly confirm
or deny the unlikely mosaic interpretation of the umbrellas.
So, in the absence of any evidence to the contrary, humans
tend to view scenes, including flat pictures, under the assump-
tion that a small change in viewpoint will not alter the spatial
layout of the scene. The generic viewpoint assumption is the
basis of a compelling and surprising form of visual art known
as anamorphic sidewalk or pavement drawings.
The static depth cues that must be considered in conjunc-
tion with the generic viewpoint assumption will be presented
here, for convenience, along a continuum from local to global.
By local, we mean that a cue can be considered quite reliably in
694 Depth Perception
isolation from other regions of the image; by global, we mean
that a cue is based on information distributed over a large
region of the image.
At the extreme local end of this continuum are the junctions
that occur when edges intersect with one another in an image.
Edge junctions contain some of the most reliable cues to rela-
tive depth in an image. For example, in Figure 1, the edges of
the large orange umbrella near the center terminate where they
meet the edges of the two umbrellas in front of it, forming
T-junctions. T-junctions are formed whenever one edge
occludes the view of a more distant edge. The edge in front is
continuous, whereas the occluded edge terminates at the inter-
section, making this a reliable local cue to depth. T-junctions
are generally considered to be a local cue for what is classically
known as the depth cue of occlusion or interposition.
L-junctions are also quite reliable cues to surface relationships
in the scene, but must be used in conjunction with a convexity
assumption, which is the assumption that, in the absence of any
other information, surfaces in the world are more often convex
(bulging out) than concave (indented). The convexity assump-
tion applied to L-junctions suggests that the side of the L with the
smaller angle belongs to the surface that is nearer to the viewer
than the side of the L with the larger angle. L-junctions can be
seen in all of the umbrellas in Figure 1, and invariably they occur
when an umbrella lies in front of a surface that is further away
from the viewer than the umbrella in question.
A second important class of local depth cues concerns the
three-edge intersections (i.e., Y- and arrow-junctions) that
occur when surfaces are joined together at corners. Corners
are extremely useful pieces of information about the three-
dimensional shape of objects, and tend to be interpreted in
the absence of other information as being convex rather than
concave. Note in Figure 1 how the three-dimensional shape
of the orange umbrella is revealed by the intersection of its
ribs. These edge intersections are ambiguous with regard to
convexity-concavity when considered in isolation, but in con-
junction with the convexity assumption, they provide rich
information about the umbrella shape.
Shadows also convey information about depth. Researchers
have found it important to distinguish between attached and
cast shadows. In general, the region of a uniformly colored
object that is relatively darker than other regions is said to be
an attached shadow, so called because this region of the object
is being blocked from receiving direct light by other parts of the
same object. Attached shadows are particularly useful for deter-
mining the three-dimensional shape of objects. For example,
in Figure 1, the shadows on the white T-shirt of the young man
standing near the front of the picture reveal the shape of his
torso. One critical ambiguity that must be resolved in order to
use this information to determine the three-dimensional shape
of the surfaces is the direction of the light source. All things
being equal, the visual system usually assumes that the light in
the scene is assumed to be shining from above, which means
that if the object has a convex surface, the region of attached
shadow lies toward its bottom. However, if the light was
actually shining from below, then the same pattern of image
shading would indicate a concavity, and the assumption
of light shining from above would lead us into error. The
interrelatedness of these assumptions in human depth percep-
tion can be vigorously exercised by viewing the hollow mask
illusion, which occurs when our assumptions of convexity,
light generally coming from above, and the familiar shape of
faces come into conflict with one another.
Cast shadows help resolve this problem. This can be seen in
Figure 1, where the shadows cast by the umbrellas indicate that
the sun is almost overhead. Cast shadows provide less infor-
mation than attached shadows do about the three-dimensional
shape of the object casting the shadow, because the shape of
the shadow is dependent on both the position of the light
source and the shape of the surface they are cast upon. How-
ever, they do provide a high-fidelity source of information
about the relative positions of the objects in the scene with
respect to the light source. The position of the inferred light
source can then be used to more accurately judge the shape of
objects on the basis of their attached shadows.
Cast shadows are also a rich source of information with
regard to relative object position. For example, if a cast shadow
borders directly on the object that is casting it, then the casting
object is very likely resting on the shadowed surface. On the
other hand, if there is a gap between the cast shadow and its
casting object, then the casting object must be farther from the
surface. The distance between an object and its cast shadow is
therefore a cue for their relative positions in space. If this cue
is combined with motion, it can be used to create powerful
illusions.
Another, somewhat more global, static cue to depth can be
derived from comparing the relative retinal size of the images
cast by similar objects. In general, it is safe to assume that when
two objects of a similar actual size differ in their distance from
the viewer, the nearer object will project a larger retinal image.
But note that the constraint of objects being of similar actual
size rests on its own assumption, namely that the viewer is
familiar with the objects and therefore knows that their sizes
are similar. When this assumption of familiar size is met, as it is
for our perception of the people in Figure 1, then we can
reliably see the people of smaller retinal size being farther
away in the depicted scene. When some of these interrelated
assumptions are violated, as occurs when two objects of similar
size are depicted in different apparent locations from the
viewer, then even familiar people can appear to vary greatly
in their perceived size, as occurs in the Ames Room illusion.
An even more global version of the familiar size cue to
depth occurs when linear perspective is used to infer relative
distance in a scene. Just as objects of the same size will have
smaller retinal sizes as they move further away from the
observer, so too will parallel edges in a scene converge toward
the horizon as their distance from the viewer is increased. The
point at which these lines converge, either within the frame of
an image or outside of it, is known as the vanishing point, and
corresponds to infinity as far as depth perception is concerned.
Note that these lines may be explicit, as occurs when depicting
architectural drawings, or they may be implicit, as occurs if we
look down a road bordered by trees of a similar height. Here
the imaginary lines connecting the tops of the trees converge
with the imaginary lines connecting the trunks to the ground at
the vanishing point. Linear perspective is the depth cue devel-
oped most vigorously in Western art made since the time of the
Renaissance until the invention of photography. It is especially
effective when used to depict the depth relations among carpen-
tered objects, presumably because these objects tend to have
Depth Perception 695
many parallel edges to calibrate the interpretation of depth.
When straight lines cannot be assumed in a carpentered scene,
then our tendency to interpret the scene as containing linear
perspective can go awry, leading to tantalizing ambiguities and
even outright impossibilities, as in the art of M. C. Escher.
When the depth cue of linear perspective is combined with
the depth cue of the relative retinal size of familiar objects, it
leads to a global depth cue called the texture gradient. A texture
refers to any collection of objects in an image, and a texture
gradient refers to the pattern of changes in the relative size and
spacing of these objects. For example, the beach umbrellas
depicted in Figure 1 form a semiregular pattern, becoming
smaller and closer together in the retinal image as they corre-
spond to actual umbrellas that are more distant from the viewer.
Texture gradients inform the viewer about depth through
gradual changes in image size and spacing, as seen in the beach
umbrellas in Figure 1, and the discontinuities in the gradient
are themselves informative because they indicate a sudden
change in the orientation of a surface relative to the observer.
For example, if we are in a room that has been tiled consistently
on the floor and wall, we can use discontinuities in the texture
gradients to determine changes in surface orientation. These
occur where the floor meets the wall and where one wall meets
another wall. In the famous study of human infant's sensitivity
to a visual cliff by Eleanor Gibson and Richard Walk, only a
texture gradient was used to signal to the infants that the
surface beneath the glass floor they were crawling on consisted
of surfaces at two different distances.
At the most global end of the continuum are several static
depth cues that require a comparative analysis of almost the
entire visual field. Proximity to the horizon (also known as
relative height or height in the plane) refers to the observation
that objects that are nearer to the horizon are generally seen as
being farther away from the viewer than are objects that are
farther away from the horizon. This means that below the
skyline, more distant objects tend to be higher in the retinal
image. Thus, in Figure 1, the people in the water are seen as
more distant than the man in the white T-shirt. Above the
horizon, these relationships are reversed, with nearer objects
in the scene being higher and more distant objects being closer.
Finally, aerial perspective is a global depth cue that is based
on relative differences in contrast and color rather than on size
or spatial position. When light travels through the air, it is
either scattered or absorbed by various particles. This means
that the light from distant obj ects travels through more air before
reaching our eyes than the light reflected from nearby objects,
with the consequence that less of a distant object's reflected light
reaches the eye. As such, distant objects will tend to be more
blurred than nearby obj ects ( relative contrast) , and they will tend
to be darker (relative brightness). They will also have a bluish
tinge (relative hue), because air molecules scatter shortwave blue
light to a greater extent than light of longer wavelengths, and
some of the blue light from the sun is scattered toward the viewer.
This can be seen in the hills at the top of Figure 1.
Not surprisingly, given the complexity and interrelatedness
of the static cues for depth, these cues tend to appear more
slowly in human development than the structural and motion-
based cues. Sensitivity to edge intersections develops relatively
early in life, with research suggesting that by 3-4 months
infants are responding to T-junctions in an adult-like manner.
However, it is not until 6-7 months that most infants can
respond reliably to linear perspective, texture gradients, famil-
iar size, or shading. The reliable use of cast shadows may not be
fully developed until 3 or more years of age.
Depth Perception in the Brain
There is currently much scientific interest in understanding
how the various cues for depth are processed and combined
by the neurons of the brain. The cue that has been studied most
thoroughly in this regard is binocular disparity, which, as dis-
cussed in an earlier section, is critical for the computation of
stereovision. Neurons that are tuned to specific binocular dis-
parities have been found in numerous regions of the visually
sensitive cortex of primates (including humans). Roughly
speaking, the further removed these neurons are from the eye
(by virtue of the number of synapses that must be crossed for
information to reach them), the more accurate and refined is
the stereovision they exhibit. Neurons in area VI, the first stage
of cortical visual processing, are sensitive to stimuli in the two
eyes that satisfy very loose criteria of similar disparity. Neurons
in the next cortical region, area V2, function under stricter
criteria, implying that more of the correspondence problem
has been solved at this point. In keeping with this trend of
increasing sophistication of processing as we move away from
the eyes, neurons in higher cortical regions are also specialized
to different kinds of disparity. For example, one class of neu-
rons may respond only to abrupt changes in depth, whereas
other neurons may register more continuously graded changes.
Conclusion
This review of depth perception in humans has emphasized
what is currently known about the way various physical
factors - both in the structures of the eye(s) and in features of
the image - are used by the brain to achieve the experience
of depth perception. The review organized depth cues into
two broad categories: those specialized for depth at near dis-
tances, where direct bodily interaction with other objects is
most likely, and those cues that work at any distance from
the eyes. The near cues included changing the shape of the
eye's lens (accommodation), changing the relative gaze direc-
tion of the two eyes (vergence), and integrating disparate
images in the two eyes (stereovision). Among the cues more
generally useful for both near and far vision, the review further
distinguished between those that involve changes in the retinal
image over time (depth from motion) and those that are
effective when they appear in a picture (static image cues).
See also: Visual Motion Perception; Visual Perception.
Further Reading
Braunstein ML (1976) Depth Perception Though Motion. New York: Academic Press.
Coren S, Ward LM, and Enns JT (2004) Sensation and Perception, 6th edn. New York:
Wiley.
Enns JT (2004) The Thinking Eye, the Seeing Brain. New York: WW Norton.
696 Depth Perception
Howard IP and Rogers B (2002) Seeing in Depth, vols. 1 & 2. New York: Oxford
University Press.
Julesz B (1971) Foundations of Cyclopean Perception. Oxford, England: University of
Chicago Press.
Parker AJ (2007) Binocular depth perception and the cerebral cortex. Nature Reviews
Neuroscience 8: 379-391 .
Wallach H and O'Connell DN (1953) The kinetic depth effect. Journal of Experimental
Psychology 45: 205-217.
Wolfe JM (ed.) (2007) Sensation and Perception. Sunderland: Sinauer.
Relevant Websites
http://hyperphysics.phy-astr.gsu.edu/HBASE/vision/accom.html - Accommodation.
http://psych.hanover.edu/krantz/art/aerial.html - Aerial Perspective.
http://psych.hanover.edu/krantz/MotionParallax/MotionParallax.html - Depth from Motion.
http://www.michaelbach.de/ot/mot_ske/index.html - Depth from Motion.
http://www.richardgregory.org/experiments/index.htm - Depth from Shading.
http://vision.psych.umn.edu/users/kersten/kersten-lab/demos/BalllnaBox.mov - Depth
from Shading.
http://en.wikipedia.org/wiki/Depth_perception - Depth Perception in General.
http://psych.hanover.edu/Krantz/art/cues.html - Depth Perception in General.
http://www.sinauer.com/wolfe/chap6/startF.htm - Depth Perception in General.
http://www.european-street-painting.com - Generic Viewpoint Assumption.
http://www.ski.org/CWTyler_lab/CWTyler/Artlnvestigations/PerspectiveHistory/
Perspective.BriefHistory.html - Linear Perspective.
http://psych.hanover.edu/KRANTZ/art/rel_hgt.html - Proximity to the Horizon.
http://www.psychologie.tu-dresden.de/i1/kaw/diversesMaterial/www.illusionworks.
com/html/ames_room.html - Retinal and Familiar Size.
www.mcescher.com - Static Image Cues.
http://www.yorku.ca/eye/disparit.htm - Stereovision.
http://cprr.org/Museum/Ephemera/Stereo-Viewers.html - Stereovision.
http://www.vmresource.com/ - Stereovision.
http://www.magiceye.com - Stereovision.
http://psych.hanover.edu/krantz/art/texture.html - Texture Gradients.
http://library.thinkquest.org/27066/depth/nlotherdepth.html - Vergence.
Developmental Psychopathology
A M Klahr and S A Burt, Michigan State University, East Lansing, Ml, USA
M Nikolas, University of Washington School of Medicine/Seattle Children’s Hospital, Seattle, WA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behavioral genetics A methodology for establishing the
genetic and environmental bases of individual differences in
observable behaviors or traits.
Equifinality Phenomenon in which different pathways may
lead to similar expressions of pathology, that is, different
initial states can lead to similar end states.
Etiology The causal processes and mechanisms underlying
the development of a behavior or trait.
Eleterotypic continuity Process when underlying
mechanisms remain constant while the actual type
or form of the manifested behavior changes with
development.
Elomotypic continuity Process when an observable
behavior remains constant while the underlying mechanism
causing the behavior changes with development.
Mediation Statistical model that seeks to explain the
relationship between an independent and a dependent
variable with the inclusion of a third explanatory
(or mediating) variable.
Molecular genetics Study of association and linkage
between DNA variants and observable behaviors and traits.
Multifinality Phenomenon in which similar initial
pathways may result in different forms of dysfunction; that
is, similar initial states lead to multiple possible end states.
Phenomenology Study of individuals' subjective
experiences of an event, that is, an individual's unique
experience of a psychological disorder.
Psychiatric comorbidity The co-occurrence of two or more
diagnosable psychological disorders within one individual.
Psychopathology Atypical and/or maladaptive behaviors
and mental/emotional processes.
Introduction
Psychopathology and its associated interpersonal, emotional,
and financial costs impact society at multiple levels and across
a variety of domains. Research investigating the causes, course,
and outcome of psychopathology has the potential to provide
enormous benefits to afflicted individuals, their families, and
society in general by informing innovative and specialized pre-
vention and treatment strategies for a variety of different disor-
ders. Developmental psychopathology has emerged as a leading
scientific framework for identifying the dynamic exchange pro-
cesses that influence both adaptive and maladaptive behavior.
The developmental psychopathology perspective represents
an integration of several distinct scientific disciplines, including
genetics, neuroscience, and psychiatry, as well as experimental,
clinical, and developmental psychology. Scientists working
within this framework aim to use these multiple levels of analy-
sis in order to uncover the mechanisms that give rise to and
maintain psychological disorders. This work is anchored on the
perspective that the processes underlying typical and atypical
development are mutually informative. That is, the continuity
between normality and pathology (or adaptive and maladaptive
behavior) as well continuity among the underlying causal
mechanisms is emphasized.
This article aims to provide an introduction to the develop-
mental psychopathology framework, including discussion of
its key concepts and principles, as well as potential challenges
to psychopathology research. Antisocial behavior is then dis-
cussed as an example, in order to further illustrate the ways in
which developmental psychopathology research is advancing
knowledge regarding the phenomenology, etiology, and treat-
ment of psychiatric disorders.
Key Concepts, Principles, and Challenges
The principles of developmental psychopathology have
become main stream in both the larger fields of psychiatry
and clinical psychology. At its core, developmental psychopa-
thology is the broad study of the processes involved in the
emergence and course of maladaptive behaviors. While there
are many features encompassing the developmental psychopa-
thology perspective, the key concepts and principles involve
(1) the approach to the study of causation; (2) the heterogene-
ity of cause, course, and outcome; and (3) the emphasis on
the trajectories and continuity between typical and atypical
development.
Causal Processes
Unlocking the etiological factors that give rise to psychopathol-
ogy undoubtedly offers great potential for advancing treatment
and prevention. Recognition of the influence of genetic factors
on psychological disorders and the innovation of behavioral
genetics has greatly contributed to these efforts. Behavioral
genetics, which relies on differences in genetic and environ-
mental relatedness among family members to estimate genetic
and environmental effects on a given condition, has been able
to quantify the magnitude of genetic and environmental influ-
ences on virtually all forms of psychopathology. Advances in
molecular genetics, specifically the mapping of the human
genome, has inspired work seeking to identify specific DNA
variants that may contribute to psychopathology. Importantly,
for both behavioral and molecular genetics, influences from
genetic and biological factors are typically conceptualized as
probabilistic (rather than deterministic).
697
698 Developmental Psychopathology
Consistent with the latter perspective, psychosocial and envi-
ronmental risk factors are also thought to be essential to the
development, course, and outcome of psychopathology. These
include main effects related to prenatal development, the family
environment, as well as neighborhood, community, ecological,
and sociocultural elements. In addition, there is growing
interest in understanding the exchanges between biological
and environmental factors that confer risk and/or protection
for psychopathology via studies of gene-environment interplay.
Examination of this interplay has been pursued specifically via
studies of gene-environment correlation (rGE) and gene x envi-
ronment (GxE) interaction processes - two related but distinct
mechanisms by which genetic and environmental factors influ-
ence one another. rGE refers to the extent to which an indivi-
dual's genetic makeup influences their exposure to particular
environmental factors. Alternatively, GxE interaction describes
genetic differences in sensitivity to environmental exposures,
such that certain environments only contribute to psychopa-
thology for individuals with particular genetic variations. Work
examining both rGE and GxE effects for psychopathology has
surged in recent years and positive findings have effectively
demonshated that genetic and environmental factors do not
operate in isolation from one another. This work has also has
reframed the nature-nurture debate in psychology and behav-
ioral sciences by adding weight to a growing consensus that
both genetic and environmental factors are important in
human development.
Heterogeneity
Two essential concepts, multifinality and equi finality, are key to
the principle of heterogeneity in developmental psychopathol-
ogy as it relates to causal mechanisms. Multifinality refers to the
notion that a single risk (or protective) factor can result in
multiple outcomes, whereas equifinality refers to the idea that
a single behavioral dimension or category of psychopathology
is influenced by multiple risk factors. Put another way, it is
unlikely that a single risk or protective factor will be the sole
necessary cause of one specific type of behavior or psychopa-
thology. This greatly complicates the task of researchers
attempting to understand the causes of psychopathology, and
yet it is critically important for untangling the various causal
pathways which can result in disorder across individuals.
In addition to heterogeneity in causal mechanisms, it is also
important to consider heterogeneity in terms of behavioral
presentation. The types of behavioral symptoms present at
any one time, as well as the onset, course, and responses to
treatment, often differ even within similar behavioral dimen-
sions and/or diagnostic categories. Thus, subtyping forms of
psychopathology in terms of symptom profile, onset, and
course has become one particularly common and useful
method for focusing on more etiologically homogenous sub-
groups within more multifaceted or multidimensional forms
of psychopathology.
Developmental Continuity and Discontinuity
The developmental psychopathology perspective also high-
lights issues of continuity and discontinuity between normality
and pathology. That is, developmental psychopathology is
focused upon understanding the dimensional aspects of
behavior (i.e., the mean and the extremes), as well as the
mechanisms that contribute to the etiology of behavioral con-
tinua. The primary notion here is that atypical or maladaptive
behavior cannot be understood outside of the context of typi-
cal development and that each is mutually informative. The
study of typical development has served to refine our under-
standing of psychopathology across many domains. For exam-
ple, the emergence of theory of mind (i.e., the ability to
attribute mental states to oneself and others) has been shown
to be an important milestone of cognitive development in
early childhood. Deficits in the development of theory of
mind are now widely recognized as one core feature of autism
spectrum disorders. In short, we cannot identify particular
behaviors as psychopathological if we are not yet clear on
which behaviors are normative at a given age. Importantly
however, while the underlying liability for psychopathology
may be conceptualized as dimensional, categorical decisions
are often required for clinical and practical purposes (e.g.,
diagnosis and treatment).
In addition, the developmental context for behavior is an
important consideration for the examination of psychopa-
thology. The relative frequency or commonality of behaviors
often changes with the developmental stage and what is typical
at one stage may be atypical (or maladaptive) at another.
Identifying specific trajectories of behavioral development
(both common and atypical) may then also aid in making
distinctions between normality and psychopathology. This
distinction is critical for identifying maladaptive behavior
itself, as well as for identifying the individuals that may be at
risk for the development of further problems versus those
whose atypical or problematic behavior may be more likely
to remit with age and development.
Challenges
While the developmental psychopathology framework has
provided numerous advantages for the examination of mal-
adaptive behavior, multiple challenges (both theoretical and
practical) remain. Put another way, the developmental psycho-
pathology framework has also aided researchers in identifying
specific issues that may introduce potential confounds or may
complicate the interpretation of findings within the field.
Measurement
Developmental psychopathology research often relies on
informant-based approaches on various measures (e.g., ques-
tionnaires, interviews, etc.) and is often limited by the extent to
which individuals are reliable reporters of their own and
others' behavior. The measurement of these behaviors is also
sensitive to the informant's own expectations and the under-
standing of 'problematic behavior,' as well as their exposure to
or familiarity with specific behaviors within a particular devel-
opmental context.
Heterotypic and homotypic continuity
Heterotypic continuity refers to the notion that the underlying
mechanisms or processes that contribute to psychopathology
may stay constant whereas the behavioral manifestation of
that liability may change with development. By contrast,
Developmental Psychopathology 699
homotypic continuity indicates that the behavioral manifesta-
tion may stay constant whereas the underlying mechanisms
and processes may change with development. Longitudinal
work that considers the developmental context and trajectories
of maladaptive behavior across time and development is thus
necessary for attempting to disentangle these processes. Such
work is difficult for researchers to conduct, however, and thus
heterotypic and homotypic continuity remain challenges for
developmental psychopathology research and theory.
Comorbidity
In addition to issues with causal continuity, the co-occurrence or
overlap between various (and presumably independent) forms
of psychopathology remains an important issue for develop-
mental psychopathology researchers. Maladaptive behavioral
dimensions tend to be correlated with one another and may
represent different manifestations of the same underlying
liability. Alternately, there may be specific types of comorbidity
patterns that represent etiologically distinct forms of psychopa-
thology. In fact, comorbidity among psychiauic disorders is
often the rule and not the exception and thus remains an
important consideration. Indeed, by complicating the clinical
picture, the presence of comorbidity can make it more difficult
to identify the etiology, course trajectory, and specific impair-
ments associated with any one particular disorder.
Sex differences
One final challenge related to developmental psychopathology
work involves sex differences in frequency and behavioral
expression. For example, autism and developmental delays
tend to be more common among males whereas depression
and anxiety tend to be more common among females. Simi-
larly, the behavioral expression of aggression looks quite dif-
ferent across sex: whereas as boys are more likely to engage
in physical aggression, girls are more likely to engage in social
aggression (e.g., gossip, name-calling). Questions remain
regarding whether these differences are due to sex or gender
differences in terms of behavioral expectations and develop-
mental context, sex differences in biological, cognitive, or social
development, or potential sex-specific etiological mechanisms
or pathways operating for various forms of psychopathology.
Onset versus course
Central to identification of the specific mechanisms involved
in psychopathology are some additional principles regarding
the potential ways in which causal processes are operating.
Among these is the notion that the biological and/or environ-
mental risk processes involved in the onset of a particular set
of behaviors may be distinct or partially distinct from those
contributing to the course or maintenance of maladaptive beha-
viors. Additionally, determination of the direction of environ-
mental/socialization effects remains the key (i.e., the extent to
which children influence their environments and vice versa).
Longitudinal designs are particularly useful for examining
these various distinctions and remain a popular and fruitful
investigative tool.
In all, these principles and challenges remain at the core of
developmental psychopathology research. We next turn to an
exemplar case - antisocial behavior - to highlight the strengths
of the developmental psychopathology approach.
Applied Example: Antisocial Behavior
Antisocial behavior represents one behavioral domain in
which the developmental psychopathology approach has
proven extremely useful. Antisocial behavior refers to a cluster
of problematic and disruptive aggressive and rule-breaking
behaviors, the core of which involves persistent violations of
societal norms and the rights of others (e.g., bullying, stealing).
Causal Processes
It is now well-established that there are significant genetic
influences on antisocial behavior, accounting for approxi-
mately 50-60% of the variance in the population. Genes may
also influence the types of environments individuals seek out
as they mature (e.g., antisocial adolescents may actively seek
out and chose to associate with delinquent peers, i.e., an exam-
ple of active rGE). Molecular genetics has also recently begun
to identify specific genetic variants that increase the likelihood
of developing antisocial behavior. To date, there are several
promising genes of interest, mainly related to dopamine and
serotonin neural transmission.
Additionally, research has also focused upon identifying the
ways in which an individual's genetic make-up may influence
their sensitivity to environmental risk factors, referred to as a
GxE interaction. Indeed, one of the first and most influential
studies of GxE focused upon examining these processes in
relation to antisocial behavior. Here, the effects of childhood
maltreatment on the development of antisocial behavior
varied, depending upon the form of a gene (specifically a
gene involved in dopamine neural transmission). That is, for
children with the low activity gene, the relationship between
maltreatment and antisocial behavior was strong. However, for
those with the high-activity form of the gene, the relationship
between maltreatment and antisocial behavior was signifi-
cantly attenuated. Thus, mistreatment may effectively act to
'turn on' genes of risk.
Researchers are also aware that environmental influences
are not necessarily independent of genetic risk. Indeed, yet
another line of work has focused upon identifying rGE pro-
cesses, where by genes exert their influence via nonrandom
exposure to particular environmental experiences. In the case
of antisocial behavior, biological parents both pass on any
genes of risk for the behaviors and may also create an aversive
environment through their own antisocial acts (i.e., a passive
rGE). In contrast, children who engage in antisocial behaviors
tend to evoke harsh, rejecting, and inconsistent parenting
practices (even when these parents are not biologically related
and thus do not share the genes for antisocial behavior,
referred to as an evocative rGE).
In sum, genetic and environmental risk and protective
factors are thought to influence the onset, development, and
maintenance of antisocial behavior through a variety of com-
plex and dynamic processes. These processes include both
environmental mediation of genetic risk (as illustrated
above) as well as genetic mediation of environmental influ-
ences. Work exploring this complex interplay between
genetic and environmental risk factors for antisocial behav-
ior will likely remain an active area of research in the years
to come.
700 Developmental Psychopathology
Heterogeneity
Researchers have identified many demographic, personologi-
cal, and psychosocial factors that are associated with antisocial
behavior. Child sex emerges as a particularly salient risk factor,
as boys and men are much more likely to display antisocial
behavior than are girls and women. Moreover, children who are
impulsive and sensation seeking, and who exhibit low levels of
prosocial behavior and high levels of emotional instability, are
at greater risk for developing antisocial behavior. A number of
environmental factors have also been related to antisocial
behavior, including prenatal complications, low birth weight,
maternal substance use during pregnancy, low socioeconomic
status, association with delinquent peers, and certain family
and neighborhood characteristics. Importantly, intervention
studies have also emerged as an important tool for examining
the etiological role of various environmental factors, as possible
causal mechanisms can be extrapolated from successful treat-
ment strategies (e.g., interventions which target the parent-
child relationship have been shown to be successful in treating
children with disruptive behavior problems). The multitude of
both genetic and environmental risk factors associated with
antisocial behavior can be interpreted as consistent with the
concept of equifmality - that different factors all contribute to
the same outcome.
That said, multifinality is also present, in that many of these
same factors are also predictive of other psychopathological
outcomes. For example, identified family and parenting char-
acteristics including discordant marital relationships, inade-
quate parental supervision, punitive disciplining strategies,
physical or sexual abuse, and neglect as well as neighborhood
risk factors (e.g., living in a high-crime area) are all predictive
of antisocial behavior as well as depression, anxiety, and inat-
tention and hyperactivity. In addition, antisocial behavior
rarely occurs independently of these other psychological pro-
blems, and is commonly comorbid with inattention and
hyperactivity.
Heterogeneity is also operating in terms of the types of
processes underpinning the development of antisocial behav-
ior. For example, some studies suggest that the impact of the
familial and contextual environment on the development of
antisocial behavior may be greater (and that of genetic influ-
ences lessened), when living in a poor, high-crime neighbor-
hood. In sum, research in antisocial behavior has shown
heterogeneity not only in terms of the types of risk factors but
also with the types of mechanisms that may underlie its
development.
Developmental Continuity and Discontinuity, Challenges,
and Future Directions
Like many other forms of psychopathology, understanding the
nature of antisocial behavior and its causes is dependent upon
both its relation to typical behavior and developmental context.
For example, physically aggressive behaviors are quite common
among toddlers and preschool children (more than 50% of
toddlers bite, kick, or hit). However, the behavior becomes
less developmentally appropriate as children age, and thus
behaviors we consider essentially normative (if unpleasant) in
3-year-olds would be markedly atypical in adolescents. By con-
trast, the less severe rule-breaking behaviors within the broader
construct of antisocial behavior (e.g., shoplifting, vandalism,
lying) occur relatively frequently among adolescents (and par-
ticularly adolescent males), but are rare during childhood.
Future research on antisocial behavior needs to consider these
issues of continuity and discontinuity.
Measurement of antisocial behavior is complicated by
many contextual factors. Often researchers rely upon parent
or teacher reports of a child's behavior, or self-report measures.
These measurement techniques are prone to many sources
of potential bias, including limited parental knowledge of
typical versus atypical child behavior, as well as influences of
informant personality and/or psychopathology, and their psy-
chological state at the time they fill out the report. For their
part, children may struggle to accurately reflect on their own
behavior, and respondents of all ages may not wish to disclose
antisocial acts (even when guaranteed anonymity). In an
attempt to circumvent some of these difficulties, researchers
often obtain data from multiple informants, including ob-
server ratings of videotaped behavior.
Regardless of the informant, however, questions also arise
regarding the demarcation between typical and atypical or
maladaptive behavior, as well as changes in the manifestation
of antisocial behavior with age. The context in which the
behavior occurs as well as the developmental timing of the
behavior is critical for making these distinctions (see above
example of physical aggression in 3-year-olds as compared to
adolescents). Furthermore, mechanisms underlying antisocial
behavior may stay constant, while the behavioral forms (e.g.,
physical aggression, rule-breaking behaviors) may change with
development, an example of heterotypic continuity. Alternately,
there may be homotypic continuity operating, such that the
types of behavior may stay relatively constant throughout the
lifespan, however, the processes and pathways influencing these
behaviors may change. In line with this, the processes that influ-
ence the onset of antisocial behavior (i.e., those that influence
aggression in toddlerhood) may be distinct or partially separable
from those that contribute to the continuation or maintenance
of these behaviors across development. These issues still need to
be resolved, and research in antisocial behavior is increasingly
relying upon longitudinal designs to capture and examine these
developmental complexities.
In all, the integration of a broad range of scientific disci-
plines within the developmental psychopathology framework
has aided in identifying the complex exchange processes that
contribute to both typical development as well as psychopa-
thology. The heterogeneity of causal processes and outcomes,
the multifinality and equifinality of risk and protective factors
involved in psychopathology, as well as comorbidity and mea-
surement issues create many challenges for the study of antiso-
cial behavior (and for psychopathology more broadly). Yet,
the recognition of these complexities has been critical to many
advances in our understanding of the onset and course of
maladaptive behavior. While much work remains to be done,
the developmental psychopathology perspective provides a
useful framework for both identifying causal mechanisms as
well as for applying this knowledge toward to the development
of effective prevention and treatment strategies.
Developmental Psychopathology 701
See also: Antisocial and Narcissistic Personality Disorder; Behavior
Genetics of Personality; Childhood Mental Disorders; Personality
Development; Psychopathology: Diagnosis, Assessment, and
Classification; Socioemotional Development; Temperament and
Individual Differences.
Further Reading
Beauchaine TP, Hinshaw SP, and Gatze-Kopp L (2008) Genetic and environmental
influences on behavior. In: Beauchaine TP and Hinshaw SP (eds.) Child and
Adolescent Psychopathology, pp. 58-92. Hoboken, NJ: Wiley.
Cicchetti D (2008) A multiple level analysis of perspective on research on development
and psychopathology. In: Beauchaine TP and Hinshaw SP (eds.) Child and
Adolescent Psychopathology, pp. 27-57. Hoboken, NJ: Wiley.
Hinshaw S (2008) Developmental psychopathology as a scientific discipline:
Relevance to behavioral and emotional disorders of childhood and adolescence.
In: Beauchaine TP and Hinshaw SP (eds.) Child and Adolescent Psychopathology,
pp. 3-26. Hoboken, NJ: Wiley.
Lahey BB, Moffitt TE, and Caspi A (eds.) (2003) Causes of Conduct Disorder and
Juvenile Delinquency. New York: Guilford Press.
Lewis M (2000) Toward a development of psychopathology: Models, definitions, and
prediction. In: Sameroff AJ and Lewis M (eds.) Handbook of Developmental
Psychopathology, 2nd edn., pp. 3-22. New York: Springer.
Quay HC and Hogan AE (eds.) (1999) Handbook of Disruptive Behavior Disorders.
New York: Kluwer Academic/Plenum Publishers.
Sameroff AJ (2000) Dialectical processes in developmental psychopathology.
In: Sameroff AJ, Lewis M, and Miller SM (eds.) Handbook of Developmental
Psychopathology, 2nd edn., pp. 23-40. New York: Springer.
Disgust
V Curtis, London School of Hygiene and Tropical Medicine, London, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Construct A hypothetical entity in brains that is assumed to
exist because of a regular pattern of activation and
behavioral response to particular cues.
Disgust A system in animal brains that orchestrates
behavior that serves to avoid infection.
Hygiene Behavior that serves to avoid sources of infection.
Parasite An organism living on or in another, dependent on
it for at least a part of its life cycle and causing damage to
the host.
Pathogen A microorganism living on or in another,
dependent on it for at least a part of its life cycle and causing
damage to the host.
Introduction
Most people have had experience with disgust. Confronted
with feces, vomit, dead animals, or bad-smelling food, we
react aversively, removing ourselves from contact with the
offending object, turning up our noses, saying 'Yuck!' and
experiencing unpleasant feelings akin to nausea. But what sort
of a thing is disgust? In common with many psychological
phenomena, disgust is a construct: a black box we assume to
exist in brains because encountering particular cues occasions a
regular pattern of response. However, disgust is a slippery con-
struct; the events and objects that occasion it are extremely
diverse, ranging from certain organic objects, animals, and peo-
ple, to unhygienic, transgressive, and immoral behavior. The
response is also highly diverse; it includes physiological changes
such as lowered heartbeat and galvanic skin response, activation
ofthebasal ganglia and insula, and a variety of facial expressions.
It also includes diverse behavioral responses, including laughter,
pointing out the object to others, distancing oneself from the
object, or approaching so as to remove or clean up the object of
revulsion. To a lay person, the one constant factor that is diag-
nostic of disgust is its accompanying affect - without the feeling
of disgust the disgust construct is not activated.
Disgust is thus something of a puzzle. Why should such a
wide variety of stimuli occasion this variety of responses and
such strong feelings? Though disgust has never been a particu-
larly popular area of scientific study, a variety of psychologists,
philosophers, and social scientists have tried to come up with
coherent explanations for this phenomenon. In this article,
I argue that these attempts at explanation fail, or are at best
incomplete, when they do not take into account the evolved
function of the disgust system. Taking a biological perspective,
I show how animals of all taxa, including humans, have had to
solve the problem of parasites: organisms smaller than them-
selves that attempt to make a living on, or in, their hosts.
Animals have evolved a variety of systems to reduce the fitness
costs of infection, including coherent self-repairing skin and
complex immune systems. They have also evolved behavioral
defenses to parasites, avoiding encountering and incorporating
them. This behavior is produced by the disgust system, which
serves to detect cues that parasites may be present, and to
orchestrate relevant protective behavior. Here I describe a selec-
tion of disease-avoidance behaviors in animals, and show the
continuity with human parasite-avoidance behavior. I also
explore the areas where human disgust has diverged from
that of other animals; where its use extends into the moral
domain. I argue that the current burgeoning of scientific work
on disgust would be more successful if it took better account of
disgust's evolutionary purposes.
Attempts at Explanation
Humans report that they are disgusted by a bewildering array
of objects and events. These include spoilt or unfamiliar foods,
bodily wastes, body contents, dead bodies, poor hygiene, certain
sexual activities, certain animals and insects, sick, deformed,
or poor people and those seen as outsiders, as well as dirty
environments. Actions that imply certain sorts of immorality,
including transgression, hypocrisy, and cruelty, are also com-
monly reported as disgusting.
There have been many attempts at understanding why all
of these things should be disgusting. Prescientific explanations
made disgusting things offences against God. Later attempts
at explanation in the philosophical, anthropological, huma-
nities, and psychological literatures have been almost as unsci-
entific. They made disgust part of the psychodynamics of the
repression of desire, or a product of the ordering of society,
or of magical folk thinking. Darwin began the scientific explo-
ration of disgust but sent scientists down a number of blind
alleys. The current standard approach in psychology explains
disgust using a complex composite of these approaches. Recent
evolutionary approaches offer a more rational basis for the
study of disgust, as I shall argue.
Ancient societies considered disgusting matter and behavior
as offenses against the divine. Early religious texts include
lists of the unclean such as the 12 impurities of the body
('Oily exudations, semen, blood, the fatty substance of the
brain, urine, feces, the mucus of the nose, ear-wax, phlegm,
tears, the rheum of the eyes, and sweat . . .') in the laws of
Manu, one of the sacred texts of Hinduism dating from about
200 BCE. Ancient Greek texts describe the pollutions of child-
birth, death, sex, and murder. The abominations of Leviticus
include body wastes and deviant sexual practices. Purification
rituals were codified to return the offender to a state of purity,
thus reducing their threat to society and to the divine order.
702
Disgust 703
Charles Darwin provided one of the earliest discussions of
the function of disgust. From linguistic evidence, he proposed
that disgust's original purpose was the avoidance of things
offensive to the taste. He asked leading questions of his corre-
spondents around the world and from this concluded that the
facial expression of disgust was universal in humans. He also
related anecdotes about disgusted turkeys and baboons.
Whilst Darwin emphasized the universality of the emotion,
Herbert Spencer was one of the earliest of a long line of
anthropologists to insist on the differences between what
different cultures found disgusting:
Here human flesh is abhorred, and there regarded as the greatest
delicacy; in this country roots are allowed to putrefy before they are
eaten, and in that the taint of decay produces disgust; the whale's
blubber which one race devours with avidity will in another by its
very odor produce nausea.
The social anthropologist Mary Douglas did most to
entrench the view that disgust is a product of culture. She
tackled the much-puzzled-over topic of the different food
taboos of different religions and from this extracted a broad
theory of culture. Following Durkheim and Mauss's idea that
classification is inherent in the organization of social life,
she decided that dirt is a by-product of order. Accordingly,
anomalous objects and events that do not fit the local cosmol-
ogy have to be rejected, and classed as dirty or impure, as
otherwise they would pose a threat to the social order. The
caste system in India is Douglas's paradigm case. Human
wastes are anomalous materials, neither alive nor dead, both
belonging to, but rejected by, the person. Hence wastes such as
feces are dirty and polluting. Humans who deal in wastes
become symbolically impure and the rejects of the social
system: the castes that henceforth have to be kept apart for
fear of social disorder [3], Douglas' approach is widely cited in
anthropological and cultural studies.
Freud had disgust serving a purpose different from that
proposed by Douglas. For him, disgust served the psychody-
namics of the repression of desire, defending the brain against
its baser tendencies. Children are taught to repress their urges,
keeping their animal id in check by learning to be disgusted.
Learning disgust of feces during toilet training represses any
temptation to want to contact or eat them. Disgust of the
bodies of others reduces the temptation to give in to lust at
every opportunity.
In 1927, Aurel Kolnai wrote the first scholarly work entirely
devoted to the subject of disgust. Using the phenomenological
approach current at that time, he provided a careful account of
the experienced emotion, painting a near complete picture of
disgust's domains. For him, biological disgusts include putre-
faction, excreta, secreta, dirt, animals, and insects - especially
crawling ones - certain foods, the human body in too close
contact, exaggerated fertility, disease, and deformity. His moral
disgusts include excess and satiety, lies, deceit and corruption,
moral weakness, and sentimentality. But Kolnai's explana-
tion was in the Freudian tradition. For him, disgust served a
balancing function, in encouraging the avoidance of excess,
surfeit, and a desire toward death.
As a practicing psychiatrist, Andres Angyal puzzled over
the things his patients found disgusting. He found a pattern
of disgust objects being "base or mean, capable of permanently
permeating anything they have contacted, leaving a lingering,
unpleasant after-sensation, even after washing the hands." For
Angyal, disgust objects often had a sense of the uncanny - they
may be cold or clammy and imply death. From a few anthro-
pological accounts he concluded that disgust is more universal
than culturally specific, and that it operates independent of any
knowledge about disease or microbes. However, the nearest he
came to an explanation of the patterns he described is that
wastes belong outside and it would be a perversion of nature
for them to be taken back in.
Today the most cited source for explanations about disgust
is the work of Paul Rozin and Jon Haidt. These authors use a
composite approach mixing Darwinian, cultural, and Freudian
ideas with folk magic. For them core disgust is centered on
all animals and their products, as potential foods. In their
most recent summary (in the Handbook of Emotions, 2008)
they suggest that disgust evolved out of distaste, and then
expanded to include reminders of our animal nature, certain
types of interpersonal contact, and moral violations. They
suggest that this core is based on a widespread fear of oral
incorporation of substances that might change one's nature,
based on the magical folk belief that 'you are what you eat.'
They explain the rejection of contaminated foodstuffs that
have come into contact with offensive objects or persons with
the law of sympathetic magic: 'once in contact, always in
contact,' and the law of similarity: when objects that look
like something disgusting then they are disgusting. For Rozin
and Haidt, cultural dietary and food preparation rules serve
to separate foods from their animal nature and help us not to
think too hard about the process that brought them to our
plate. Otherwise we would not be able to consume them.
These authors suggest that the fear of oral incorporation of
animal products has since spread beyond the mouth to the rest
of the body. Hence inappropriate sexual acts, poor hygiene,
death, and violations of the ideal body 'envelope' or exterior
form (e.g., gore, deformity, obesity) are also found disgusting.
However, now the concern is Freudian in origin, where disgust
represses a drive toward death. They postulate that humans
have to deal with the existential terror of knowing that they are
animals and, as such, are condemned to die. Disgust develops
to repress thoughts of mortality. Hence certain sexual acts and
envelope violations remind us of our animal nature, and poor
hygiene places us below the level of humanity. Disgust thus
protects both the body and the soul from pollution. They
describe a further category of interpersonal disgust as a response
to strangeness, disease, misfortune, and moral taint. They also
set out evidence that disgust may be elicited by certain moral
violations, especially in the domain of purity and sanctity, and
maintain the view that disgust is unique to humans.
A number of writers in the humanities have also explored
the phenomenon of disgust. William Ian Miller continues the
Freudian concern with mental balancing, concluding that
"ultimately the basis for all disgust is us - that we live and
die and that the process is a messy one emitting substances and
odors that make us doubt ourselves and fear our neighbours."
He also concurs with Kolnai that we are disgusted by that
which is too much alive. Robert Rawdon-Wilson explored
disgust in all its facets, and found it has many positive uses,
for example in humor that ridicules immorality, and in mark-
ing where social boundaries can and cannot be transcended.
704 Disgust
However, he ultimately labeled it 'the Hydra' because he found
it too complex to explain.
None of these approaches provide a satisfying scientific
solution for the puzzle of disgust. None can explain why such
a wide variety of objects and actions cue disgust, or why they
produce typical behavioral reactions of disgust. If we are
to understand disgust we need to look elsewhere; we need to
make reference to the process that designed brains and behav-
ior - evolution. We need to ask what adaptive purpose is served
by the disgust system in the brain.
A small, but increasing number of scientists, including
Dan Fessler, Debra Lieberman, Mark Schaller, Richard
Stevenson, and my group, are exploring disgust from an evolu-
tionary perspective. For us, there is a simple and parsimonious
solution to the puzzle of disgust: disgust systems evolved to
defend animals from attack by parasites, the tiny, usually invis-
ible, predators that attack by stealth and eat their hosts alive.
It orchestrates behavior that encourages animals to avoid
the sources of pathogens - whether in the environment,
in other animals, or the parasites themselves. That disgust in
humans is a product of evolution does not, of course, mean
that it does not vary from individual to individual, or from
culture to culture. Indeed evolution has designed such systems
to be responsive to local conditions. Individuals inherit differ-
ing propensities to disgust and individuals tune their disgust
responses over their lifetime according to experience and local
cultural rules. This is one of the reasons why previous work has
found disgust to be a hydra, having multifarious forms in
multiple settings. But disgust does have a single basic function:
it evolved to orchestrate the avoidance of parasites.
The Problem of Parasites
Parasites are everywhere. Viruses, bacteria, plasmodia, rickettsia,
and multiple species of worms and ectoparasites swarm
throughout, and over, all animals. The total biomass of para-
sites on planet earth is greater than that of predators. Parasites
cause damage to their hosts by consuming resources, releasing
toxins, manipulating their behavior, and spreading to kin and
community. Parasites thus exert a selection pressure on their
hosts - encouraging them to develop systems to minimize the
damage they suffer to their evolutionary fitness. All taxa have
immune systems that serve to track down, immobilize or kill
pathogenic parasites. All taxa that have been studied also dis-
play behavior that serves to minimize contact with parasites,
before they can infect the host. This has been dubbed 'the
behavioral immune system.' In humans (and, debatably, in
other animals), this pathogen-avoidance behavior is orche-
strated by the disgust system.
The disgust system has a difficult task; it has to recognize a
possible risk of infection without being able to detect tiny or
invisible parasites directly. One way of doing this is to avoid
close contact with conspecifics, especially those that show signs
of sickness. Another is to avoid other animals that may be
carrying parasites, or indeed, anything that looks like a para-
site. A third way is to avoid environments and objects where
parasites might be lurking.
Pathogen/parasite-avoidance behavior has been found
in the animal kingdom wherever it has been studied. Many
animals avoid others that show signs of sickness. So, for exam-
ple, the sociable Caribbean spiny lobster Panulirus argus,
refuses to share a den with other lobsters who are infected
with the lethal PaVl virus, which is transmitted by physical
contact. Bullfrogs, Rana catesbeiana, suffer from a yeast infec-
tion that reduces their mobility and can kill them. Given the
choice, healthy tadpoles avoid going anywhere near those
that have the infection. Similarly, when experimenters injected
Killifish, Fundulus diaphanous, with black inkspots to mimic
the effects of a common parasite, other killifish avoided shoal-
ing with them. Social primate species will generally only accept
a new member into the troop after a long period of quarantine,
which includes stressful attacks to expose the state of health of
the candidate member. Candidates showing signs of sickness
remain marginalized.
Like the animals that avoid sick conspecifics, humans also
avoid others that show signs of sickness, and we also find signs
of sickness in others disgusting. Figure 1 shows a photograph
of a man's face which we manipulated to look ill by painting
on spots, and mimicking signs of fever. This was scored as twice
as disgusting as the same individual in a healthy state by over
40 000 respondents in a web survey. Figure 2 shows an
infected wound, which was scored as 27% more disgusting as
a clean burn in the same experiment. Figure 3 shows that an
empty underground train was scored as much less disgusting
than one that was full of people. Few humans are prepared
to come into close contact with and share bodily fluids with
others, whether they are healthy or not, except in particular
overriding circumstances, such as for pair bonding or child
Figure 1 Average disgust score of two facial images.
Figure 2 Average disgust score of two wounds.
Figure 3 Average disgust score of two images of an underground train.
Disgust 705
care. In a more recent wide-ranging web survey, we found that
signs of infection on genitals and poor genital hygiene were
amongst the strongest elicitors of disgust.
Animals can, in some cases, detect and avoid or remove
parasites directly. C. elegans, the tiny and much-studied nema-
tode worm which has only 302 neurons to its name, can tell
the difference between pathogenic bacteria, from which it flees,
and prey bacteria, on which it feeds. Fish detect and swim away
from the cercariae of eye flukes that can cause blindness. Ani-
mals also expend a lot of precious energy on removing damag-
ing ectoparasites. Cattle swish their tails to drive away biting
tsetse flies, fish scrape themselves to remove fish lice, vampire
bats scratch to remove batflies, and impala use their teeth as
tick combs. If impala or doves are experimentally prevented
from grooming themselves, lice and ticks multiply. Over 250
species of bird are known to 'ant'; rubbing crushed insects
such as ants or millipedes over their plumage. This behavior
distributes compounds such as formic acid that protect from
parasitic bacteria, fungi, and arthropods Chimpanzees have
been seen engaged in penile hygiene-wiping themselves off
with leaves after mating. Some animals will pay highly for a
groom to rid themselves of parasites; cleaner fish remove ecto-
parasites but also bite off bits of healthy flesh, and a study of
long-tailed Macaques, Macaca fascicularis, revealed a biological
market system where males paid for sex at the going rate in the
currency of time spent grooming.
Humans also expend considerable effort and funds in
grooming themselves and others to remove ectoparasites
such as lice and scabies mites. Figure 4 shows a head louse,
which was found significantly more disgusting than a wasp in
the photographic comparison. Humans recognize and avoid
parasitic species of worm and find them disgusting. Figure 5
shows Ascaris worms which were scored as more disgusting
than caterpillars. Humans are further disgusted by species of
animals that can vector parasites such as flies, cockroaches,
rats, bats, snails, and pigs.
Animals also have to be careful not to come into contact
with places where parasites might be lurking. When Daphnia
are forced by fear of predation to swim near the muddy bottom
Figure 4 Average disgust score of two insect images.
of a tank, they acquire more parasites. Ants avoid making nests
in the vicinity of ant carcasses; birds abandon nests that are
infested with parasites. Many insects have tricks such as 'frass-
flinging' or 'turd-hurling' to keep their environment free of
feces, penguins eject projectile poo, and tapirs, badgers, and
blind mole rats use latrine sites. Sheep avoid feeding on pasture
that is heavily dunged, and it has been suggested that the
reason why many herbivores, such as reindeer and caribou,
migrate is to avoid parasite build-up in calving grounds. Anec-
dotal evidence suggests that human nomads also move on
when camping grounds get too insanitary.
Like other animals, humans avoid substances in the envi-
ronment that might harbor parasites. In our web survey, urine
and excrement, as well as sticky door handles and filthy buses,
were rated highly disgusting. Figure 6 shows that a plate of
food looking like bodily wastes scored more highly than a plate
of what looked like inorganic matter.
Humans have a further ability, that it seems is not found in
other animals: to track contamination. When an object of disgust
comes into contact with a previously nondisgusting object, the
second object can become disgusting. So for example tooth
brushes, eating implements, or bedclothes known to have been
used by strangers tend to be rejected. This response is what Steven
Pinker has called 'instinctive microbiology.' Objects that harbor
microbes can deposit microbes on the surface of other objects;
hence these others become as disgusting as the original source.
Since the problem of parasites is common to all animals,
including humans, it is not surprising to find that the basic
avoidance behavior of humans is similar to that of other ani-
mals. Like other animals, humans avoid close contact with
most others of the same species, especially if they are sick;
they avoid other species that are parasites or parasite vectors;
and they avoid environments and objects where parasites and
pathogens are likely to be encountered. This gives us a parsi-
monious explanation for why humans should have a disgust
system. From this perspective, then, disgust is the construct in
the brain that takes cues of potential parasite presence and then
orchestrates the behavior that serves to avoid them.
Parasite detection and avoidance was a hard problem to
solve for a number of reasons. Apart from the problem posed
by their size, there is also the problem of their indirect effect.
While animals can learn from painful experience with preda-
tors, sickness does not usually immediately follow pathogen
contact, so it is hard to learn what to avoid from experience of
what has caused illness. A further problem animals have to
solve is that it can be costly to avoid parasites. Refusing to
consume rotten food may cause one to miss an opportunity
for a scarce meal, and avoiding a sick person may cause one to
miss a mating opportunity, for example. Over evolutionary
706 Disgust
history, the genes of those animals which, on average, did best
at solving these parasite-posed problems did better at passing
on their genes.
Have we now solved the puzzle of disgust by invoking the
disease avoidance function? Well, not quite. It is easy to see
why the main disgust elicitors should be avoided: spoilt foods,
bodily wastes, body contents, dead bodies, poor hygiene,
certain sexual activities, certain animals and insects, sick,
deformed, or poor people and those seen as outsiders, as well
as dirty environments can all harbor pathogens and hence pose
a risk of infectious disease. Objects that have contacted any of
the above may have also been contaminated with microbes;
hence the disgust associated with contagion. Questions, how-
ever, remain about why unfamiliar foods should elicit disgust
and also about the evolutionary purposes of moral disgust.
Disgust of Unfamiliar Food
Humans are omnivores who live in a huge range of ecological
zones. In principle, humans could eat almost any living thing;
however, given the potential for foods, especially those of
animal origin, to contain pathogens and parasites, it makes
sense to be neophobic. Since the current social group is still
alive, it is probably a good rule of thumb to eat what others in
the current environment are eating. This is, in fact, what hap-
pens in humans (similar patterns are observed in another
omnivore, the rat). Children are prepared to eat a huge variety
of foods that are fed to them by their family until the age of
about 2 years. Then an extreme reluctance to try new foods sets
in. This helps explain why food rules differ between cultures
and why disgust is often evinced by people offered unfamiliar
types of food, especially those of animal origin. Though dis-
gust's core concern with infection prevention is constant, the
way in which it is expressed in individuals is developmentally,
and adaptively, tuned.
The final piece of the disgust puzzle is why certain sorts of
immorality - including transgression, hypocrisy, and cruelty -
are also commonly reported as disgusting.
Moral Disgust
We recently asked 77 UK-based adolescents to list what they
found morally disgusting. Top ten on their lists came rape,
racism, murder, torture, bullying, pedophilia, discrimination,
necrophilia, genocide, and exploitation. High on the list also
came thieving, abuse of trust, and corruption. Two concerns
can be distinguished here. One is moral outrage concerning
taking unfair advantage, especially in the maltreatment of the
weak. Some of these outrages also involve exposure to bodily
fluids and so are doubly disgusting. While the latter concern
has already been explained, why is it that unfairness, cruelty,
and exploitation are found to be morally disgusting?
One explanation that has been offered is that the word
disgust is being used as a metaphorical sense. The disgust of
moral offenses is like that felt for infective organic material but
is not the same. People use the word disgust for a feeling of
revulsion for moral violations. But this is not, in fact disgust; it
is more akin to anger. However, a recent series of fMRI studies
have suggested that there is considerable similarity in the acti-
vation patterns of moral and organic disgust. People offered
unfair deals in an ultimatum game, for example, show activity
in the anterior insula, a locus that is characteristic of disgust.
The same facial muscles are involved in the expression of both
moral and organic disgust. More recently, a series of studies
have shown interactions between the two; priming with an
organic disgust, or sitting beside a smelly waste bin makes
moral judgment more severe, while hand washing can assuage
feelings of moral guilt.
It thus seems likely that while the proper domain of disgust
is the avoidance of the likely sources of infection, it has evolved
an extended domain in humans that causes aversion to mor-
ally reprehensible behavior. One explanation may be that
instead of responding to the likely presence of a parasitic
organism that threatens bodily resources, the brain is respond-
ing to cues that an individual is behaving like a parasite, using
unfair means to take more than his or her fair share of
social resources. Racists gain unfair advantage by denigrating
others; pedophiles, rapists, and bullies take advantage of the
weak, while necrophiliacs abuse bodies that cannot defend
themselves. Revulsion from such parasitic behavior leads to
shunning of the perpetrators, and their exclusion from the
benefits of social life. The threat of punishment by exclusion
is a powerful one, and one that helps keep individuals behav-
ing cooperatively. Disgust thus serves to police the boundaries
of society, keeping social parasites out, in a way that is func-
tionally equivalent to the purpose it serves in keeping parasitic
organisms out of the host body.
While this explanation fits the pattern of moral disgusts, by
what mechanism could disgust of the organic sources of infec-
tion have been exapted to extend to revulsion for the social
parasite? Here we can turn again to animal models for some
insight. While few animals are as ultrasocial as humans, the
eusocial insects including ants, termites, wasps, and bees live
colonial and cooperative lives. Like humans, insects have to deal
with threats from parasitic species, which threaten them as indi-
viduals and as a colony. Ants, for example, clean food of parasitic
fungal spores; bees remove sick individuals from the nest and
prevent sick individuals entering; termites warn of parasites
through a special dance. Further, individuals of the same species
can act as parasites. In the social Hymenoptera, including most
ants, Apis honeybees, Meliponinae stingless bees, and Vespinae
wasps, workers retain ovaries even though they cannot mate.
These workers can lay unfertilized, male-destined eggs and
thereby challenge the reproductive primacy of the queen. How-
ever, other conspecifics do not tolerate this selfish behavior and
eat the eggs of the rebel. Another example is caste fate policing.
In many bee species, individuals that do not accept their fate to
become a worker and try to develop into a supernumerary
queen are discriminated against or killed by conspecifics.
The punishment of cheaters is one of the mechanisms that
allow societies to keep cooperating. In humans, punishment
by disgust - the avoidance and shunning of social cheaters - is a
powerful disincentive to selfish behavior. Social parasites such
as rapists and thieves are subjected to public disgust and revul-
sion when they are exposed. Indeed, such public contempt is
not always required. We humans police ourselves to great extent,
being disgusted with ourselves for selfish behavior. We fear
the embarrassment and shame we may feel if we behave badly,
Disgust 707
for example subjecting others to our own disgusting bodily
emanations, or subjecting others to our own dishonest or greedy
parasitic tendencies. Having the human ability to imagine the
disgust of others leads to plans being made to behave coopera-
tively, so as to avoid shame and embarrassment at all costs.
Explaining Disgust
I have argued in this article that the explanation for the con-
struct of disgust in the human brain is the evolutionary need to
avoid damage from parasitic infection. This need is something
we have in common with all animals. I have also argued that,
beyond its proper domain of parasite avoidance, humans have
extended the function of disgust to apply to moral revulsion at
social parasitism.
This Darwinian explanation differs from those offered pre-
viously by anthropologists, psychoanalysts, and psychologists.
It is even somewhat at odds with Darwin himself.
While Darwin was surely right that the expression of
disgust is near universal in humans and that disgust can be
observed in animals, he was incorrect to suggest that disgust
originated in distaste. Parasites have many routes of access to
the resources of a host body and humans, like all of their
animal ancestors, must police those portals of entry, by remov-
ing ectoparasites from the skin, avoiding contact with diseased
sexual organs, or avoiding breathing in the exhaled droplets of
diseased conspecifics. Avoiding consuming spoilt or contami-
nated food is only one amongst many of the strategies animals
use to avoid parasites. Humans must have inherited all of
these multiple aversions from their animal ancestors. It is
thus highly unlikely that human disgust originated in human
distaste, which then later diverged into a variety of other
aversions, as Rozin and Haidt, following Darwin, have argued.
However, does it make any sense to suggest that animals
can be disgusted, as Darwin proposed and Rozin and Haidt
refute? The answer to this question depends entirely on how
one defines disgust. If it is defined, as I propose, as a system in
brains that recognizes cues to infectious diseases and orches-
trates appropriate avoidance responses, then disgust is ubiqui-
tous in the animal kingdom. The nematode worm, C. elegans,
fleeing parasitic microbes in its petri-dish, is exhibiting disgust-
driven behavior. However, if we define disgust as an emotion
and if having an emotion depends on being able to report its
characteristic qualia of feelings, then disgust must be confined
to humans, since only they can report feelings. In my view, it is
more helpful to see feelings as an epiphenomenon of such
motivation systems. Feelings are used in the theatre of the
conscious brain to learn the consequences of behavior and to
imagine its future consequences. Knowing that one is likely to
feel disgusted by rotten food tomorrow encourages one to store
it correctly today. Knowing that one may step in feces tomor-
row encourages one to save to build a latrine next year. Know-
ing that one might meet, and be disgusted by, sick people helps
to encourage one to stay at home in times of plague. While the
feeling of disgust is probably confined to the small number of
higher animals with conscious imaginations, all animals have
disgust systems.
For Rozin and Haidt, the paradigmatic source of core
disgust in humans is the consumption of animals and their
products. From an evolutionary perspective, however, animals
are just one likely source of parasites, amongst others. Other
humans harbor more human pathogens and hence should be
more central to the disgust system than other animals (canni-
balism is especially disgusting, for example). Animals should
be disgusting inasmuch as they pose a risk of transmitting
infection. So pigs share more parasites with humans than
herbivores such as sheep, and hence pig meat is more likely
to be treated with disgust than lamb, for example. Animal
products such as fish and meat are more risky to consume
than vegetable matter, especially if spoilt. Interestingly, it has
been hypothesized that microbes may have evolved to exploit
our aversion to rotten meat by producing large amounts of
volatile warning chemicals. These heighten our disgust, thus
allowing the smell-emitting microbes to feast on the nutrient-
rich material that we have rejected. Our disgust systems
may thus be being fooled into overreacting to gone-off animal
foods. The avoidance of animal materials as one source of
pathogens provides a more parsimonious explanation for
their rejection than Rozin's invocation of magical thinking
such as you are what you eat.' Indeed it is clear that 'animal
nature' is not a key issue since not all animals disgust us
and humans will sometimes compare animals favorably with
themselves. Further, Rozin and Haidt's suggestion that people
distance themselves from meat preparation so as not to be too
disgusted to eat it may be true for modern American societies
but does not hold in most developing countries, and was
probably untrue for most of recorded history, where families
had no choice but to butcher their own produce.
Rozin and Haidt propose a magical 'law of similarity' for
why children will refuse to eat fudge shaped like feces, despite
the children knowing that it was fudge. Here, again, no magical
thinking is needed to explain this phenomenon; in the envir-
onments in which our brains evolved, things looking like feces
simply were feces. The instinctive reaction cannot simply be
overridden by the experimenter's reassurance as to edibility.
Similarly the law of contagion - 'once in contact, always in
contact' - is better explained by 'instinctive microbiology' than
magical beliefs. Disgusting objects are disgusting because they
may contain invisible pathogens. These microbes can be trans-
ferred by simple contact. Hence, it is not surprising that those
of our ancestors who took note of objects with a recent history
of contact with disgusting objects thrived better than those
who did not. Though the same argument should apply to
animal behavior, we know of no experiments that have tried
to ascertain whether animals record, remember, and avoid
objects that have become contaminated in this way.
Freud argued, as do Rozin and Haidt, that disgust is part of
an internal psychodynamic regulation system, where the
dysfunction of one part of the brain is relieved by the func-
tioning of another. For Freud inappropriate appetites and mal-
adaptive lust after sex are countered by learning to be disgusted
by bodily wastes such as feces and sexual fluids. Evolutionary
psychologists recognize that there are conflicts internal to brains
but do not make one system regulate another. They rather
recognize such internal conflicts as trade-offs. Animals have to
constantly balance a range of evolutionary needs. People who
are hungry, for example, have to balance their need for food
with the danger of eating unsafe foods, and will lower their
disgust thresholds to expand the range of what they can eat.
708 Disgust
Rozin and Haidt follow Freud in positing a psychodynamic
use for disgust in controlling our own reactions to the fact of
our mortality. For Rozin and Haidt, disgust helps to keep us
from dwelling on the terrifying truth of our own animality and
of our future annihilation. Fessler and Navarette have challenged
their use of the so-called Terror Management Theory (TMT),
showing that death disgust is negatively correlated with age,
when TMT predicts the opposite, and that priming thoughts of
mortality do not increase disgust. They argue, as do we, that
disgust is better seen as an adaptation for disease avoidance.
The approach to disgust in the social sciences tends to
follow the work of Mary Douglas, who held that we categorize
items as disgusting when they are 'out of place,' when they are
anomalous and do not fit the local cosmological system, as we
saw. For Douglas, societies came first, and disgust is one of
their products. Yet it cannot be true that disgust is a human
cultural production, since as we have seen, disgust predates
humans and certainly predates societies. Douglas does, how-
ever, point us toward some of the symbolic uses of disgust in
society where, for example, the caste system is extrapolated
from the principle that those with contaminating jobs should
be kept apart from the rest. Street sweeping, house cleaning,
and waste collection remain low-status jobs in most countries
today. Douglas's work on the symbolic and ritual uses of
disgust in making social separations between pure and pol-
luted, in group and out group remains deeply insightful. How-
ever, from the Darwinian perspective, disgust is not purely a
product of culture but a factor that shapes culture - disgust
came first. Cultures may have elaborate rules about purity and
pollution, but these tend to be based on our organic disgusts.
I have argued that the work of Darwin, Freud, Douglas,
Rozin and Haidt, and their followers have sent disgust studies
down a number of blind alleys. Disgust is not unique to
humans, it does not originate in distaste, it is not a product
of magical thinking, nor of the cultural constructions of societies,
neither did it arise to psychodynamically balance inappropriate
appetites or fears of death. There is one simple explanation for
why humans and all animals avoid others, certain other animals,
and contaminated environments, and that is the threat of infec-
tious disease. Humans have extended the domain of disgust, using
it to police and punish social parasites, hence reinforcing the
cooperative behavior that makes social life possible. This expla-
nation for disgust should be preferred on grounds of parsimony;
it provides a simple, single explanation for all of the diversity of
disgust-related phenomena we have been considering.
Disgust and Its Uses
Does it matter in the end whether disgust is an adaptation
to the threat of parasites as I contend? Since disgust has tenta-
cles that go deeply into the human psyche, influencing our
everyday hygiene and our everyday aversions, as well as being
one of the sources of both prejudice and moral behavior,
understanding it fully must be an important task for life scien-
tists. Further, disgust may hold the key to improving our ability
to fight infectious disease on a planet where infections still kill
over 13 million children every year.
There is still controversy in neuroscience over the exact parts
of the brain that are activated in the case of disgust. One reason
for this may be that neuroscientists still tend to take disgust
stimuli off the shelf, rather than selecting them in a princi-
pled manner from the stimuli that should be disgusting
according to evolutionary theory; that is, those containing a
threat of infection or social parasitism. Until they do this,
it will be impossible to develop a detailed and replicable
picture of the mechanics of disgust and its component parts
using brain scanning.
Beyond better science, there are practical benefits to adopt-
ing an adaptive view of disgust. Those who have to deal with
disgusting material as part of their job would benefit from a
recognition that disgust is a part of human nature. Those who
have to deal with wastes, care for the sick, or groom others
have to do emotional labor in overcoming disgust and would
benefit if this fact were acknowledged and increased support
provided. Tendencies to marginalize such individuals can also
be recognized and counteracted, in the same way that societies
have increasingly come to reject racism or other types of
prejudice.
In the domain of public health, it would be helpful to
recognize that evolved responses to infection avoidance may
not be a good guide to the best way to avoid disease in
our modern environments. For example, quarantine and the
avoidance of physical contact were intuitive, but not useful
responses in the HIV/AIDS pandemic. Recent campaigns have
successfully harnessed the power of disgust to get people to
wash their hands with soap in a variety of settings to avoid
diarrheal and influenza pathogens. Regular panics about the
contamination of food are costly for companies and countries
and might be avoided if we understood more of the operation
of intuitive disgust.
In the moral domain politicians have a sorry history
of making capital by exploiting a tendency to xenophobia -
disgust of the outsider, as an aid to entrench their own power.
The more such behavior is recognized and is generally
understood, the less effective it will become.
Probably our most important scientific task in the domain
of disgust is to tease apart the story of the role that disgust plays
in the evolution of human morality. This is the frontier to
watch where exciting discoveries are going to be made in the
next decade.
See also: Anxiety Disorders; Evolutionary Psychology;
Obsessive-Compulsive Disorder.
Further Reading
Angyal A (1941) Disgust and related aversions. Journal of Abnormal and Social
Psychology 36: 393-412.
Cashdan E (1994) A sensitive period for learning about food. Human Nature 5: 279-291 .
Combes C (2001) Parasitism: The Ecology and Evolution of intimate Interactions.
Chicago: University of Chicago Press.
Cremer S, Armitage SAO, et al. (2007) Social immunity. Current Biology 17(16):
R693-R702.
Curtis V (2007) Dirt, disgust and disease: A natural history of hygiene. Journal of
Epidemiology and Community Health 61(8): 660-664.
Curtis V, Aunger R, and Rabie T (2004) Evidence that disgust evolved to protect from
risk of disease. Proceedings of the Royal Society B 271 (supplement 4): SI 31-SI 33.
Darwin C (1872) The Expression of the Emotions in Man and Animals. Chicago:
University of Chicago Press.
Disgust 709
Douglas M (2002) Purity and Danger. New York: Routledge.
Drummond RP, Stevenson R, Case T, and Oaten M (2009) Can the emotion of disgust
be harnessed to promote hand hygiene? Experimental and field-based tests. Social
Science & Medicine 68: 1006-1012.
Fessler D and Navarette C (2003) Domain specific variation in disgust sensitivity across
the menstrual cycle: Evidence in favour of an evolutionary account of sexual disgust.
Human Behaviour and Evolution 24(6): 406-417.
Joyce R (2006) The Evolution of Morality. Massachusetts: MIT.
Kolnai A, Korsmeyer C, and Smith B (2004) On Disgust. Peru, IL: Open Court Publishing.
Miller Wl (1997) The Anatomy of Disgust. Cambridge, MA: Harvard University Press.
Murphy FC, Nimmo-Smith I, and Lawrence AD (2003) Functional neuroanatomy of
emotions: A meta-analysis. Cognitive, Affective & Behavioral Neuroscience 3(3):
207-233.
Oaten M, Stevenson RJ, and Case T (2009) Disgust as a disease-avoidance mechanism.
Psychological Bulletin 135(2): 303-321.
Rozin P, Haidt J, and McCauley CR (2008) Disgust. In: Lewis M and Haviland JM (eds.)
Handbook of Emotions, pp. 757-776. New York, NY: Guilford Press.
Wilson RR (2002) The Hydra’s Tale: Imagining Disgust. Edmonton, Alberta, Canada:
University of Alberta Press.
Zimmer C (2000) Parasite Rex. New York: Touchstone.
Dissociative Disorders
E Cardena Lund University, Lund, Sweden
© 2012 Elsevier Inc. All rights reserved.
Glossary
Depersonalization disorder Persistent and recurrent
episodes of feeling estranged from the self and/or the
environment.
Dissociation Experienced loss of information or control
over mental processes that, under normal circumstances,
would be available to conscious awareness, self-attribution,
or control.
Dissociative amnesia A dissociative disorder characterized
by a nonmedically caused loss of personally significant
memories.
Dissociative fugue A type of psychogenic amnesia that also
includes sudden wandering away from one's residence or
job alongside with other alterations of consciousness.
Dissociative identity disorder The most severe of
the dissociative disorders, characterized by a lack
of integration and control among different
psychophysiological states or identities, and a number
of other symptoms.
Spirit possession The experience, often accompanied
by amnesia, in which one's identity is substituted
by that of another being. It most often occurs as part
of religious practices that do not involve any
psychopathology.
Trance An alteration of consciousness featuring
apparent lack of awareness or unresponsiveness to the
surroundings, sometimes accompanied by automatic
movements.
Dissociation
At the same time that Aristotle was explaining learning as a
conjoining or association of different psychological elements,
Greek sibyls in the Delphi and other oracles had episodes in
which they seemed to have their ordinary identity substituted
by that of Apollo or other Gods. All throughout history, there
have been descriptions in which ordinary processes such as
memory and personal identity seemed to be in abeyance, or
parallel psychological strands vied for the person's cognition
and behavior. Those phenomena are currently being studied
under the rubric of dissociation, a complex term encompassing
a broad domain of psychological processes and experiences. In
contrast with association, the assembling of different processes
or parts, dissociation implies either that processes that were
conjoined were split asunder or that they did not become
integrated or associated in the first place as would be expected.
Before focusing on the dissociative disorders (DD) per se, it is
important to understand what is meant by dissociation, includ-
ing its pathological manifestations.
Dissociation is a term used in various disciplines (e.g., disso-
ciation of chemical processes). In psychology, it generally
refers to a discrepancy between indexes of information (e.g.,
variations between implicit and explicit perception or mem-
ory). In the areas of clinical and personality psychology and
neurology, this idea can be subdivided into ( 1 ) the apparent
absence of conscious awareness of impinging stimuli or ongo-
ing behaviors (e.g., becoming unresponsive to or unaware of
the surroundings as a consequence of a ritually induced altered
state of consciousness or an epileptic seizure), (2) the seeming
coexistence of separate mental systems or identities that are not
integrated with the self (e.g., instances of dissociative identity
disorder (DID) and spirit possession or, in neurology, cases of
'alien hand'), and (3) ongoing behaviors or perceptions incon-
sistent with an honest introspective verbal report (e.g., the
'blindsight' phenomenon in which an individual with damage
in the striatal cortex reports being blind in a hemifield while
behaviorally showing awareness of occurrences in that hemi-
field, or a similar uncoupling of reports and behavior in a
patient with a conversion - or pseudoneurological - disorder).
In order to differentiate dissociation from the more general
concepts of nonconscious processes, a requirement is that
these events are not just part of ordinary processes such as
overlearning.
Another descriptive notion of dissociation is that of con-
sciousness alterations characterized by experiential discon-
nection with all or parts of the self (e.g., depersonalization,
out-of-body experiences) or the environment (e.g., derealiza-
tion, a sense of unreality of the environment and other beings).
Some authors use dissociation to refer to any alteration in the
state of consciousness, but this is inappropriate; mystical states,
for instance, produce a sense of enhanced connectedness rather
than the opposite.
Dissociation has also been used as an explanatory theoretical
construct. In this sense, based on psychodynamic theory, the
assumption is that dissociation is an intentional mechanism
with the goal to experientially and/or cognitively distance one-
self from traumatic or other material that is threatening. Natu-
rally, it is perfectly reasonable to postulate that dissociative
processes may at times be consciously controlled or even
unconsciously intentional, whereas at other times, they may
not have any specific purpose and just be epiphenomena of
attentional or other cognitive changes.
Figure 1 is a bidimensional heuristic to characterize disso-
ciative phenomena. Axis x indexes whether the processes caus-
ing the phenomena are mostly biological (e.g., neuroendocrine
processes) or psychological (e.g., psychological trauma, hyp-
notic suggestions) processes, whereas axis y groups them as to
whether they can be considered pathological or innocuous. The
remaining of this article focuses on the upper right-hand quad-
rant, that is, pathologies whose etiology is mostly psychological
(naturally, this does not mean that there are no biological
710
Dissociative Disorders 71 1
Pathological
Blindsight
Commissurotomy
Organic amnesia
Epileptic fugues
Hemineglect
Neurological
State-dependent
learning
Sleep amnesia
Dissociative identity
disorder (MPD)
Depersonalization/
derealization
Dissociative
amnesia
Dissociative fugues
Conversion
disorders
Psychological
Hypnosis
Out-of-body
experiences
Automatisms
Implicit perce otion/memory
Normal
Figure 1 The domain of dissociation. Modified with permission from
Cardena E (1994) The domain of dissociation. In Lynn SJ and Rhue J
(eds.) Dissociation. Clinical and Theoretical Perspectives, pp. 15-31.
New York: Guilford.
correlates of these experiences, but that psychological processes
such as exposure to trauma play the major role). With the
exception of DID, which does not have a clear parallel among
neurological disorders, the latter should be considered as dif-
ferential diagnoses for all of the other DD, as they may have, at
least superficially, similar presentations.
I would like to emphasize that dissociation should be con-
sidered a particular form of psychobiological functioning
rather than something necessarily pathological. For instance,
a number of dissociative alterations in consciousness and iden-
tity can be part of ritual or religious activities, are not necessar-
ily associated with pathology, and can even foster creative
endeavors as in the case of some instances of automatic writing
or drawing. The clinician should look at other factors such as
how controllable, organized, desired, and culturally appropri-
ate the dissociative experience is to determine whether it is
pathological.
A comprehensive definition of pathological dissociation is
an experienced loss of information or control over mental
processes that, under normal circumstances and for the age
and developmental level of the person, would be available to
conscious awareness, self-attribution, or control. These phe-
nomena are characterized by (a) involuntary and unwanted
intrusions into awareness or behavior of memories or experi-
ences, with an accompanying loss of continuity in subjective
experience (so-called 'positive' dissociation); and/or (b) an
inability to access information or control mental functions,
manifested in symptoms such as gaps in awareness, memory,
or self-identification, that are normally amenable to such
access/control (so-called 'negative' dissociation), and/or (c) a
sense of experiential disconnectedness, which may include
perceptual distortions, about the self or the environment.
With respect to distal and proximal variables associated
with the propensity to dissociate, there is evidence for both
genetic and nonshared environmental contributions. With
regard to the former, studies have found that the propensity
to dissociate has a heritability component and that the COMT
gene is likely to be implicated. As far as early environmental
contributions are concerned, there is a body of literature
showing that disorganized attachment and related factors
such as dissociative behaviors from the caretakers (e.g., sud-
denly 'freezing,' acting as if they were someone else, or their
inability to have affective control of their inner states) are
strong predictors of the adult dissociation of their children.
With regard to proximal factors, there is abundant evidence
that traumatic events are closely related to acute and chronic
dissociation. If the traumatic events are not very severe and
chronic, such as the possible consequences of short-term natu-
ral disasters, most people experience just short-lived dissocia-
tive episodes. In contrast, particularly virulent events such as
early abuse or being a victim of prolonged brutalization can
give rise to chronic dissociative phenomena. Although a few
authors have argued against a trauma-dissociation link propos-
ing that dissociative individuals just fantasize traumas that
never occurred, a voluminous body of literature with retrospec-
tive and prospective studies on clinical and nonclinical samples
does not support their case.
Dissociation in general, and the DD in particular, can be
evaluated in a number of ways, including behavioral observa-
tions in attachment tests (e.g., stopping all behavior or 'freezing')
or in the clinical situation (e.g., suddenly seeming to become a
very different person without an obvious reason), and a number
of valid and reliable instruments, including standardized and
semistandardized interview and questionnaires, many of them
translated into various languages. Research on measures of neu-
roendocrine dynamics may also be able to serve as reliable
indicators of dissociation in the future.
The Dissociative Disorders
The fairly universal phenomenon of spirit possession was partly
replaced in the West by mesmeric/hypnotic and spiritualist/
channeling phenomena. Although these phenomena were
mostly manifestations of experiences and beliefs that did not
bring about serious distress or dysfunction, there have always
been cases that were pathological even within the cultural
standards of the time. At the inception of psychology as a
science at the end of the nineteenth century, many of the
foremost theoreticians and clinicians in Europe (e.g., ). M.
Charcot, F. W. H. Myers, Pierre Janet, Sigmund Freud) and the
United States (e.g., William James, Morton Prince) sought to
explain and treat such unusual phenomena as patients present-
ing with strange paralyses, or behaviors or reservoirs of knowl-
edge for which they claimed no conscious knowledge or control
but which were psychologically meaningful and did not have a
medical basis. When these manifestations were associated with
dysfunction or distress, they were subsumed under the rubric of
hysteria, a sexist term referring to the uterus, and later became
known as the dissociative and some somatoform disorders
(mainly pseudoneurological or conversion symptoms) in the
Diagnostic and Statistical Manual (DSM) taxonomy. However,
the International Classification of Diseases (ICD) did not split
asunder the DD from conversion. The rationale for the DSM
was that pseudoneurological symptoms seem to be medical,
the same as other symptoms such as hypocondriasis. On the
other hand, it has been shown that the disorders covered in
the following sections have high positive correlations with
712 Dissociative Disorders
pseudoneurological and similar symptoms, and both have
strong associations with a history of trauma and seem to share
underlying psychological processes. Conversion and somato-
form disorders in general will not be further considered here,
but it is worth pointing out that the earlier observation of a
relation between hypnotic suggestion and conversion symp-
toms has been supported by brain imaging and other research
procedures. It has also been proposed that self-suggestion pro-
cesses underlie at least some types of conversion, mediated by
changes in the frontal lobe affecting the cingulate cortex. In a
related fashion, research has shown that individuals with DD
and posttraumatic disorder are more hypnotizable than other
psychiatric groups, so hypnotic susceptibility may very well be a
predisposing factor in the inception of various dissociative
phenomena.
The initial interest in dissociative processes and disorders
during the late nineteenth and early twentieth centuries was
short-lived. The focus on alterations in consciousness and
cognitive processes were replaced by an emphasis on behavior
in academia and on unconscious processes in clinical psychol-
ogy. It was not until the latter part of the twentieth century,
when a number of cultural and intellectual forces came
together, that research on consciousness at large, and dissocia-
tion in particular, reemerged. I can mention the development of
parallel-processing cognitive and neurodevelopmental models
consistent with dissociative processes, as well as the increasing
realization that traumata (e.g., war, accidents, early abuse) are
far more common than realized earlier and that these events are
clearly related to acute and chronic dissociative phenomena.
With regard to the DD, the last years have witnessed increas-
ingly careful and sophisticated treatment, development of valid
and reliable measures, and research which has most recently
included brain imaging techniques.
A word now about the denial by some authors of the
validity and prevalence of DD, especially DID (erstwhile
known as multiple personality disorder), which they claim to be
iatrogenic and/or culturally created, mostly produced by thera-
pists' leading suggestions and circumscribed almost exclusively
to Anglo-Saxon countries. The critiques have been based
almost exclusively on a reconsideration of the literature rather
than on new research and are contradicted by various studies
showing that these disorders are found across the globe,
although with some variations (e.g., the experience of being
chronically possessed by an external agency rather than having
alter identities), and that hypnosis is not a tool used for diag-
nostic purposes, but for therapy. Furthermore, if therapists'
suggestions were the main cause of symptom presentation,
considering that many if not most DID patients have received
earlier diagnoses, they should continue to manifest the initially
diagnosed condition, rather than providing refractory to treat-
ment until dissociative processes are considered (there is now
budding research on the efficacy of therapy for DD). There are
also both cognitive and neurophysiological studies that are
inconsistent with the theory that DD are forms of malingering
or mere role-playing. Naturally, these studies do not eliminate
the influence of culture in the shaping of a syndrome nor the
possibility of iatrogenic factors in a few diagnoses, but the
same could be said for any psychiatric condition. Of especial
value in this controversy have been a number of longitudinal
studies showing that dissociative phenomena in adulthood are
strongly predicted by such earlier variables as traumata (e.g.,
abuse, loss, anatomical malformation requiring painful and
intrusive medical procedures), and disturbances in attachment
to the early caretaker.
With regard to the prevalence of the DD, although it has
been at times stated that hysteria was circumscribed to the
Victorian times, epidemiological research in various countries
has shown that patients in outpatient and inpatient units could
receive a primary or secondary DD diagnosis within a range of
10-45%, depending on their history of trauma and other seri-
ous adversities. In the general population, it has been reported
that up to about 11% might fulfill the criteria for DD, although
this figure probably would vary considerably across cultures
with different levels of exposure to trauma and different criteria
for dysfunction. In the following sections, the DD according to
the DSM-IV are described.
Before this, I should mention, albeit briefly, pathological
dissociation among children. There has not been nearly as
much research with children and teens as there has been with
adults for a number of reasons, including the difficulty of differ-
entiating what may be age-appropriate manifestations (e.g., not
having a sense of self across time in early childhood) from actual
pathological manifestations, the lack until relatively recently of
valid measures for this group, and the differing presentations of
pathological dissociation of children and adults. Regarding the
last point, an expert has written that trance (i.e., seeming unre-
sponsiveness to the surroundings) may be the most prevalent
form of dissociation among children, a phenomenon that
does not seem to be quite as central in the DD of adults. Also,
rather than the multiplication of identities found among adults,
dissociative children seem to fail to reflect on, or mentalize, their
own psychological processes and those of others, besides having
uncontrolled alterations of consciousness.
Dissociative Amnesia
The definition of this disorder by the DSM-IV centers on one or
more instances of amnesia for important personal informa-
tion, which cannot be explained by a medical condition, ordi-
nary forgetfulness, or the developmental amnesia for the first
years of life. This diagnosis is given when the amnesia is the
only or central symptom, rather than just being part of a
syndrome as is the case in acute stress disorder (ASD) and
posttraumatic stress disorder (PTSD), or in more complex
forms of DD such as fugue and DID.
The material that is forgotten is typically related to intense
stress or trauma (e.g., a very painful divorce, a war battle) and
can adopt different forms. Amnesia can be localized in time,
either to a previous event such as the loss of a close person, or
to a period of time such as not remembering some years of life
during which chronic abuse occurred. Amnesia can be selective in
the sense that certain aspects of a trauma that would be expected
to be remembered cannot be accessed (e.g., not recalling how a
brawl concluded in the person shooting someone else). System-
atized amnesia is characterized by a theme connected to trauma
or stress, for instance, not recalling the various visits of a perpe-
trator during the holidays or events having to do with an imped-
ing marriage. Lastly, in the rarely encountered generalized or
global amnesia, the person forgets much or most episodic mem-
ory, not being able to remember even one's name or other basic
Dissociative Disorders 713
personal information. The great majority of amnesia episodes
are not generalized or global, and the material that is inaccessible
at some point may become accessible later on, as the basic
mechanism of memory would predict, notwithstanding some
pronouncements about the impossibility of 'recovered mem-
ories.' Besides these symptom presentations, it is worth stating
that a history of trauma and dissociation have also been asso-
ciated with impaired working memory. Currently, the mechan-
isms through which dissociative amnesia occur are far from
clear, but there is some evidence that both failures to store
memories and inability to retrieve memories may be implicated.
This disorder, the same as other DD, is commonly asso-
ciated with other symptoms, including depression and anxiety,
and typically manifests early in life up to early adult years. There
are various medical conditions (e.g., concussions, vascular pro-
blems, dementia) that also produce amnesia, but the memory
problems in these cases are not so clearly related to severe stress
or trauma and may include such memory problems as antero-
grade amnesia, or inability to learn, that is not a typical feature
of dissociative amnesia.
Dissociative Fugue
This disorder is characterized by generalized amnesia, confu-
sion about one's identity and sudden wandering away from
home or place of residence. Sometimes, the confusion/
amnesia about one's identity may give place to an alternate
identity. Because of the nature of the disorder, these indivi-
duals may first be encountered by law agencies or acute care
hospital units, become homeless, or be taken advantage of by
other individuals (there is some evidence that prostitutes and
exotic dancers dissociate more than average, although it is not
clear whether this symptom preceded their profession, may
result from it, or both). During the fugue episode, the person
may seem partially or totally unresponsive to the external
environment and flees away from a source of stress or trauma
(e.g., an upcoming wedding that produces strong conflict, or a
violent household).
In a differential diagnosis, the clinician should consider the
postictal episodes of wandering known as epileptic porioma-
nia, and psychotic episodes in which, for instance, voices may
tell the person to flee. The pattern of symptoms of these other
conditions and the close association with stress or trauma in
the case of fugue help with the differential diagnosis. Cur-
rently, there is a suggestion for the DSM-5 that dissociative
fugue should be considered a special case of dissociative amne-
sia, which I have also advocated for earlier.
Depersonalization Disorder
The DSM-IV defines depersonalization disorder as persistent
and recurrent experiences of feeling separated from one's body
or psychological processes, but with intact reality testing. Dur-
ing a depersonalization episode, individuals may feel discon-
nected from sensations, feelings, or thoughts, or as if they were
not quite real or were dreaming. Also commonly found are
sensory alterations such as a strange quality to colors or
sounds. Technically, depersonalization refers to this experience
as it relates to the self, whereas derealization relates to the
surrounding environment or to other people, for instance,
'feeling' that others are not quite real. These experiences com-
monly co-occur.
Depersonalization disorder, in which these experiences are
recurrent, chronic, and the central aspect of the person's
suffering or dysfunction, must be distinguished from deperson-
alization episodes, which may be isolated and innocuous, or be
part of a larger syndrome such as in DID or panic disorder in
which great fear is often associated with depersonalization. With
respect to reality testing, whereas in psychosis, the person may,
for instance, believe that he/she is dead, in depersonalization, the
person does know that he/she is not dead but feels that way.
Depersonalization episodes can happen in connection with
a seizure disorder, may be triggered by great anxiety or expo-
sure to trauma, or be the by-product of ingestion of psychoac-
tive drugs or of mental techniques such as meditation and
hypnosis. Although depersonalization episodes may have
appeared initially as part of a mood or an anxiety disorder,
they may remain after the other symptoms have cleared, thus
becoming a unique disorder.
Recent studies on depersonalization disorder have shown
that a frequent etiological factor is early emotional abuse
(rather than, say, physical and/or sexual abuse as in the case
of DID, see below) and, more proximally, severe stress or
trauma. The disorder is associated with memory and attention
problems, functional abnormalities in brain areas related to
sensory integration and body schema, blunted neuroendocrine
response, and hypothalamic-pituitary-adrenal dysregulation.
Because depersonalization and derealization episodes often
happen in tandem or may alternate and there are no clinical
reasons to differentiate them, it has been recommended that
the DSM-5 expand its definition of depersonalization disorder
to also include derealization episodes, which are currently
diagnosed as DDNOS (DD not otherwise specified).
Dissociative identity disorder
The phenomena subsumed under DID have been previously
called double or multiple personality, but the DSM-IV changed
the name to underline that the core of the condition is not
the presence of many personalities, but the failure to integrate
various aspects of the individual into a single personality. This
terminological change has the important implication that a
unified personality is an achievement rather than a given.
Thus, the current therapeutic goal in the treatment of DID is to
integrate different aspects or states of the person so that the
person can partake of a common memory, self-attribution,
and control, rather than trying to find the real personality, as
was the goal earlier on.
DID is defined by the DSM-IV as the presence of two or
more distinct identities or personality states (or alters), with
characteristic patterns of perceiving and behaving, and which
recurrently take control of the individual. Another diagnostic
requirement is psychogenic amnesia, which in the case of DID
can take an additional form besides the ones described earlier,
namely that one identity may remember what another one did
but not vice versa. DID is the most severe of the DD and
typically subsumes a number of other symptoms, including
depersonalization, anxiety, depression, and affective lability
(including attempts at self-injury), chronic anxiety, various
phobias, conversion and other somatization problems, and
714 Dissociative Disorders
substance abuse, eating, sexual, and personality disorders
(especially avoidant and borderline disorders within the last
category). This multiplicity of symptoms has led some authors
to propose that the lack of identity integration is not as impor-
tant as the multiplicity of symptoms, which helps explain why
patients with DID have a history of many previous diagnoses,
often of a mood disorder or schizophrenia, before receiving the
DID one. It has also been observed that the severity and pat-
terning of symptoms can vary across individuals and times.
There are DID patients who can function quite well and may
even go undetected by those around them, whereas other
patients may at times become frankly psychotic.
Other authors have remarked that the presentation of dis-
tinct identities is not easy to detect (contrary to the popular
stereotype DID patients are socially avoidant rather than his-
trionic) and instead propose as diagnostic criteria simpler dis-
sociative symptoms, including hearing voices that are not
attributed to the self and may, at times, exert control on it,
amnesias, pseudoneurological or other somatization symp-
toms, depersonalization/derealization, becoming unrespon-
sive to the surroundings, and experiencing various alterations
in the sense of self. At this point, it is seems that the DSM-5 will
maintain a minimalistic approach to DID rather than have
many symptoms as part of its definition.
As mentioned earlier, some authors have questioned the
validity of this diagnosis but have failed to provide empirical
support for an iatrogenic hypothesis. On the other hand, a study
found that the validity and reliability of the DID diagnosis
was comparable to that of some noncontroversial diagnoses,
and memory and neuroimaging studies have been consistent
with the DID patients' reports that they experience alternate, not
consciously integrated psychophysiological states.
Regarding etiological factors, in a number of studies, the
vast majority of DID patients have reported severe and chronic
forms of early abuse. Some critics have countered that such
reports, coming from highly hypnotizable individuals are sus-
pect, but other studies have shown that the search for some
type of independent corroboration, which may be very difficult
after many years have elapsed, has been substantially consis-
tent with the patients' reports. Nonetheless, it is clear that the
vast majority of children who suffer severe abuse or neglect
do not develop DID, thus other factors have to be involved.
A genetic predisposition to dissociate interacting with some
form of disorganized or disrupted attachment seems to also
be important etiological factors. I should underline that trau-
mata are important particularly if they occur early in life, when
a sense of an integrated self is being developed and is thus
vulnerable to disruptions.
Pathological Spirit Possession and Trance
In the DSM-IV, dissociative trance disorder (DTD) was defined
as 'worthy of further study,' although still lacking enough
evidence to consider it a DD on its own rather than just
being an example of DDNOS. DTD includes trance, namely
marked alterations of consciousness or loss of the usual sense
of identity characterized by a lack of or unusually narrow
awareness of the surroundings and stereotyped behaviors
or movement, but without replacement by another identity.
DTD also includes possession trance, a temporary alteration of
consciousness defined by the replacement of the usual sense of
identity by another one (which may be recognized as an ances-
tor or a spiritual force), stereotyped behaviors attributed to the
possessing entity, and typically full or partial amnesia for the
possession episode.
The alterations of consciousness just described are usually
not pathological and are associated with religious or other
rituals, following a cultural syntax for such events. Research
shows that most experients, under these circumstances, evi-
dence good psychological adjustment. Nonetheless, in a few
cases in both industrialized and nonindustrialized societies,
they may occur outside of cultural norms, be dysfunctional,
and produce distress, thus becoming pathological. A case can
be made that pathological spirit possession is the nonWestern
version of DID (and, with logical equivalence, that DID is a
Western version of pathological spirit possession). Although
this formulation has been made without much empirical
grounding, a recent study in Uganda found that pathological
spirit possession sufferers reported greater traumatic events and
various types of dissociative symptoms than a comparison
group, a pattern similar to that of DID patients in the West.
Both for conceptual and now for empirical reasons, it has
been recommended that the DSM-5 criteria for DID include an
expansion of the phrasing to cover pathological spirit posses-
sion. An obvious advantage of this change would be to have a
more specific diagnosis than the vague DDNOS. DTD would
remain, however, one of the examples of the DDNOS.
DD Not Otherwise Specified
Studies in both Western and non-Western cultures have found
that the catch-all bag diagnosis of DDNOS is the most preva-
lent DD diagnosis, which suggests that the DSM classification
is incomplete in some important ways. The proposed changes
to the DSM-5, if accepted, will partly ameliorate the problem
by having derealization without depersonalization being
included within depersonalization disorder, and pathological
spirit possession within DID. Nonetheless, it is very likely that
some presentations will still remain outside of the main diag-
nostic entities. One example is likely to be individuals who are
mostly unable to regulate their states of consciousness, espe-
cially when dealing with even minor stresses, which may be
experienced by them as severe stresses or even traumas. They
may then have mixed episodes of 'blanking out' (i.e., seem-
ingly becoming unaware or unresponsive to the environment),
either because they do not seem to have conscious experiences
at that time or they are absorbed in some distressing memory
or fantasy, alongside with bouts of depersonalization, amne-
sia, feeling that they cannot control different aspects of their
selves, and so on.
Currently, the DSM-IV provides the following examples of
DDNOS: cases similar to DID that do not fulfill all of its criteria
(for instance, experiencing multiple identities without amne-
sia); derealization without depersonalization; dissociative
states in those subjected to coercive persuasion; DTD (includ-
ing both the pathological possession and trance manifestations
described earlier); loss of consciousness, stupor, or coma not
attributable to a medical condition; and Ganser's syndrome or
the giving of approximate answers to questions (when it is
associated with alterations of consciousness and is not the
Dissociative Disorders 715
product of ignorance or temporary confusion ). Two further
examples of DDNOS are being proposed for the DSM-5: acute
dissociative reactions to stressful events and acute states char-
acterized by mixed dissociative and psychotic (e.g., hallucina-
tions, delusion) symptoms.
I would like to comment further on two of the current or
proposed examples.
The first one refers to nonmedical loss of consciousness,
stupor, or coma. This is a case in which the frontier between
dissociative and conversion symptoms is so blurred as to have
no meaning. For instance, how can one really distinguish
between a pseudoneurological petit mal or absence seizure
from stupor or loss of consciousness? Another important
issue is that even though there is a common assumption that
psychological disorders are typically individual, the literature
on mass unexplained medical illness shows that such symp-
toms as fainting or hallucinations can spread within a group of
susceptible individuals in both Western and non-Western
countries.
The second point is that at least as currently phrased, the
proposed example of 'acute dissociative reactions to stressful
events' has a great overlap with ASD, an entity listed under the
Anxiety Disorders and defined as an acute posttraumatic reac-
tion that includes dissociative and PTSD (i.e., reexperiencing,
avoidance, or hyperarousal) symptoms.
Worthy of mention also are a number of culture-bound
syndromes (e.g., latah, amok) that include dissociative phe-
nomena and are not covered by the DDNOS examples
provided. Thus, a clinician is well advised to consider that
dissociation can occur in all major areas of functioning (self-
image, volition, memory, identity, perception, emotion)
above and beyond the descriptions provided by psychiatric
taxonomies. Whether these phenomena are pathological or
not requires not only consideration of the concept of dissoci-
ation in general, but of the semantic networks of diseases,
values, and ethnoepistemologies present in every culture,
along with such aspects as the ability to control and organize
the experience.
In sum, after a hiatus of a few decades, the interest in the
research, evaluation, and treatment of the DD has been
rekindled. We have barely begun to investigate and try to
understand what dissociation can reveal to us about how
humans achieve epistemological and phenomenological inte-
gration and what produces its willed and unwilled disruptions.
Wherever these studies may lead, there is good evidence now
that DD are not nearly as rare or exotic as clinicians used to
think even some years ago. The founder of the most important
psychopathology journal, Morton Prince, may have had the last
word when he wrote in 1906-1907 that dissociative phenom-
ena in their mild forms are 'an almost every-day clinical affair.'
See also: Posttraumatic Stress Disorder; Somatoform Disorders.
Further Reading
American Psychiatric Association (2000) Diagnostic and Statistical Manual of
Mental Disorders, 4th edn., text rev. Washington, DC: American Psychiatric
Association.
Brown RJ, Cardena E, Nijenhuis E, Sar V, and van der Hart 0 (2007) Should conversion
disorder be re-classified as a dissociative disorder in DSM-V? Psychosomatics
48: 369-378.
Cardena E and Carlson E (201 1 ) Acute Stress Disorder revisited. Annual Review of
Clinical Psychology 7: 245-267.
Cardena E and Gleaves D (2007) Dissociative disorders. In: Hersen M, Turner SM, and
Beidel D (eds.) Adult Psychopathology & Diagnosis, 5th edn., pp. 473-503.
New York: Wiley.
Cardena E, van Duijl M, Weiner L, and Terhune D (2009) Possession/trance
phenomena. In: Dell PF and O’Neil JA (eds.) Dissociation and the Dissociative
Disorders: DSM-V and Beyond, pp. 171-181. New York: Routledge.
Cardena E and Weiner L (2004) Evaluation of dissociation across the lifespan.
Psychotherapy 41: 496-508.
Dell PF and O’Neill J (eds.) (2009) Dissociation and Dissociative Disorders: DSM-V and
Beyond. London: Routledge.
Diseth TH (2006) Dissociation following traumatic medical treatment procedures in
childhood: A longitudinal follow-up. Development and Psychopathology
18:233-251.
Foote B, Smolin Y, Kaplan M, Legatt ME, and Lipschitz D (2006) Prevalence of
dissociative disorders in psychiatric outpatients. American Journal of Psychiatry
163: 623-629.
Gleaves DH (1996) The sociocognitive model of dissociative identity disorder:
A reexamination of the evidence. Psychological Bulletin 120: 42-59.
Holmes EA, Brown RJ, Mansell W, et al. (2005) Are there two qualitatively distinct forms
of dissociation? A review and some clinical implications. Clinical Psychological
Review 25: 1-23.
Lanius RA, Vermetten E, Loewenstein RJ, et al. (2010) Emotional modulation in PTSD:
Clinical and neurobiological evidence for a dissociative subtype. American
Journal of Psychiatry 167: 640-647.
Loewenstein RJ (1991) Psychogenic amnesia and psychogenic fugue: A comprehensive
review. In: Tasman A and Goldfinger SM (eds.) Review of Psychiatry, vol. 10,
pp. 189-222. Washington, DC: American Psychiatric Press.
Ottnow-Lewis D and Putnam FW (eds.) (1996) Child and adolescent psychiatric clinics
of North America: Dissociative identity disorder/multiple personality disorder.
Putnam FW (1997) Dissociation in Children and Adolescents: A Developmental
Approach. NewYork: Guilford Press.
Simeon D, Gross S, Guralnik 0, Stein D, Schmeidler J, and Hollander E (1997) Feeling
unreal: 30 cases of DSM-lll-R depersonalization disorder. American Journal
of Psychiatry 1 54: 1107-1113.
Van Ommeren MV, Sharma B, Komproe I, et al. (2001 ) T rauma and loss as determinants
of medically unexplained epidemic illness in a Bhutanese refugee camp.
Psychological Medicine 31: 1259-1267.
Relevant Websites
www.isst-d.org - International Society of Trauma and Dissociation.
www.istss.org - International Society for Traumatic Stress Studies.
www.dsm5.org/ProposedRevisions/Pages/Default.aspx - Webpage for proposed
DSM-5 diagnoses.
Divorce
S Stack, Wayne State University, Detroit, Ml, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Agreeableness A tendency to be warmhearted, a team
player, supportive, trusting, and generous; one of the five
principal dimensions of personality in the five-factor model.
Crude divorce rate The number of divorces per 1000
persons in the general population.
Divorce A dissolution of a marriage.
Five-factor model of personality A model of personality
orientations based on five principal dimensions of
personality including openness, conscientiousness,
extroversion, agreeableness, and neuroticism.
Mania A psychiatric disorder marked by long-term
tendencies toward abnormally elevated expansive
or irritable moods, and supplemented by such
characteristics as grandiosity, racing thoughts, and
pressure to keep talking.
Social integration The strength of the bonds between
the individual and the social order, including the number
of shared beliefs an practices.
Suicide The act of taking one's own life knowingly and
intentionally.
Over the last half century, the percentage of the population
that reported that "having a good marriage and family life
is extremely important" has remained relatively constant at
70-85%. Nevertheless, given the rising number of divorce
cases, fewer people realize this cultural value.
Divorce is a significant indicator of the quality of marriages
and, more generally, of the quality of life. Divorce often breeds
feelings of rejection, loss, and failure. On scales of traumatic
life events, divorce is second only to the death of a spouse. For
adults, divorce increases the odds of many social problems,
including financial strain, depression, a sense of disorientation,
and suicide. Divorce is also associated with lower well-being for
children both in the short term on such indicators as school
performance and psychological distress, and in the long term
on indicators such as social mobility, psychological problems,
and divorce as an adult. As divorce becomes more prevalent,
it can interfere with the production of the next generation of
appropriately socialized and trained persons for society.
While divorce is associated with many stress-producing
consequences, approximately a million couples continue to
obtain divorces every year. Recent national survey data indicate
that 43% of first marriages end in divorce within 15 years.
At least half of the persons currently getting married will ulti-
mately have their marriages end in divorce. Divorce is becom-
ing part of the normal 'life course' or socialization experience
of people. Most people will ultimately and eagerly seek a
divorce at some point in their lives. It is not clear, however,
if the benefits of ending an unhappy marriage are, in fact,
greater than the costs of a divorce.
The Extent of Divorce
Descriptive Data and Trends
The United States
Currently, there are 23 million divorced adults in the United
States and an even larger number of remarried, formerly
divorced persons (four of ten marriages involve a bride and/
or a groom who were previously divorced). The number of
children of divorced parents is approaching the number of
children from intact homes.
Divorce rates measure the flow of marital breakdown. They
refer to divorces per 1000 general population in a particular
year. Rates vary across geographic space. The rates for the South
and West tend to be higher than those for the Midwest and
Northeast. These differentials have been narrowing over time.
In 2007, states with the highest rates of divorce were Nevada
(6.5) and Arkansas (5.9) and states with the lowest rates were
Massachusetts (2.3) and Rhode Island (2.8), states with a high
proportion of Catholics.
A second measure of the incidence of divorce is the stock of
divorced persons (the percentage currently divorced or the
percentage ever divorced). The percentage of the population
that has ever been divorced varies little across regions. This
stock measure is lower in the Northeast at 19% compared to
26-27% elsewhere.
Gender is related to the odds for remaining divorced.
Females, being somewhat less likely to remarry than males,
are more apt to be divorced than males. Using a stock measure
for 2008, for example, 8.5% of males over 15 were divorced
compared to 11.1% of females over 15.
The risk of being divorced varies over the life course. Table 1
provides data for the percentage of nine gender/age groups
who were divorced in 2008. The percentage of males who are
currently divorced increases with age and peaks at 15.0% for
males aged 50-54. For females, there is a similar age gradient,
but the peak is at 55-64. In all age ranges, the percentage of
females who are currently divorced is higher than that for
males. Divorced males opt to remarry at a higher rate, and
sooner, than do divorced females.
As the proportion of persons in society who opt to marry
decreases, it is increasingly important to conceptualize divorce
in the context of the population at risk: the married popula-
tion. This is especially important in describing the incidence
of divorce between Caucasian and African-Americans.
For example, in 2008, 10.5% of Caucasians were divorced
and 57.7% were married, the ratio being approximately one
divorced white for every six married ones. In 2008, 12.1% of
716
Divorce 717
Table 1 Percentage of selected age groups who are currently
divorced by gender, 2008, United States
Age group
Percentage of males divorced
Percentage of females
divorced
20-24
0.8
1.4
25-29
3.0
4.6
30-34
6.4
8.1
35-39
9.1
11.4
40-44
11.9
15.2
45-49
14.2
17.5
50-54
15.0
17.6
55-64
14.0
18.7
65-74
9.5
13.9
Table 2 The percent divorced and married by race, 1990-2008
1990
2000
2008
Caucasians % divorced
8.1
9.8
10.5
Caucasians % married
64.0
62.0
57.7
Blacks % divorced
10.6
11.5
12.1
Blacks % married
45.8
42.1
35.1
Source: US Bureau of the Census (2009) Statistical Abstract of the US, 2009.
Washington: US Government Printing Office.
blacks were divorced, but only 35.1% were married, the ratio
being approximately one divorced black for every three that
were married. Table 2 presents trend data on the marital status
of whites and blacks indicating that the ratio of the stock of
divorced persons to the stock of married persons has increased
over the two decades for both races. It is also noted that in
2008, among adults, four of ten blacks and two in ten whites
fell into the category 'never married.'
As the proportion of married persons in society dwindles,
the divorce rate (divorces per 1000 general population) would
be expected to go down since the population at risk of divorce
is smaller. This is what has been happening in the last two
decades in the United States and many other industrialized
nations.
There is some inequality in who initiates divorce. Wives
initiate approximately six out of every ten divorces, while
men initiate about three in ten. Joint actions constitute the
remainder of the filings. When children are present, wives
initiate nearly the same percentage of filings.
Half of all divorces involve at least one child under the age
of 18. The average number of children per divorce has been
falling since 1964, a reflection of a decline in overall family
size. The median duration of marriages that end in divorce has
held relatively steady since 1970: 7years.
Prior to the twentieth century, the custody of children was
uniformly given to the father who was legally responsible for
providing for his children. Given the spread of Freudian ideas
on child development, this pattern totally reversed in the twen-
tieth century, with the courts uniformly awarding custody to
mothers. Today, while fathers are increasingly awarded joint (or
shared) legal (decision-making) custody, it is still unusual for
fathers to obtain physical (residential) custody of their children.
However, joint physical custody is becoming more common.
Trends
The incidence of divorce has increased substantially over the last
century and a half. The earliest data on the estimated percentage
of marriages that ultimately ended in divorce are for 1867.
Ultimately, 5.3% of the persons who married in 1867 became
divorced. This figure doubled to 12.0% for those married in
1900. Currently, most marriages will ultimately end in divorce.
Between 1921 and 1965, the divorce rate had been rising
very slowly in the United States. The two noticeable departures
from this pattern were a drop in the depths of the Great
Depression and a short-term substantial upswing in the after-
math of WW II. As the postwar baby boomers began to wed,
the divorce rate nearly doubled between 1965 and 1975. It
reached a record high of 5.3 in 1981. Since that time it first
leveled off and has been declining since 1992. Currently, the
American divorce rate is 3.7. The decline is associated with
the aging of a large cohort, the 'baby boomers,' who are now
well past their divorce-prone years. In addition, marriage rates,
generating a population at risk for divorce, have fallen. The
population of persons living together without marriage has
increased more than tenfold in a space of a generation, thus
further decreasing the stock of persons at risk for divorce.
Comparative perspectives: other nations
The United States has one of the highest divorce rates in the
world. This has been true for over a century. In 1869, the rate
was 1.2, a rate higher than some industrialized nations in
the 1990s.
Table 1 presents data on divorces per 1000 population for
selected nations of the world. Nations with a high proportion
of Catholics (e.g., Ireland, Italy) tend to have a low divorce
rate; Ireland had no legal provision for divorce until 1997.
Norms against divorce are especially strong among Catho-
lics and loosen with economic development. The Russian
Federation has surpassed the United States and now has the
highest divorce rate of this group, with the United States in
second place.
It has not been fully understood why the rate of divorce in
the United States is so high. One possibility is the high level of
individualism, a quest after the maximization of individual
goals and pleasures. The country is relatively low in 'collectiv-
ism.' For example, the proportion of the labor force that
belongs to unions, a collective form of organization, is the
lowest in the industrialized world. If the costs of a marriage
outweigh its benefits, Americans are apparently more likely
than the members of most other nations to seek a divorce.
Between 1990 and 2007, the divorce rate decreased in
7 nations, in 14, it increased, and in 4, it remained rela-
tively constant. In four nations, the increase was 100% or
more, including Ireland where divorce was legalized in 1997
(Table 3).
The Causes of Divorce
Macrosociological Explanations
Changes in divorce legislation
In colonial America, there were no divorce courts and divorces
were rare events. Divorces were a legislative matter, and each
and every divorce was debated in the colonial congress!
718 Divorce
Table 3 Divorces per 1000 population circa 1990, and circa 2007
selected nations, and trend (if >10% change)
Nation
Divorce rate
1990
Divorce rate
2007
Trend
Australia
2.5
2.3
Constant
Austria
2.1
2.5
Up
Belgium
1.9
2.8
Up
Canada
3.1
2.2
Down
Denmark
3.1
2.6
Down
Finland
2.9
2.5
Down
France
1.9
2.2
Up
Greece
0.9
1.2
Up
Hungary
2.9
2.5
Down
Ireland
0.0
0.8
Up+
Italy
0.5
0.8
Up
Israel
1.3
1.9
Up
Japan
1.3
2.0
Up
Luxembourg
1.1
2.3
Up-t-
Netherlands
1.9
2.0
Constant
New Zealand
2.7
2.3
Down
Norway
2.4
2.2
Constant
Poland
0.9
1.7
Up
Portugal
1.0
2.4
Up+
Russian
Federation
3.9
4.8
Up
Spain
0.6
2.8
Up+
Sweden
2.2
2.3
Constant
Switzerland
2.2
2.6
Up
UK
2.9
2.4
Down
USA
4.7
3.6
Down
Source: United Nations (1992) Demographic Yearbook. New York: United Nations;
United Nations (2009) Demographic Yearbook 2007. New York: United Nations.
Notes: up+ = more than a 100% change.
Further, one alternative to marriage, living alone, was either
banned or heavily taxed. With the development of judicial
divorce, more lenient divorce legislation, female labor force
participation, and other changes, divorce was more common
by the late nineteenth century.
An analysis of 1 7 industrial nations found that the degree
of liberality in the divorce law was the variable most closely
associated with the variance in divorce rates; it was more
important than economic conditions and female labor force
participation. However, the liberality of divorce legislation was
largely determined, in turn, by the strength of Catholicism.
Divorce law itself may be a reflection of the status of religion
in society.
Economic factors
Two conflicting theories link the business cycle to the divorce
rate. In the first view, economic prosperity should increase the
divorce rate since divorce is less costly in prosperous times.
Prosperity provides unhappy couples the financial means for a
divorce. Proponents of this first view often note the sizable
drop in the divorce rate during the worst years of the Great
Depression. According to the second view, economic prosper-
ity should decrease the divorce rate since it reduces stress on
families. In this perspective, it is economic depression that
increases the divorce rate. Financial strain adds tensions to
family life that in turn increase the probability of divorce.
While surprisingly little rigorous empirical testing has been
done on these opposed perspectives, the weight of the evi-
dence, from the post-WW II era, suggests that economic pros-
perity (e.g., low unemployment) has a slightly negative effect
on divorce. The size of this effect is typically less than that
of the age structure and the extent of female labor force
participation.
Women's increased labor force participation is perhaps the
most frequently cited factor underlying the increase in the
divorce rate since World War II. As economic opportunities
for women have increased, more women have obtained an
independent financial base. They are less dependent on mar-
riage for monetary support and more apt to seek a divorce from
an unhappy marriage. Increased female labor force participa-
tion has often been the leading determinant of shifts in the
incidence of divorce. However, other nations have even higher
rates of employed women than the United States and such
nations as Sweden often have lower divorce rates. There is
more to the divorce equation, then, than this factor.
The age structure of the population can affect divorce
through influencing economic opportunities for different age
cohorts. As the proportion of young people in society expands,
this sets in motion certain rigidities in the supply of labor. An
oversupply of labor results in more competition for employ-
ment. An abundance of young workers results in the fall of
their incomes, more financial strain on their marriages, and,
hence, a higher incidence of divorce.
Social integration
Factors such as religiosity, migration, urbanism, and war can
affect the strength of the bond between the individual and
society. The extent of shared beliefs and practices can be
affected by these conditions; when bonds to social institutions
like religion and neighborhoods are weakened, life can become
less meaningful. Urban life can promote a sense of anonymity
undermining integration and increasing tolerance for divorce.
In such contexts, the probability of divorce increases.
A study of divorce rates in 3000 counties found that indi-
cators of social integration were more important than socio-
economic variables in explaining the variation in divorce rates.
In particular, the higher the incidence of church membership,
the lower the rate of divorce. Also, the higher the rate of
population growth, the higher the rate of divorce. Finally, the
higher the urban percentage, the higher the incidence of
divorce.
As divorce becomes more commonplace, the normative
climate surrounding divorce is liberalized. The stigma asso-
ciated with divorce may decline, resulting in an increase in
divorce; divorced persons can provide role models and encour-
age others to follow suit. Overtime data for the United States
confirm a positive association between divorce and the previ-
ous level of divorce. This suggests some support for a cultural
perspective.
Individual-Level Explanations
Personality, psychiatric disorders, dissimilarities
According to the selection hypothesis, persons who succeed in
long-term marriage are a select group. Such persons have solid
personalities before and during marriage, which help their
Divorce 719
marriages to be satisfying and rewarding. Thirteen studies are
available that measure personality in large samples of married
persons and then keep track of who gets divorced as time goes
by. A recent meta-analysis reviewed all 13 over time studies,
which measured personality using the dominant 5-factor
model. Three of the five major dimensions of basic personality
orientation were predictive of divorce. The best predictor was
agreeableness (a tendency to be warm, a team player, support-
ive, trusting, and generous). Agreeable persons were signifi-
cantly less apt to become divorced. The second best predictor
was neuroticism. Neurotic persons (e.g., chronically depressed,
overanxious) were considerably more apt than nonneurotic
persons to eventually divorce. Finally, conscientious persons
(tendency to be well organized, industrious, efficient, enter-
prising) were less apt to become divorced. Importantly, these
understudied personality dimensions were at the same level of
importance in predicting divorce as more traditional major
predictors, including socioeconomic status. The risk of divorce
also has a hereditable component, perhaps operating through
personality constructs.
One of the best predictors of divorce is age at first marriage:
the higher the age, the lower the odds of divorce. For example,
in the United States, marrying at over 25 years of age lowers the
risk of divorce by 24% (compared to marrying under 18). Age
at first marriage predicts the risk of divorce across 17 nations
for which data are available. However, to the extent that age
captures maturity, it may be, in part, a proxy for aspects of the
conscientious and agreeable personality types.
While there is little systematic, overtime research, evidence
suggests that long-standing, serious psychiatric problems in-
crease the odds of divorce. Analyses using the national co-
morbidity surveys determined that 11 of the 14 psychiatric
disorders investigated were associated with an elevated risk of
divorce over time. The manic disorder (marked by abnormally
and persistently elevated, expansive, or irritable moods and
supplemented by such considerations as grandiosity, racing
thoughts, pressure to keep talking, requiring <3 h sleep) was
the best predictor of divorce. It elevated the odds of divorce by
3.2 times. Persons diagnosed with major depression had an
enhanced risk of divorce 1.7 times that of the general popula-
tion. The greater the number of psychiatric disorders, the
greater the risk of divorce.
Studies tend to find that the greater the differences in status
between spouses, the higher the odds of marital disruption.
This is especially true for dissimilarities in religious preference,
but also extends to differences in age, education, and race.
Communication between the spouses is a function, in part,
of similarities on these characteristics. The associated value
conflicts can increase divorce risk.
Social integration
Social isolation increases risk of divorce. Measured in terms
of the number of ties to friends and community organiza-
tions, the greater the 'integration' the lower the probability of
divorce. This is mainly for the first 7 years of marriage. Stron-
ger results are found for 'normative integration' or the num-
ber of close relatives and friends who have experienced a
divorce. In an 8-year panel study of 2000 married couples,
the probability of divorce was only 9% among those with no
significant others divorced, whereas the probability of divorce
was 16% among those who had both a friend and relative
who had divorced. As normative standards permit or encour-
age divorce in one's social networks, the likelihood of divorce
increases.
One particularly strong predictor of divorce is low religious
integration. Catholics have a lower incidence of divorce than
Protestants, given the stronger norms against such in the for-
mer. From national survey data, being a member of any reli-
gious group versus not having a religious affiliation lowers the
risk of divorce by 13%.
Economic factors
The occupational status, education level, and income of the
husband lower the risk of divorce. Having some college educa-
tion lowers the chances for a divorce by 13%. However, only
income tends to remain significant if one holds the other
variables constant. National survey data in the United States
indicate that persons living in households with incomes over
$50 000 were 30% less apt to divorce than their counterparts
over the first 10 years of marriage. Financial strains place the
family under stress which, in turn, increases the probability of
marital dissolution.
Table 5 presents data on the percentage of men who are
divorced in each income class. Since age status is confounded
with income status (older men tend to make more money than
their younger counterparts), the data refer to a particular age
range containing the "baby boomers,' those 55-64 in 2008.
While these men in Table 4 are about the same age, the
percentage of them who are divorced varies from a low of 8.3
for those with personal incomes of over $ 100 000 to a high of
18.5 for those men with the lowest incomes. Importantly, as
one goes from the income group of men earning $40 000-
$75 000 to a group above the median or typical income
($75 000-$99 999) for men of this age, the percentage
divorced drops from 11.3 to 8.8. That income is associated
with the odds of being married is indicated by the ratio of the
percentage married to the percentage divorced. This ratio
improves from 3.3 for the poorest group to 10.4 for the richest
group of men.
Table 4 The odds of divorce for persons diagnosed with selected
psychiatric disorders, national comorbidity surveys, United States
Disorder
Increased
risk for
men
Increased risk of
divorce for women
Increased risk of
divorce, total
population
Mania
3.3x
4.8x
3.2x
Major
depression
1.7x
1.7x
1.7x
Generalized
anxiety
2.3x
1.4x
1.7x
Drug abuse
Number of
disorders
1.4x
1.4x
1.4x
One disorder
1.3x
1.7x
1.3x
Two
disorders
2.3x
1.7x
1.5x
Three
disorders
2.5x
2.5x
1.9x
720 Divorce
Table 5 The percentage of men, ages 55-64, who are currently
divorced, and currently married by personal income range, 2008, United
States, current population reports
income range
Number of
(A)
(B)
Ratio of
males,
percent
percent
(A)/(B)
55-64
married
divorced
$1 — $5 000
411 000
62.3
18.8
3.31
$5 000-14 999
888000
65.5
17.0
3.85
15 000-24999
1 306 000
71.4
15.1
4.72
25 000-39999
2255000
73.0
15.0
4.87
40 000-74 999
Above median
3998 000
79.9
11.3
7.1
income
ranges:
75 000-99999
6567 000
84.3
8.8
9.6
100000+
1 805 000
86.3
8.3
10.4
Source: http://www.census.gov/population/www/socdemo/hh-fam/cps2008.html.
Table 6 Top five perceived causes of divorce reported by divorced
persons themselves, marital instability over the life course survey, United
States
Perceived cause
Total (%)
Males (%)
Females (%)
Infidelity
21.6
15.6
25.2
Incompatibility
19.2
19.5
19.1
Drinking/drugs
10.6
5.2
13.7
Grew apart
9.6
9.1
9.9
Personality problem
9.1
10.4
8.4
Divorced persons’ perceptions of causes of divorce
When queried about what caused their divorce, divorced per-
sons tend to give little credence to widely referenced factors in
the academic research such as age at first marriage or conflict
resolution skills. Infidelity, for example, rates high on the list, a
relatively unstudied factor in academic research. Table 6 pro-
vides national data on the top five causes of divorce as reported
by divorced persons themselves.
Further insight into the cross-cultural causes of divorce is
provided by anthropologists. Betzig reviews contributing fac-
tors to divorce as recorded by anthropologists in 186 societies,
including the Aztec and Native American tribes such as the
Apache, Huron, and Pawnee.
Family dynamics and process
Premarital cohabitation
While premarital cohabitation might be seen as 'preparation'
for a successful marriage, persons who cohabit before marriage
are at relatively high risk of divorce. This risk factor for divorce
is consistent across 17 nations. Both cohabitation and di-
vorce may be indicators of a predisposition to go against
normative behavior regarding marital conventions.
Children
On the one hand, children are often thought to decrease the
odds of divorce since they can give an unhappy couple 'a
reason to stay together,' or perhaps problems concerning
children per se help to prevent a focus on marital problems,
thereby lowering the risk of divorce. On the other hand, chil-
dren can increase stress; children also represent a source of
numerous economic and social demands on parents. In a
study of over 40 000 marriages across 17 nations, it was deter-
mined that the greater the number of children that a married
couple has, the lower the risk of divorce.
Esteem for spouse
Spouses in happy marriages tend to hold each other in higher
esteem than do divorced persons. The former group is consid-
erably more likely to describe their mate's traits as superior to
their own than the latter group. This factor of support for the
spouse is related to having an agreeable personality.
Sexuality
A badly neglected variable, sexual adjustment distinguishes
happily married couples from divorced spouses. The former
report significantly more enjoyment and higher actual/pre-
ferred sex frequency ratios than the latter. Sexual incompatibil-
ity is one of the more frequent explanations that divorced
persons themselves give for the breakdown in their marriages.
Further, adultery is a much-cited reason for divorce among
divorcees.
Consequences of Divorce
Probably, the best researched area in divorce studies is that of
the consequences of divorce. The effects of divorce include
psychological, economic, and social consequences. The litera-
ture on the effects of divorce is divided first into that discussing
effects on children and that discussing effects on adults. For
children, effects are further subdivided into short- and long-
term impacts. Finally, the sources of variation in the extent to
which people are affected by divorce are discussed for both
children and adults.
Effects on Children
Short-term consequences
Psychological impacts
Meta-analyses of over 100 relevant studies found that parental
divorce increases the risk of psychological maladjustment and
low self-esteem in children.
There is some evidence that the children of divorced parents
have lower peer popularity at school than the children of intact
families. This may be a social consequence of psychological
impairment.
School performance. Two meta-analyses of over 100 studies
found that parental divorce is associated with decreased school
performance. Contributing factors to low-grade point averages
included mothers with high levels of depression and low edu-
cation levels.
The 'task overload' often found in single-parent, divorced
families is associated with weaker supervision, increased labor
force participation of the custodial parent to offset the loss of
the income of their former spouse, and less parental monitor-
ing of school work.
Divorce 721
Criminal behavior
Divorce tends to be associated with a higher incidence of
criminal behavior. A meta-analysis of 72 studies determined
that children from divorced homes have higher rates of felony
thefts, crimes against persons, drug abuse, and other offenses
than children from intact homes. Divorce reduces the amount
of supervision and control of children, both within the family
and in neighborhoods with high proportions of single-parent
families.
Long-term consequences
Psychological impacts: depression
Life-course approaches to divorce and depression have estab-
lished a link between parental divorce and the odds for depres-
sion among children after they become adults. This association
is independent of other marital and socioeconomic life out-
comes of the children.
Divorce
A systematic analysis of 40 000 marriages in 17 nations deter-
mined that parental divorce increases the risk of divorce among
children when they reach adulthood. The children of divorced
parents tend to have more favorable attitudes toward divorce
than the children of intact families. Divorced parents can con-
stitute a negative role model for their children. Children grow-
ing up in families with constant bickering may not adequately
learn the interpersonal skills that contribute to successful
marriages.
In addition, personality orientations and disorders have
genetic components. The intergenerational transmission of
divorce may involve the transmission of genes that pro-
mote disagreeableness, low conscientiousness, and forms of
neuroticism.
Social mobility
The experience of family disruption during childhood substan-
tially increases men's odds of ending up in the lowest occupa-
tional stratum. Family disruption also weakens the association
between the father's occupational status and the occupational
status of the son. These relationships hold for both blacks
and whites.
At the macrolevel of analysis, the increase in divorce rates
since the 1960s is tied to the weakening of the association
between sons' class of origin and class of destination. As
more sons have been coming out of divorced homes, fewer of
them are achieving the amount of occupational advancement
that one would ordinarily expect on the basis of their father's
occupation.
Explaining the variation in effects among children
Some children of divorced parents fare better than others in
their level of well-being. Amato's review of 180 relevant studies
suggests five indicators of well-being: academic achievement,
behavior problems, psychological adjustment, self-esteem, and
quality of social relations. Evidence-based investigations have
reported the most consistent results for two moderating factors:
the quality of the custodial parent's mental health and the level
of conflict between the former husband and the former wife.
First, the hypothesis that the well-being of the children of
divorced parents is positively linked with the postdivorce
psychological adjustment of the custodial parent is supported
in nearly all studies. Better-adjusted parents are able to foster
higher levels of well-being in their children. Further, the higher
the quality of the relationship between the custodial parent
and the child, the greater the well-being of the child.
Interparental conflict generally lowers the well-being of the
child. Nearly 90% of the relevant studies support the notion
that children's well-being is inversely correlated with the level
of postdivorce conflict between their parents.
A majority of studies support the notion that the greater the
economic well-being of the postdivorce household, the greater
the well-being of the child.
The life-stress perspective contends that it is not a single
stressor that erodes children's well-being, but a series or accu-
mulation of stressors. Changing schools, moving, financial
strain, loss of contact with grandparents, and so on, may
ultimately wear down the child and threaten his/her basic
well-being. Two-thirds of the relevant investigations confirm
this notion that the greater the number of stressful life events,
the lower the well-being of the child.
Generally speaking, children of divorced parents are more
apt to have better well-being if their custodial parent is in good
mental health, interparental conflict is low or absent, and the
number of stressful life events is low.
Effects on Adults
Psychological and physical health
Generally, research has found a strong association between
divorce and psychological impairment. For example, national
survey data find that the divorced have a level of depression
that is 40% higher than the national average. Levels of pro-
blems such as loneliness, guilt, anger, and anxiety are higher
among divorced persons than among married persons. These
problems are often attributed to such factors as a deterioration
of the standard of living after divorce, a relative lack of con-
fidants, and a deep sense of loss.
Divorced persons have a risk of suicide that is 2-4 times
greater than married persons depending on the gender and age
group one inspects. For example, divorced males aged 40-44
have a suicide rate of 104.4/100 000, 4.58 times that of married
males of the same age (20.3/100 000). At the societal level, the
relationship between suicide and divorce rates is like a socio-
logical 'law.'
Turning to physical health, research on both the LInited
States and other nations has tended to find a higher rate of
both morbidity and mortality among the divorced than among
the married. The increased risk of illness among the divorced
has been linked to suppressed immunological functioning,
which is especially pronounced among persons who are still
attached to their exspouse.
Economic effects
In an egalitarian marriage where both husband and wife earn
$50 000 a year, they may be able to make a $2000 per month
mortgage or rent payment on a large home place in a desirable
neighborhood. However, after a divorce, all else being equal,
each would half their pooled income. A home place with
a $1000 per month payment might be all each could
afford. It would probably be smaller and in a less desirable
722 Divorce
neighborhood, probably with poorer schools for their chil-
dren. On the whole, the custodial parent is faced with
an ~30% or more reduction in their income and standard of
living. Remarriage is the main avenue for restoring the predi-
vorce standard of living.
Social effects
An immediate target for criminal violence is the exspouse, a
person who is often the perceived source of considerable anger.
Crimes of violence occur among one-quarter of all divorcing
couples. At the aggregate level, the divorce rate is a fairly
reliable predictor of the crime rate in cities.
Explaining the variation in effects among adults
Gender differences
Women experience greater negative financial consequences
from divorce than do men. The degree of this difference, how-
ever, may be changing as the income gap between men and
women's predivorce income narrows.
While both men and women experience an increase in
depression following divorce, the causes of the depression
tend to be somewhat different. For men, loss of emotional
support is cited ahead of financial problems. In contrast, for
women, financial problems are cited ahead of loss of emo-
tional support.
Economic
The financial consequences of divorce are cushioned by the
economic resources, such as skills and degree of labor force
participation, of the custodial parent. Custodial parents with
higher levels of education and training, and who increase the
number of hours worked, fare better than their counterparts.
Support systems/social isolation
A recent meta-analysis of 21 studies determined that the
higher the degree of social relationships following divorce,
the lower the level of maladjustment. Maladjustment indica-
tors include depression, anxiety, negative affect (e.g., unhap-
piness, life dissatisfaction), generalized psychological distress,
and physical symptoms. Social support involving specific
relationships (e.g., one-on-one relationships such as that
with a best friend) buffer against maladjustment. Network-
based or secondary social support such as that gleaned from
membership in groups (e.g., church, circle of friends, neigh-
borhood, and support group) may especially promote posi-
tive adjustment. For older divorced persons, relationships
with their adult children, who can now provide support, are
often critical to divorce adjustment.
Relations with exspouse
Interspousal hostility and bitterness over the divorce impedes
divorce adjustment. Positive attachment, such as dependence,
to a former spouse may account for long-term psychological
distress in some divorces. A relationship with one's ex marked
by indifference may be the healthiest mode. The development
of a new intimate relationship with someone of the opposite
sex improves divorce adjustment.
Cultural beliefs
A belief in the immorality of divorce is associated with higher
levels of psychological stress among divorcing couples. Adher-
ence to liberal gender roles improves divorce adjustment for
women.
Sexuality
An important aspect of divorce adjustment is reestablishing an
intimate sexual relationship. Research at the national level has
found that males were considerably more sexually involved
than females, especially at older ages. Persons low in religios-
ity, high in education, and high in political liberalism were
more sexually active than their counterparts.
Conclusions
Divorce in the Context of Rising Cohabitation Rates
While there have been declines in divorce recently in some
nations, this trend is associated with parallel declines in mar-
riage rates and upticks in the incidence of cohabitation, a key
alternative to marriage. Table 7 provides the essential data for
establishing a context for the study of relationship breakdown
in the future.
In the space of under 50 years, the number of cohabiting,
unmarried American couples increased more than 15-fold
from 439 000 in 1960 to over 6 million in 2008. In the same
period, the percentage of persons who were divorced rose.
Marriage became correspondingly less popular. A total of
88% of males were married in 1960, but only 65.7% were
married by 2008. Importantly, the marriage rate, here, the
number of marriages for every 1000 unmarried women 15
and over, declined by 50% from 73.5/1000 in 1960 to just
39.9/1000 in 2004.
Table 7 Trends in divorce and associated social indicators, 1960-2004
Year
% divorced
males
% divorced
female
% married
male
% married
female
Married rate per 1000
married female
Number of
cohabitants
(1000s)
% kids with both
parents
% births to
unmarried
mothers
1960
1.8
2.6
88.0
87.4
73.5
439
88
5.3
1970
2.2
3.5
89.3
86.9
76.5
523
85
10.7
1980
4.8
6.6
84.2
81.4
61.4
1589
77
18.4
1990
6.8
8.9
74.1
73.0
54.5
2856
73
28.0
2000
8.3
10.2
69.0
71.6
46.5
4736
69
33.2
2004
8.2
10.9
65.7
67.3
39.9
5080
68
34.6
2008
9.0
11.7
Ma
Na
Na
6800
Na
Na
Divorce 723
The rise in unwed births and rise in divorce is associated
with a substantial drop in the percentage of children under 1 8
who live with both biological parents: from 88% in 1960 down
to 68% in 2004. For African-Americans, this figure is substan-
tially less, with only 35% of children now living with both
biological parents.
Importantly, there has been a sevenfold increase in the
percentage of children born to unmarried mothers: 5.3% in
1960 rising to 34.6% by 2004. For African-Americans, fully
68% of births are to unwed mothers.
Cross-national research spanning 17 nations has consis-
tently found that cohabitation before marriage raises the odds
of divorce after marriage. Far from being a way to reduce the
chances of divorce, it apparently backfires and enhances
the chances of divorce. In the United States, over half of all
first marriages are now preceded by cohabitation.
With so many children being born to unwed mothers, and
with the fall in the marriage rate and rise in unmarried cohabi-
tation, future research on divorce needs to include not just the
breakups experienced by married couples, but also the break-
ups of the unions of unmarried couples. In some nations, the
proportion of unmarried cohabiting couples is now greater
than the proportion of married couples.
It seems unlikely that these trends which began a genera-
tion ago are apt to reverse. They are best understood as an
expression of a highly individualist culture. Cultural attitudes
of young people, the group constituting the future of the
family, increasingly support new marriage and family forms.
For example, after increasing over the years, a majority of youth
in the United States now report that "having a child without
being married is experimenting with a worthwhile lifestyle and
not affecting anyone else." Academic research needs to respond
to these trends and begin the systematic study of union disso-
lution among cohabitants.
See also: Aggression; Alcohol: Psychosocial Effects; Anxiety
Disorders; Big Five Model and Personality Disorders; Depression;
Homicide; Love and Intimacy; Marital Dysfunction; Personal
Relationships in Everyday Life; Personality Disorders; Sexual Behavior;
Stress and Illness; Suicide.
Further Reading
Amato P (1993) Children’s adjustment to divorce: Theories, hypotheses, and empirical
support. Journal of Marriage and the Family 55: 23-38.
Amato P (2000) The consequences of divorce for adults and children. Journal of
Marriage Family 62: 1269-1287.
Amato P (2001) Children of divorce in the 1990s: An update of the Amato and Keith
(1991) meta analysis. Journal of Family Psychology 1 5: 355-370.
Amato P and Previti D (2003) People's Reasons for divorcing. Journal of Family Issues
24: 602-626.
Becker G (1 993) A Treatise on the Family, 2nd edn. Cambridge: Harvard University Press.
Betzig Laura (1989) Causes of conjugal dissolution: A cross-cultural study. Current
Anthropology 30: 654-676.
Biblartz T and Rafferty AE (1993) The effects of family disruption on social mobility.
American Sociological Review 58: 97-1 09.
Clarke-Stewart A and Brentano C (2006) Divorce: Causes and Consequences.
New Haven, CT: Yale University Press.
Dronkers J and Harkonen J (2008) The intergenerational transmission of divorce in
cross-national perspective. Population Studies 62: 273-288.
Fine MA and Harvey JH (2006) Handbook of Divorce and Relationship Dissolution.
London: Lawrence Erlbaum Associates.
Jerskey BA, Panizzon MS, Jacobson KC, et al. (2010) Marriage and divorce: A genetic
perspective. Personality and Individual Differences 49: 473-478.
Kessler RC, Walters EE, and Forthofer MS (1998) The social consequences of
psychiatric disorders, III: Probability of Marital Stability. American Journal of
Psychiatry 1 55: 1 092-1 096.
Krumrei E, Coif C, Martin S, Fogo W, and Mahoney A (2007) Post divorce adjustment
and social relationships: A meta analytic review. Journal of Divorce and Remarriage
46: 145-166.
Phillips R (1988) Putting Asunder: A History of Divorce in Western Society. New York:
Cambridge University Press.
Roberts B, Kuncel N, Shiner R, Caspi A, and Goldberg L (2007) The power of
personality: The comparative validity of personality traits, socioeconomic status,
and cognitive ability for predicting important life outcomes. Perspectives on
Psychological Science 2: 313-345.
Stack S (2000) Suicide: A 15-year review of the sociological literature. Suicide and
Life-Threatening Behavior 30: 1 63-1 76.
Stack S and Gundlach J (1992) Divorce and sex. Archives of Sexual Behavior
21: 359-367.
Whitehead BD and Popenoe D (2005) The State of Our Unions: The Social Health of
Marriage in America. http://marriage.rutgers.edu/Publications/Print/printsoou2005.
htm.
Drugs, the Brain, and Behavior
D E Presti, University of California, Berkeley, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Agonist A chemical that binds to a neurotransmitter
receptor and mimics the action of an endogenous
neurotransmitter.
Antagonist A chemical that binds to a neurotransmitter
receptor and has no effect other than to block the action of
endogenous neurotransmitters and other receptor agonists.
Drug A chemical that, in small amounts, has significant
effects on physiology.
Hallucination Perceptual experiences in the absence of
associated external sensory stimuli.
Neurotransmitter A chemical that is released by nerve cells
and diffuses extracellularly to bind to receptors on nerve cell
membranes, thereby mediating intercellular
communication of neural signals.
Neurotransmitter receptor Proteins located in
nerve cell membranes that specifically bind
neurotransmitters, undergo a resulting conformational
change, and thereby mediate changes in cell membrane
electric potential or initiate various intracellular
biochemical processes.
Synaptic cleft At a synapse, the narrow gap between a
neuron and an adjacent neural cell.
Uptake transporter Proteins located in nerve cell
membranes that function to remove neurotransmitter
from the synaptic cleft and move it into the interior of
the cell; often located on the presynaptic axon terminal,
thus returning neurotransmitter into the cell from
which it was released.
Introduction
Drugs are chemicals that, in small amounts, have significant
effects on physiology. Psychoactive drugs are chemicals that
have effects on the psyche or mind, presumably by way of their
interaction with the nervous system and, in particular, the
brain. The human brain is likely the most complex object
thus far encountered in the universe. Its complexity is mani-
fest in its ~100 billion neurons interconnected by trillions of
synapses. In addition, the human brain contains several
hundreds of billions of various kinds of glia cells, some of
which are intimately involved in cellular signaling processes.
Astrocyte glia cells, for example, play central roles in cell-to-cell
signaling at synapses, exhibiting many of the properties for-
merly considered unique to neurons.
Each and every synapse is the setting for an intricate molec-
ular drama involving release of chemical neurotransmitters,
rapid diffusion of neurotransmitter across the narrow gap
(cleft) separating the presynaptic from the postsynaptic neu-
ron, binding of neurotransmitter to receptor proteins - located
in the membranes of the postsynaptic and presynaptic neurons
and of astrocyte glia cells - and resulting conformational
changes of the receptor proteins. If the receptor is an iono-
tropic receptor, the effect will be ion flow across the cell
membrane, a change in the membrane voltage, and a resulting
change in the excitability of the cell. If the receptor is a meta-
botropic G-protein-coupled receptor (GPCR), the result will
be the modulation of intracellular processes that include
ion-channel gating, enzyme activation, and gene transcription.
The neuronal signaling process is terminated when the neuro-
transmitter is taken back up into the neuron that released it, or
into other nearby cells, via membrane proteins called uptake
transporters, or is enzymatically degraded.
Dozens of different molecules have been identified as neu-
rotransmitters in the human brain. Glutamic acid or glutamate
is the primary excitatory neurotransmitter in the human
brain. GABA (y-amino-butyric acid) is the major inhibitory
neurotransmitter. Other neurotransmitters include acetylcho-
line, serotonin, dopamine, norepinephrine, epinephrine, his-
tamine, glycine, adenosine, anandamide, nitric oxide, and
the opioid polypeptides. Additional neurotransmitter mole-
cules no doubt remain to be discovered. The diversity of chem-
ical neurotransmitters and the numerous different receptor
proteins - typically, many different types of receptor proteins
for each type of neurotransmitter - are another facet of the vast
complexity of the brain.
As far as is presently known, psychoactive drugs exert their
influence by interacting with the signal transmission process
at synapses. There are various ways in which this can happen.
A drug can mimic the effect of a neurotransmitter because
sufficient similarity in molecular structure allows the drug to
bind to and activate a receptor in a manner similar to that of
endogenous neurotransmitter. Such a drug is called a receptor
agonist. Or it could bind to a receptor and block the receptor,
preventing its being activated by endogenous neurotransmit-
ter. Such a drug is called a receptor antagonist. Some drugs
enhance neurotransmitter release or leakage from the presyn-
aptic neuron. Others block uptake of neurotransmitter from
the synaptic cleft, or interfere with the synthesis or degradation
of neurotransmitter, or interact with the chemistry of the brain
in ways not yet elucidated.
Psychoactive drugs include the following chemicals: com-
monly used nonmedical substances such as caffeine, alcohol,
and nicotine; medically prescribed substances such as amphet-
amine, morphine, and benzodiazepines; mind-altering sub-
stances that have been excluded from accepted medical use
and are declared illegal, such as heroin, cannabis, and psyche-
delics like lysergic acid diethylamide (LSD) and mescaline; and
medications used to treat mental-health conditions, such as
antidepressants and antipsychotics. Among these drugs, some
724
Drugs, the Brain, and Behavior 725
are frequently used for their pleasure-producing effects (alco-
hol, caffeine, etc.), some are used medicinally (antidepressants,
antipsychotics, etc.), some are used for both (amphetamine,
opioids, benzodiazepines, cannabis, etc.), and some are more
prone to fostering problematic relationships such as addiction
(nicotine, alcohol, benzodiazepines, opioids, etc.).
Historically, psychoactive dmgs were all natural products,
ingested in the form of plants or fungi, or extracts of these
organisms. The word 'drug' comes from the Old English word
'drogge' and the Old French word 'drogue,' words that may
have earlier had the meaning of 'dry' or 'dried plant.' Beginning
in the 1800s, chemists isolated and identified a number of
molecules from plants that had psychoactive effects. Mor-
phine, caffeine, nicotine, atropine, cocaine, and mescaline
were all identified as psychoactive compounds from plants by
nineteenth-century European chemists. Beginning near the end
of the nineteenth century and continuing to the present day,
medicinal chemists have synthesized previously unknown
molecules having psychoactive effects. Some of these new
molecules have been marketed by the pharmaceutical industry
as drugs for the treatment of various conditions.
In the United States, many psychoactive drugs are classified
according to the Controlled Substances Act, a federal law first
instituted in 1970 and copied in some form by all of the states.
The US Controlled Substances Act classifies psychoactive dmgs
that have been declared to have potential for abuse (problem-
atic use leading to possible addiction) into five categories or
schedules. Schedules II, III, IV, and V are legally available for
medical use with an appropriate physician's prescription.
Examples are amphetamine, morphine and other opioids,
and the benzodiazepines. Schedule I dmgs have been declared
categorically illegal and are not available for legal medical
use. Examples are heroin and the psychedelics. Alcohol and
nicotine (or tobacco) are not regulated by the Controlled
Substances Act. Some other countries have similar classifica-
tions of psychoactive dmgs, such as the Misuse of Dmgs Act in
the United Kingdom. This UK Act classifies drugs having abuse
potential into one of three classes: A, B, or C, with Class A dmgs
considered to be potentially the most harmful. Many countries
worldwide have signed on to an international classification
of controlled substances: the United Nations Convention on
Psychotropic Substances, instituted in 1971.
Abuse and addiction are toxic aspects of certain dmgs. Even
drugs that have substantial medicinal benefits may also have
significant toxic effects. Indeed, as the pioneer pharmacologist
Paracelsus (1493-1541) said nearly 500years ago: "Everything
is a poison. The difference between a poison and a medicine
depends on the dose."
In terms of both physiology and behavior, the effects of any
psychoactive substance may be impacted by both the mental
state or set of the user (expectations, prior experience, mood,
etc.) and the physical setting of use (alone, social, therapeutic
context with a healer, spiritual ritual, etc.). Some dmgs are
more susceptible to these 'set-and-setting' effects than others.
Thus, caffeine may generally have a small set-and-setting effect
and alcohol an intermediate one, while for substances such as
cannabis and the psychedelics the set-and-setting effect may be
very large.
This article will briefly describe the essence of what is cur-
rently known about the neurochemical effects of various
psychoactive dmgs, as well as something of their effects on
behavior. Sometimes knowledge of neurochemical effects can
be mechanistically related to behavioral actions, and some-
times there is little understanding as to how the known neuro-
chemistry and a dmg's effects on behavior are related. As
knowledge about the brain increases, we may find that many
of the neurochemical actions of psychoactive dmgs are far
more complex than are presently understood.
Caffeine
Caffeine (1,3, 7-trimethylxanthine) is by far the most widely used
psychoactive dmg in the world. It is found in a number of plants,
the most well-known being coffee ( Coffea arabica, native to
Africa), tea ( Camellia sinensis, native to China), and cacao
(Theobroma cacao, native to South and Central America), from
which chocolate is made. Other caffeine-containing plants
include kola [Cola acuminata, native to Africa), guarana (Paullinia
cupana, native to South America), and yerba mate [Ilex paraguar-
ienis, native to South America) . Closely related to caffeine are two
other methylated xanthines with similar psychoactive effects.
These are theophylline (1,3-dimethylxanthine), found in tea,
and theobromine (3,7-dimethylxanthine), from cacao. Caffeine
is believed to act by antagonizing (blocking) receptors for the
inhibitory neurotransmitter adenosine, thereby producing a
stimulant effect on brain activity.
Nicotine
Nicotine comes from the tobacco plant - Nicotiana tabacum,
Nicotiana rustica, and related species - native to the Americas. It
is generally taken into the body by smoking the dried leaves of
the tobacco plant. It may also be absorbed through the nasal
mucosa or oral-buccal cavity, using a snuff preparation or
chewing tobacco. The major known neurochemical effect of
nicotine is to bind as an agonist at the nicotinic-type receptor
for the neurotransmitter acetylcholine. This produces complex
effects in the brain that are presumably related to the behav-
ioral effects of relaxation, mental stimulation, and focused
attention experienced by users of tobacco. Nicotine also stimu-
lates the cardiovascular system, producing an increase in heart
rate and blood pressure. Chronic exposure to nicotine can
result in cardiovascular disease and increased risk of heart
attack and stroke. Tobacco smoking is related to increased
risk of lung diseases such as cancer and emphysema, appar-
ently due to irritant and carcinogenic substances in smoke. The
carbon monoxide component of smoke produces additional
stress to the cardiovascular system and other body organs by
decreasing the oxygen-carrying capacity of the blood.
Sedative-Hypnotics
Sedative-hypnotics produce relaxation and sometimes eupho-
ria in low doses, drunkenness and sleep in higher doses, and
possible coma and death if the dose is sufficiently large.
Alcohol (ethyl alcohol or ethanol) is the most widely used
sedative-hypnotic drug. Ethyl alcohol is a natural product of
726 Drugs, the Brain, and Behavior
the fermentative metabolism of carbohydrate by yeasts. As this
process occurs easily and naturally in the environment, it is
likely that human consumption of alcoholic beverages goes
far back in history.
Synthetic pharmaceutical sedative-hypnotics include the ben-
zodiazepines (diazepam (US brand name, Valium), alprazolam
(Xanax), lorazepam (Ativan), temazepam (Restoril), etc.), bar-
biturates (phenobarbital, secobarbital, amobarbital, etc.),
and other substances such as chloral hydrate, meprobamate
(Miltown), and methaqualone (Quaalude). Pharmaceutical
sedative-hypnotics have been prescribed medically as treat-
ments for anxiety (anxiolytic agents), as aids to sleep, as muscle
relaxants, and as antiseizure medications. A variety of primarily
hypnotic drugs have also been introduced to the marketplace
in recent years. There include zolpidem (Ambien), zaleplon
(Sonata), and eszopiclone (Lunesta).
General anesthetics are powerful sedative-hypnotics used to
induce loss of conscious awareness and insensitivity to pain
during surgical procedures. They are often volatile organic
compounds inhaled as part of a gas mixture during surgery.
Historically, diethyl ether and chloroform were used in this
way. Contemporary examples include halogenated com-
pounds such as desflurane, enflurane, halothane, isoflurane,
and sevoflurane.
GHB, y-hydroxy butyric acid, is a sedative-hypnotic whose
chemical structure is similar to that of the neurotransmitters
GABA and glutamate. In the United States, GHB has been
declared a Schedule I drug because of its use in recreational
settings. However, it is also available as a Schedule III pharma-
ceutical for the treatment of certain sleep disorders.
Kava kava ( Piper methysticum, native to South Pacific
islands), valerian ( Valeriana officinalis), chamomile ( Matricaria
recutita), lavender ( Lavandula augustifolia), catnip ( Nepeta cat-
aria), and hops ( Humulus lupulus) are examples of plants with
sedative-hypnotic properties.
Many sedative-hypnotic substances are believed to exert
their effect on the brain by interacting with receptors for the
neurotransmitter GABA. While not acting as direct receptor
agonists, their effect at these receptors enhances the action of
GABA as an inhibitory neurotransmitter and thus results in
decreased excitation in the brain. Some (alcohol, for example)
may also have actions at glutamate receptors, reducing the
action of glutamate, which is the brain's major excitatory neu-
rotransmitter. In addition, GHB may itself have direct inhibi-
tory neurotransmitter action in the brain, with its own distinct
receptors. The various plants listed above may contain chemi-
cal constituents that act at the GABA receptor. They may well
also contain multiple psychoactive molecules that contribute
to their overall effects. This is an important distinction between
plants and molecules: Although a single molecular species may
be associated with the primary medicinal effects of a plant, the
plant will always have more complex effects, as the plant con-
tains numerous different chemicals, a number of which may
possess physiological activity.
Cocaine
Cocaine comes from the plant Erythroxylum coca and several
related species, native to South America. Cocaine blocks
reuptake transporter proteins for the neurotransmitters dopa-
mine, norepinephrine, and, to a lesser extent, serotonin. In the
brain, this results in the activation of circuits that increase
alertness and wakefulness, produce euphoria, and decrease
appetite. In the peripheral nervous system, cocaine activates
the sympathetic branch of the autonomic nervous system, with
resultant increased heart rate, blood pressure, pupil size, and
bronchial airway size. The leaves of the coca plant have been
used by indigenous peoples of South America for centuries
(and possibly millennia) as a stimulant and for other medici-
nal qualities. Cocaine, extracted and purified from the coca
plant, has far more intense effects on the body than does
coca leaf. Sympathomimetic effects may damage the cardiovas-
cular system and may be acutely toxic by precipitating a heart
attack or stroke. Excessive stimulation of the central nervous
system (CNS) may result in stimulant psychosis (paranoia,
other delusions, hallucinations) and seizures.
Amphetamine
Amphetamine and its close chemical relative, methamphet-
amine, are synthetic substances that cause the neurotransmit-
ters dopamine, norepinephrine, and serotonin to leak out of
the presynaptic nerve terminal into the synaptic cleft. The
leakage occurs not via the usual process of synaptic-vesicle
fusion and exocytosis of neurotransmitter, but by way of the
reuptake transporter protein somehow passing neurotransmit-
ter from the interior of the axon to the exterior. This results in
activation of the same brain and peripheral nervous system
circuits as with cocaine and substantially similar behavioral
effects. There are a number of pharmaceutically manufac-
tured substances that can be considered molecular relatives of
amphetamine and that work in similar ways. They have been
prescribed as CNS stimulants for the treatment of narcolepsy,
as appetite suppressants, and as treatment for attention deficit
hyperactivity disorder (ADHD). Ephedrine and pseudoephed-
rine (from plants of the genus Ephedra, found worldwide), cath-
inone (from the khat plant, Catha edulis, native to northeast
Africa and Yemen), and the synthetic compounds methylpheni-
date (Ritalin) and phenylpropanolamine are examples of other
related molecules with both central and sympathetic nervous
system stimulant effects resulting from cellular actions similar
to those of amphetamine.
Opioids
Opioids (or opiates) are a group of naturally occurring and
synthetic chemicals that act on the nervous system like opium,
the gummy secretion from the unripe seedpod of the opium
poppy, Papaver somniferum, native to central Europe and the
region around the Mediterranean Sea, and now grown through-
out the world. Natural opium contains morphine and codeine.
The discovery of morphine by Friedrich Wilhelm Sertiirner
(1783-1841) in the early years of the nineteenth century (the
reported date of discovery ranges from 1803 to 1817) was a
landmark event in the history of science. It was the first time
a chemical was isolated and purified from a plant and shown
to account for some aspect of the medicinal properties of the
Drugs, the Brain, and Behavior 727
plant. This became a defining event for organic natural-
products chemistry and contributed to the perspective that
purified chemicals were a more valuable basis for understand-
ing the actions of medicinal substances than whole plants.
Opioids act on opioid receptors in the brain to reduce the
perception of pain (analgesia), suppress cough, constrict
pupils, and slow respiration. They may also produce relaxa-
tion, euphoria, and a dreamy state of consciousness. Medici-
nally, opioids are invaluable as analgesics and cough
suppressants. Acting at opioid receptors in the digestive system,
opioids slow intestinal motility and thus are valuable treat-
ments for diarrhea. The most significant potentially toxic
aspects of opioids include the risk of dependence and addic-
tion, and the acute depression of respiration, which can lead
to death.
A number of opioids are either chemical derivatives of
morphine, codeine, or other related molecules in opium, or
are synthesized from nonopioid precursors. Heroin (diacetyl-
morphine) is a simple chemical derivative of morphine that
allows it to cross the blood-brain barrier about 3 times more
efficiently than morphine. It was first introduced in 1898 as a
pharmaceutical medicine for the treatment of pain and cough
by the Bayer company. Hydromorphone (Dilaudid), oxyco-
done (Percodan, OxyContin), and hydrocodone (Vicodin)
are all derived from opium precursors. Methadone, meperidine
(Demerol), and propoxyphene (Darvon) are examples of syn-
thetic opioids, not derived from opium-based precursors. The
synthetic opioids fentanyl, sufentanil, and carfentanil range
from being about 100 times to nearly 10 000 times more
potent than morphine.
The known opioid receptors are widely distributed in the
brain and spinal cord, and are also found in the digestive
system and elsewhere in the body. Receptors are GPCRs and
are generally characterized as one of three subtypes: mu, delta,
or kappa. The neurotransmitters acting at opioid receptors are a
class of small peptide molecules collectively called endogenous
opioid polypeptides or endorphins (from 'endogenous mor-
phine'). These range in size from the 5-amino-acid-long
leu-enkephalin and met-enkephalin to the 31-amino-acid-
long fl-endorphin. The endorphins are derived from the post-
translational processing by specific enzymes of three larger
polypeptide products of gene transcription and translation:
pro-opiomelanocortin, pro-enkephalin, and pro-dynorphin.
Cannabinoids
The cannabinoids are a class of chemical substances found in
the Cannabis plant, the source of marijuana (dried buds and
leaves) and hashish (concentrated resinous exudate from the
buds and leaves). The primary psychoactive cannabinoid in
cannabis is delta-9-tetrahydrocannibinol, often referred to sim-
ply as THC. The cannabis plant is one of the most ancient
human cultivars, with medicinal use documented in Chinese
medical texts from more than 3000 years ago. Some botanists
recognize distinct species of Cannabis sativa, Cannabis indica,
and Cannabis rudeleris, while others believe cannabis has been
cultivated by humans for so long that no distinct wild-type
species remain. Whether they are considered different species
or not, certainly different varieties exist. The plant is believed
to have originated in central Asia and now grows throughout
the world.
The psychoactive effects of cannabis may include relaxa-
tion, sedation, intensification of thoughts and feelings, altera-
tions of perception, and increased appetite. Medicinally,
cannabis has been used as an analgesic, an anti-inflammatory,
a treatment for migraine headache, an antiemetic (to decrease
nausea), an anticonvulsant, a muscle relaxant, a treatment for
glaucoma, and an appetite stimulant. It was widely used as a
medicine prior to the initiation, in 1937, of repressive legisla-
tion by the US government. Recent years have witnessed a
resurgence of interest in the medicinal properties of cannabis
and consideration of new laws allowing resumption of legal
medical use and clinical research.
For many years, the neurochemical mechanism of action of
THC remained obscure. The last decade of the twentieth cen-
tury saw the discovery of specialized receptors responding to
THC (cannabinoid receptors). Cannabinoid receptors are
widespread in the human brain, as well as elsewhere in the
body, such as in the immune and endocrine systems. Thus far,
at least two endogenous compounds have been implicated as
neurotransmitters for the cannabinoid system. These are anan-
damide and 2-arachidonyl glycerol (2-AG), both of which are
long-chain, polyunsaturated compounds derived from mem-
brane lipids. The function of the cannabinoid signaling system
remains largely unknown, although it is likely to be involved in
things such as appetite, memory, and the perception of pain.
A very exciting aspect of endocannabinoid neurochemistry
is the notion that cannabinoid signaling at synapses is involved
in the modification of synaptic strength. These effects are
mediated by way of a novel mechanism that has been termed
retrograde signaling. In locations where this occurs - and these
locations appear to be pervasive throughout the brain -
conventional neurotransmission via the abundant excitatory
neurotransmitter glutamate may trigger the postsynaptic on-site
synthesis of endocannabinoids. These then diffuse into the
synaptic cleft and carry signal information in a retrograde
fashion back to the presynaptic neuron or to other nearby
neurons and glia cells. The effects of these endocannbinoid
transmitters at cannabinoid receptors may then alter the prop-
erties of the presynaptic neuron as well as other nearby cells.
Psychedelics or Hallucinogens
The psychedelics are among the most interesting and poorly
understood of all the psychoactive drugs. They produce a
variety of complex effects on the brain and mind, including
intensification of thoughts and feelings, alterations of sensory
perception, and loosening of psychological defenses. Reported
effects range from euphoria, therapeutic insight, and mystical
experiences, to anxiety and panic. The effects of psychedelic
drugs on behavior and mental experience, to a greater degree
than is the case for other drugs, are significantly influenced by
the mental state and the physical setting of the user.
As their properties came under investigation by twentieth-
century biomedical science, many of these compounds were
initially called psychotomimetics, because it was believed their
effects were much like the symptoms of psychosis. This term
was eventually eclipsed in favor of hallucinogen, highlighting
728 Drugs, the Brain, and Behavior
that effects may include alterations of perception, including
hallucinations. In 1956, the psychiatrist Humphry Osmond
(1917-2004), in correspondence with the author Aldous
Huxley (1894-1963), proposed the word 'psychedelic' (from
the Greek roots 'mind manifesting') to describe the unique
effects of these compounds on consciousness. In their plant
forms, psychedelics have been used by cultures around the
world for thousands of years to facilitate states of conscious-
ness conducive to mystical experiences. This has led to the use
of the word 'entheogen' (generating god within) to emphasize
their ritual use in spiritual settings.
Although it has not been found as a natural product
in plants or fungi and its profound psychoactive properties
were not known until it was synthesized and tested by Albert
Hofmann (1906-2008) in 1943, LSD is perhaps the most
widely known psychedelic chemical. It is one of the most potent
psychoactive substances known, active in quantities of a few
micrograms. Psychedelics identified from plants and fungi
include lysergic acid amide from morning glories (family Con-
volvulaceae), psilocybin and psilocin from a variety of mush-
rooms (many from the genus Psilocybe), dimethyltryptamine
(DMT) from a variety of plant species found throughout the
world, and mescaline from several species of cacti, one of
which is peyote (Lophophora williamsii). Numerous chemical
relatives of the compounds named above have been synthe-
sized, tested, and found to have psychedelic effects in humans.
This broad category of substances is sometimes additionally
delineated as the 'classical' psychedelics or hallucinogens, to
distinguish them from other substances discussed later that
also possess psychedelic and hallucinogenic effects.
The therapeutic utility of psychedelic plants and fungi has
been appreciated for centuries in cultures where shamanic
practitioners provided medical and psychotherapeutic care.
And prior to their being declared illegal in the United States,
these compounds were researched and utilized for their
psychotherapeutic properties. Clinical research during the
1950s-1960s found them to be of value in the treatment of
alcoholism and in psychotherapeutic work addressing death in
terminally ill patients. There is ample reason to believe that
when used with therapeutic intent in carefully controlled set-
tings, they can be of extraordinary value. Such use may exclude
some individuals having preexisting mental illness, as their
powerful effects on the brain and mind may exacerbate preex-
isting symptoms or tendencies toward psychopathology.
Neurochemically, psychedelics have been found to bind as
agonists to various serotonin receptor subtypes, especially
type-2A serotonin receptors. Psychedelics also bind to other
neurotransmitter receptors, including those for dopamine and
norepinephrine. The connection between their neurochemical
effects and their profound effects on mental function remains
largely obscure.
Methylenedioxymethamphetamine
Methylenedioxymethamphetamine (MDMA), popularly known
by the sheet name 'ecstasy,' and its close chemical relative
methylenedioxyamphetamine (MDA), are often considered
together with the psychedelic drugs. While MDMA does
have some of the mind-manifesting characteristics of the
psychedelics, its overall properties are distinct enough to
warrant a separate category. Chemically related to metham-
phetamine, MDMA produces CNS stimulation, euphoria, and
sympathetic nervous system stimulation. The methylenedioxy
moiety confers upon the molecule additional psychoactivity
that is psychedelic-like, characterized by intensification of
thoughts and feelings and enhanced feelings of connection
with others. For many individuals, there is a reduction in
anxiety and an enhanced ability to verbalize feelings. These
qualities led to the use of MDMA as an adjunct to psycho-
therapy prior to its being declared an illegal Schedule I sub-
stance in 1985, an act that followed media attention to its use
as a recreational drug in dance clubs. Though illegal, MDMA
continues to be used for both its psychotherapeutic and
recreational properties by some individuals. Recently it has
been the subject of clinical research for the treatment of
posttraumatic-stress disorder.
While MDMA has been found to interact with a number of
CNS neurochemical systems, the primary neurochemical
effect of MDMA appears to be the release of serotonin and
dopamine from nerve terminals into the synapse. Release is
not via the usual process of storage vesicle fusion with the cell
membrane, as occurs with a nerve impulse, but is by way
of leakage of neurotransmitter out of the cell through the
reuptake transporter. Animal studies have indicated that
MDMA may result in oxidative damage to serotonin nerve
terminals. The implications of these findings for human use
remain to be determined.
Nitrous Oxide
Nitrous oxide (N 2 0) is a gas (sometimes called 'laughing gas')
widely used in medicine and dentistry for its anesthetic
and analgesic properties. It is used by dentists as a sole anes-
thetic agent and is a component of the inhalation anesthesia
(together with potent volatile sedative-hypnotic general anes-
thetics) for major surgeries. In addition, in subanesthetic doses
it can produce a powerful psychedelic-like state of conscious-
ness. Its various psychoactive properties were documented in
1800 by the British chemist Humphry Davy (1778-1829).
Davy's work, published as a book of more than 300 pages in
length when he was only 2 1 years old, represented the first
detailed study of the effects of a defined chemical substance
on the human mind. Although nitrous oxide has been found to
interact as an antagonist at the N-methyl-D-aspartate (NMDA)-
type of glutamate receptor (a variety of ionotropic glutamate
receptor occurring throughout the brain), the mechanism of its
psychoactive effects remains obscure.
Dissociative Anesthetics
This class of psychoactive drugs consists of ketamine and PCP
(phenyl cyclohexyl piperidine or phencyclidine). These are
synthetic compounds introduced into medicine to produce
an anesthetic loss of bodily sensation without depressing the
vital functions of respiration and cardiovascular action, as do
the general anesthetics. While PCP is no longer used medically,
ketamine is used in both human and veterinary surgical
Drugs, the Brain, and Behavior 729
procedures. For human surgeries, it is perhaps used most often
in infants, the elderly, and in trauma cases, where depressing
vital functions maybe particularly dangerous. At subanesthetic
doses in humans, ketamine catalyzes a change of conscious-
ness having some hallucinogenic and psychedelic characteris-
tics and may be accompanied by a loss of body awareness.
Recent research has suggested that subanesthetic doses of
ketamine may have prolonged antidepressant effects, that is,
elevation of mood in persons suffering from clinical depres-
sion. The primary neurochemical effect thus far discovered
for these drugs is antagonist action at the NMDA-type gluta-
mate receptor.
Anticholinergics
These substances act as antagonists at muscarinic-type recep-
tors for the neurotransmitter acetylcholine. Such receptors are
found in the brain and in the parasympathetic branch of the
autonomic nervous system. Psychoactive effects can range
from feelings of disorientation to dreamlike alterations of con-
sciousness to powerful hallucinations. Effects on the auto-
nomic nervous system produce a cluster of symptoms that
include dry mouth, difficulty urinating, constipation, blurry
vision, and orthostatic hypotension. Hyoscyamine (named
after henbane), atropine (a mixture of d and l stereoisomers
of hyoscyamine, and named after Atropa belladonna), and sco-
polamine (named after Scopolia) are anticholinergics found in
a number of species of plants from the family Solanaceae.
These include A. belladonna (deadly nightshade), Hyoscyamus
niger (henbane), Mandragora officinarum (mandrake), Scopolia,
Brugmansia, Datura, and Brunfelsia.
Atropine, scopolamine, and tincture of belladonna are used
to treat intestinal motility problems such as diarrhea and irri-
table bowel syndrome. Several synthetic anticholinergic drugs
are used as adjunctive treatments for Parkinson's disease. Sev-
eral of the phenothiazine antipsychotics and tricyclic antide-
pressants (TCAs) described below have anticholinergic effects
in the brain and autonomic nervous system.
Salvia divinorum
This plant is put into a class of its own, as there are no other
substances that are known to have its peculiar effects on mental
experience. A member of the mint family, its Latin name liter-
ally means 'sage of divination,' and it has been used for cen-
turies by shamanic healers in southern Mexico (its only known
indigenous location) for ritual spiritual purposes. It was intro-
duced to the modern world when the shaman Maria Sabina
(1888-1985) told Gordon Wasson (1898-1986) about it in
the 1950s. Its effects are powerful and of a hallucinatory
nature, although of a very different sort from the classical
psychedelics or the anticholinergics. The historical use of the
plant was by way of chewing the leaves in a ritual context with
a shaman. This plant was unknown outside of a relatively
narrow ritual context and likely would have remained so with-
out the aid of the Internet. Contemporary use includes smok-
ing of the dried leaves or of concentrated psychoactive extracts.
The primary psychoactive molecular component has been
identified as salvinorin A, a terpenoid compound acting in
the brain as an agonist at kappa-opioid receptors.
Antipsychotics
In Indian ayurvedic medicine, from ancient times to the pres-
ent, extracts of the snakeroot plant, Rauwolfia serpentina, have
been used to treat symptoms of psychosis (troubling halluci-
nations and delusions). In the mid-twentieth century, the
chemical reserpine was isolated and identified from Rauwofia
serpentina and found to cause decreases in the activity of mono-
aminergic neurons using the neurotransmitters dopamine,
norepinephrine, and serotonin.
The first pharmaceutical antipsychotics were the phenothia-
zines such as chlorpromazine (Thorazine), introduced in the
1950s. Other phenothiazine antipsychotic drugs include per-
phenazine, thioridazine, trifluoperazine, and pimozide. Non-
phenothiazine antipsychotics introduced in the 1960s and
1970s include haloperidol (Haldol) and thiothixene (Navane).
The primary neurochemical effect of all these compounds is
to block various types of dopamine receptors. This is believed
to lead to a reduction of psychotic symptoms and also to the
production of various motor side effects, such as Parkinsonian-
like symptoms and movement dyskinesias.
More recently, several antipsychotic medications have been
introduced that have fewer motor side effects. These so-called
'atypical' antipsychotics include clozapine (Clozaril), risperi-
done (Risperdal), paliperidone (Invega), olanzapine (Zyprexa),
quetiapine (Seroquel), ziprasidone (Geodon), and aripipra-
zole (Abilify). In addition to being antagonists at dopamine
receptors, these drugs are also antagonists at type-2 serotonin
receptors. The serotonin-receptor antagonism is believed to com-
pensate for some of the dopamine-blocking effects in the motor
circuitry of the brain, thereby reducing problematic movement
symptoms. Many of the atypical antipsychotics have been asso-
ciated with endocrine-system effects related to problematic
weight gain and increased risk of diabetes.
Antidepressants
Extracts of the plant Saint John's Wort, Hypericum perforatum,
have been used for centuries in Europe for its antidepressant
effects. This plant also facilitates wound healing when prepara-
tions are used topically. Its healing properties were mentioned
in the ancient medical texts of Hippocrates, Pliny, and Galen.
The first pharmaceuticals developed and marketed specifi-
cally for their antidepressant effects were the monoamine oxi-
dase inhibitors (MAOIs), discovered in the 1950s. Examples of
MAOIs currently on the market are isocarboxazid, phenelzine,
tranylcypromine, and selegiline. Via inhibition of the enzyme
MAO, these compounds may produce enhanced activity in
circuits utilizing the neurotransmitters serotonin, norepineph-
rine, and dopamine.
A second class of contemporary pharmaceuticals developed
and marketed for their antidepressant effects are the TCAs,
so called because their molecular structures contain a three-
ringed central core. They include imipramine, amitriptyline,
730 Drugs, the Brain, and Behavior
desipramine, nortriptyline, doxepin, and clomipramine. TCAs
inhibit monoamine reuptake transporters, primarily for nor-
epinephrine and serotonin.
Newer-generation antidepressants include the serotonin
selective reuptake inhibitors (SSRIs), represented by fluoxetine
(Prozac), paroxetine (Paxil), sertraline (Zoloft), fluvoxamine
(Luvox), citalopram (Celexa), and escitalopram (Lexapro).
Other antidepressant drugs include trazodone (Desyrel), nefa-
zodone, bupropion (Wellbutrin and Zyban), venlafaxine
(Effexor), desvenlafaxine (Pristiq), mirtazapine (Remeron),
and duloxetine (Cymbalta). These later compounds interact
in various ways with monoamine neurotransmitter receptors
and reuptake transporters.
The prevailing hypothesis regarding antidepressant mecha-
nism is that some sort of change in serotonin and/or norepi-
nephrine synaptic chemistry underlies their clinical action, but
exactly what sort of change it is remains unclear. As these
neurotransmitters act on a variety of GPCRs, it is likely that
the ultimate antidepressant effects involve modulation of gene
transcription, resulting in production of nerve growth factors
and subsequent synaptic plasticity and neurogenesis.
Mood Stabilizers
These substances are used in the treatment of bipolar or manic-
depressive disorder. The most well-known such mood stabi-
lizer is lithium, administered as a salt that produces lithium
ion upon ingestion. The therapeutic effects of lithium in the
treatment of manic depression were discovered in 1949, and it
has been widely used as a medication in psychiatry for more
than half a century. The most robust neurochemical effect of
lithium thus far elucidated is its interference with G-protein-
coupled pathways using phosphatidylinositol as an intracellu-
lar messenger, thereby producing a damping effect on many
GPCR systems.
Several chemicals, first discovered for their antiseizure
properties and used to treat epilepsy and other seizure condi-
tions, are also used to stabilize mood in manic-depressive
disorder. These include carbamazepine (Tegretol), valproic
acid (Depakote), lamotrigine (Lamictal), and oxcarbazepine
(Trileptal). This connection between unstable mood and sei-
zures suggests that similar neuronal properties of hyperexcit-
ability may be involved in both conditions.
Drugs and the Brain
The functioning of the brain can be understood in part by way
of a symphony of complex chemical interactions. Psychoactive
drugs are believed to work by influencing the chemistry of the
brain, primarily by perturbing activity at synapses. This gives
rise to short-term behavioral effects such as relaxation, alert-
ness, euphoria, analgesia, sleep, and so forth, presumably
related to changes in activity in various brain networks.
Repeated perturbation of the brain's chemistry brings about
changes in the circuitry of the brain and results in lasting
changes in behavior: improved mood from antidepressants,
decreased psychosis from antipsychotics, symptoms of depen-
dence and addiction from a variety of drugs, movement dis-
orders from antipsychotics, and so forth. Changes in the
brain's circuitry presumably reflect the strengthening and
weakening of synapses through mechanisms such as changes
in the number of receptors, changes in the amount of neuro-
transmitter released, and modulation of intracellular biochem-
ical pathways activated by membrane receptors. Growth of
dendrites to form new synapses or deterioration of existing
synapses may also be involved. Many of these processes may
be regulated by calcium entering the cell via ionotropic recep-
tors or gene transcription modulated by way of GPCRs.
Withdrawal, Dependence, and Addiction
Changes in brain physiology resulting from repeated exposure
to psychoactive drugs may result in withdrawal symptoms
when use of the dmg is decreased or stopped. It is likely that
extended use of any psychoactive drug will bring about
homeostatic changes in brain physiology that result in symp-
toms of withdrawal if use of the drug is stopped. In some cases
the withdrawal symptoms will be clinically significant and in
other cases they may not. For example, repeated use of seda-
tive-hypnotics may produce rebound withdrawal symptoms
of arousal, anxiety, and insomnia, and in more severe cases,
hallucinations, delerium, and seizures. Situations of clini-
cally significant withdrawal are taken to indicate physical
dependence.
Some psychoactive dmgs, especially those that produce an
experience of pleasure, euphoria, or reduced anxiety, appear to
influence neural circuits in the limbic system of the brain.
These so-called reward circuits contribute to reinforcement of
behaviors important for survival, such as eating, drinking,
sexual activity, and finding shelter. Euphorigenic psychoactive
drugs may hijack these circuits and produce unusually intense
stimulation. The resulting homeostatic changes in the circuits
might then produce symptoms of withdrawal that include
cravings (intense desires) for the drug, dysphoria, anxiety,
and compulsive drug-seeking actions. Clinically significant
compulsive dmg use is termed addiction.
The reward circuits connect the ventral tegmentum with the
nucleus accumbens and the frontal cortex. The primary neuro-
transmitter in these circuits is dopamine. Various dmgs known
to be euphorigenic and potentially addiction-producing have
been shown to increase the release and activity of dopamine in
these circuits. While different psychoactive dmgs (sedatives,
stimulants, opioids, etc.) have a variety of different effects on
the brain and behavior, it is possible that their actions on the
dopaminergic reward circuits may be part of a common neuro-
chemical mechanism involved in addiction.
Whether or not an individual develops an addictive rela-
tionship with a dmg is in part dependent on the dmg's phar-
macology and its direct effects on brain neurochemistry. These
factors are modulated by genetic risk factors for addiction, as
well as psychological and sociocultural factors, such as avail-
ability of coping strategies for stress and extent of dmg use by
others in one's surroundings.
Drugs and the Mind
Let 'mind' refer to the collection of mental processes - including
thoughts, feelings, and perceptions - experienced by the
Drugs, the Brain, and Behavior 731
organism. Of such mental processes, one may be conscious
(aware) or unconscious. There is abundant evidence that the
mind is intimately associated with brain physiology: damage
to the brain results in specific changes to mental processes, and
chemical perturbations of the brain with psychoactive drugs
produce specific effects on mental processes. Exactly how mind
is related to brain physiology is one of the great unsolved
problems of psychology and neuroscience - the so-called
mind-body problem. Drugs that affect conscious awareness
and other qualities of the mind are powerful probes to the
mind-brain connection.
Pioneering neuroscientist, psychologist, and philosopher
William James (1842-1910), whose thinking on the nature
of the mind was profoundly influenced by experiences with
nitrous oxide, said more than a century ago, in words that are
every bit as relevant today:
One conclusion was forced upon my mind at the time and my
impression of its truth has since remained unshaken. It is that our
normal waking consciousness, rational consciousness as we call it, is
but one special type of consciousness, while all about it, parted from
it by the filmiest of screens, there lie potential forms of conscious-
ness entirely different. We may go through life without suspecting
their existence; but apply the requisite stimulus, and at a touch they
are there in all their completeness, definite types of mentality which
probably somewhere have their application and adaptation. No
account of the universe in its totality can be final which leaves these
other forms of consciousness quite disregarded. How to regard them
is the question
See also: Addictions and Adolescence; Alcohol: Psychosocial Effects.
Further Reading
Brunton L, Chabner B, and Knollman B (2010) Goodman & Gilman's The
Pharmacological Basis of Therapeutics. New York: McGraw-Hill.
Chevaleyre V, Takahashi KA, and Castillo PE (2006) Endocannabinoid-mediated
synaptic plasticity in the CNS. Annual Review of Neuroscience 29: 37-76.
Franks NP (2008) General anesthesia: From molecular targets to neuronal pathways of
sleep and arousal. Nature Reviews Neuroscience 9: 370-386.
Griffiths RR, Richards WA, McCann U, and Jesse R (2006) Psilocybin can occasion
mystical-type experiences having substantial and sustained personal meaning and
spiritual significance. Psychopharmacology 187: 268-283.
Hofmann A (1979) LSD, My Problem Child: Reflections on Sacred Drugs,
Mysticism, and Science. Sarasota, FL: Multidisciplinary Association for
Psychedelic Studies.
Hyman SE, Malenka RC, and Nestler EJ (2006) Neural mechanisms of addiction:
The role of reward-related learning and memory. Annual Review of Neuroscience
29: 565-598.
James W (1 902) The Varieties of Religious Experience : A Study in Human Nature
(The Gifford Lectures on Natural Philosophy and Religion, Delivered at the
University of Edinburgh, Scotland, in 1901-1902). London: Longmans, Green,
and Company.
Meyer JS and Quenzer LF (2005) Psychopharmacology: Drugs, the Brain, and Behavior.
Sunderland, MA: Sinauer.
Nichols DE (2004) Hallucinogens. Pharmacology & Therapeutics 101: 131-181.
Ott J (1 993) Pharmacotheon: Entheogenic Drugs, Their Plant Sources and History.
Kennewick, WA: Natural Products Company.
Perrine DM (1996) The Chemistry of Mind-Altering Drugs: History, Pharmacology, and
Cultural Context. Washington, DC: American Chemical Society.
Roth BL, et al. (2002) Salvinorin A: A potent naturally occurring nonnitrogenous kappa
opioid selective agonist. Proceedings of the National Academy of Sciences of the
United States of America 99: 11934-11939.
Snyder SH (2009) Neurotransmitters, receptors, and second messengers galore in
40 years. The Journal of Neuroscience 29: 12717-12721.
Zarate CA, et al. (2006) A randomized trial of an N-methyl-D-aspartate antagonist
in treatment-resistant major depression. Archives of General Psychiatry
63: 856-864.
Relevant Websites
www.csp.org - Council on Spiritual Practices.
www.dea.gov - Drug Enforcement Administration (DEA).
www.erowid.org - Erowid Information Vaults.
www.heffter.org - Heffter Research Institute.
www.maps.org - Multidisciplinary Association for Psychedelic Studies.
www.nida.nih.gov - National Institute on Drug Abuse (NIDA).
Dyslexia
M J Snowling, D C Gooch, and L M Henderson, University of York, York, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Causal modeling The use of statistical models to assess the
causes of a skill/disorder within a conceptual framework.
Comorbidity The tendency for two or more developmental
disorders to occur together.
Dyslexia A learning disorder primarily characterized by
reading delay and spelling difficulties, usually accompanied
by phonological deficits.
Heritability An index of the amount of variance in a
quantitative trait or skill that is due to genetic factors.
Phonological pathway A set of mappings between
orthography and phonology.
Poor comprehender Term given to describe a child who can
decode accurately but has specific difficult in understanding
what they read (also known as reading comprehension
impairment).
Semantic pathway A set of mappings between orthography,
semantics, and phonology.
Treatment nonresponders Term given to child who show
no or at least a poor response to intervention.
Reading is a uniquely human accomplishment, essential to
educational achievement, career prospects, and ultimately
adult well-being. It is therefore crucial to understand why
children have reading difficulties. Reading is not just a simple
matter of recognizing and understanding individual words as
our eyes pass over text. The purpose of reading is to understand
text. Being a successful reader requires building up a mental
representation of the text, a process that requires integration
of multiple sources of information, from lexical features
to event knowledge and prior experience. In developing this
skill, children make a transition from learning to read, to
reading to learn.
Models of Reading
The 'triangle model' provides a framework for understanding
the development of word recognition (Figure 1). According to
this model, the process of learning to read consists of creating
mappings between orthography (print) and phonology (speech
sounds; the phonological pathway), and between orthography
and phonology via semantics (the semantic pathway). The pho-
nological pathway is crucial for reading unfamiliar words which
can only be read by translating letter strings into phonological
forms (referred to as orthography-phonology mappings). The
semantic pathway, on the other hand, is important for accurate
and fluent reading and in English, for reading irregular and
inconsistent words (e.g., yacht), which cannot be read accu-
rately using orthography to phonology mappings. Early in
reading development, children rely heavily on the phonological
pathway but there is an increasing need, as development pro-
ceeds, for the semantic pathway to be used in order to develop
reading fluency and automatic word recognition.
The simple view of reading (see Figure 2) suggests that
reading is the product of two distinct but related processes,
which are essential for learning to read; word recognition and
language comprehension (i.e., understanding the message that
the print conveys). Children can vary in their aptitude on these
two dimensions and this model can be used to describe differ-
ent patterns or 'profiles' of performance. Recently, this model
has been adopted by the government in England to provide a
framework to guide the teaching of reading and also to help
identify children whose reading skills are developing out of
step, as illustrated by cases of dyslexia and reading comprehen-
sion impairment. Children who fall within either the upper or
lower left quadrants are those who have dyslexic difficulties
because of poor word recognition skills. Children with good
word reading but poor comprehension (lower right quadrant)
are often referred to as poor comprehenders. As reflected in the
dimensional nature of the simple view model of reading, both
dyslexia and reading comprehension impairment are best
thought of as continuum, rather than distinct categories with
clear cut-off points. Whilst dyslexia is well understood, research
into reading comprehension impairment is in its infancy.
What Is Dyslexia?
The Diagnostic and Statistical Manual of Mental Disorders
(DSM V) of the American Psychiatric Association plans to use
the term dyslexia rather than reading disorder (RD) which was
used previously. DSM V defines dyslexia as difficulties in the
"accuracy or fluency of reading that are not consistent with
the person's chronological age, educational opportunities,
or intellectual abilities" that "without accommodations, sig-
nificantly interferes with academic achievement or activities
of daily living that require these reading skills."
The definition of dyslexia used by the International Dyslexia
Association (IDA), on the other hand, highlights both the neu-
rological and cognitive underpinnings of the disorder as well as
its symptoms. The IDA defines dyslexia as "a specific learning
disability of neurological origin which is characterized by diffi-
culties with accurate and/or fluent word recognition and by poor
spelling and decoding abilities" resulting from "a deficit in the
phonological component of language that is often unexpected
in relation to other cognitive abilities and the provision of effec-
tive classroom instruction." In line with this, a recent govern-
ment review in England led by Sir Jim Rose also highlights
difficulties in phonological awareness, verbal memory, and ver-
bal processing speed as being characteristic of dyslexia.
732
Dyslexia 733
Figure 2 The simple view of reading.
The Discrepancy Definition
For years dyslexia has been considered as a specific reading
disability that is unexpected in relation to an individual's IQ.
However, the use of this IQ discrepancy definition of dyslexia
is controversial. Recent research evidence suggests that, at a
cognitive level of description, children with reading difficulties
of lower and higher IQ, share deficits in phonological proces-
sing and make equivalent responses to intervention pro-
grammes targeting phonological skills. Moreover, dyslexia
can occur across a wide range of intellectual abilities. The use
of the IQ discrepancy definition means that many children
who have significant reading difficulties that interfere with
daily living and scholastic achievement fail to fulfill diagnostic
criteria even though their reading is significantly below age
expectations. For these reasons, the DSM V does not make
any recommendations about using IQ discrepancy in the defi-
nition of dyslexia.
What Is Reading Comprehension Impairment?
Children referred to as 'poor comprehenders' present the mirror
image of dyslexia: they have difficulty understanding what they
read, despite having age appropriate word reading accuracy and
fluency, normal phonological skills and nonverbal ability. Thus,
poor comprehenders can read passages of text as accurately and
as fluenfiy as their peers but have significant difficulties when
questioned about the content of the text. Given the demands on
comprehension in primary and secondary school, poor com-
prehenders are at a significant disadvantage.
In the DSM V, reading comprehension impairment is
separated from dyslexia. Instead it is to be considered as a
communication disorder or a learning disability characterized
by impairments in both reading and listening comprehension
(in the absence of decoding difficulties). The precise definition
remains vague in comparison to dyslexia. However, the major-
ity of research studies adopt variants around the following
selection criteria for poor comprehenders:
• Reading comprehension below the average range on stan-
dardized tests (i.e., below a standard score of 85).
• Word and/or passage reading accuracy in the average range
on standardized tests (i.e., above a standard score of 90).
• A discrepancy between reading comprehension and word/
passage reading accuracy (e.g., one standard deviation in
standard scores or 1 year in age equivalent scores).
Prevalence
RD affects a sizeable minority of the child-population. Preva-
lence rates vary depending on the criteria used to define the
disorders, the measures used to assess reading skills and envi-
ronmental factors. Prevalence rates of dyslexia vary between
3% and 15% of school age children. Several epidemiological
studies have reported that dyslexia is only slightly more fre-
quent in males than in females (about 1.5:1). However, a
metareview including all such studies showed that dyslexia is
more frequently diagnosed in males than in females and stud-
ies of referred samples have found male to female rations as
high as 6:1. This overrepresentation of males in referred sam-
ples has often been attributed to a higher frequency of comor-
bid behavioral disorders, such as attention deficit hyperactivity
disorder (ADHD), in males with dyslexia than in females with
dyslexia. Fewer studies have investigated prevalence rates of
poor comprehenders, however, recent studies suggest that
this RD affects between 7.5% and 10% of primary school
children in unselected samples. In contrast to dyslexia, reading
comprehension impairment appears to affect similar numbers
of males and females. Of course dyslexia and reading compre-
hension impairment are best thought of as extremes of the
normal distribution of reading ability rather than distinct cate-
gories; hence prevalence rates depend on the arbitrary cut-off
points used to define the disorder.
Referral Routes and Assessment
It is important to consider the routes by which children with
dyslexia and poor comprehenders are referred for assessment
since these will account for the kinds of diagnosis and interven-
tions prescribed. Children with dyslexia are readily recognized
by teachers and parents; hence the most common referral route
is through educational professionals. Children with dyslexia are
also occasionally referred for assessment by optometrists, gen-
eral practitioners, occupational therapists, and speech language
therapists, particularly if the child has a comorbid condition.
In comparison, poor comprehenders often go unnoticed in the
classroom and are rare in clinically referred samples of children
with reading difficulties. However, this is likely to be a reflection
of referral bias and the fact that there are currently no estab-
lished referral routes for these children. Assessments for RDs are
734 Dyslexia
usually conducted by educational psychologists or chartered
research psychologists trained in educational assessment.
A comprehensive assessment of reading should comprise
measures of reading accuracy (reading aloud a list of single
words which increase in difficulty), decoding (reading aloud
nonsense words graded in difficulty), reading fluency (reading
a list of single words which increase in difficulty at speed),
prose reading and comprehension (passage reading followed
by an assessment of understanding). Spelling performance
should also be considered. Given the evidence that a phono-
logical deficit is central in dyslexia, assessments should include
tests of phonological skills; phonological awareness (e.g., pho-
neme deletion/isolation); phonological/verbal memory (e.g.,
nonword repetition or digit recall) and naming (e.g., Rapid
Automatized Naming (RAN)). Furthermore, difficulties in
reading comprehension are often accompanied by deficits
in broader oral language (OL), for example, semantics (vocab-
ulary) and grammar; hence it is also important to assess
these skills.
Causes of RDs
So far we have focused on the behavioral features of RDs; a
proper understanding of these difficulties requires consider-
ation of their causes A general 'causal modeling' framework
can be used to consider causation at biological, cognitive, and
behavioral levels, with environmental factors interacting at
each level. Figure 3 illustrates several factors thought to be
causally associated with dyslexia (to the left) and reading com-
prehension impairment (to the right). Within this model, solid
lines represent causal pathways that have been empirically
validated, while dotted lines represent putative causes which
are not yet proven.
Cognitive Explanations
Cognitive theories seek to explain the processes underlying
behavioral features of disorders. Indeed, cognition provides
an important bridge between biological and behavioral levels
of explanation; with respect to dyslexia, a great deal of research
during the past 40 years has focused on cognitive explanations
of the disorder.
Phonological Deficit of Dyslexia
An important cognitive theory of dyslexia proposes that a core
phonological deficit underlies the reading difficulties charac-
teristic of dyslexia. The phonological deficit hypothesis sug-
gests that a failure to develop adequate phonological skills
interferes with the learning of grapheme-phoneme correspon-
dences, which are critical in reading acquisition; within the
triangle model discussed earlier, poorly specified phonological
representations compromise the development of the 'phono-
logical pathway' to reading. A considerable amount of research
has found that children with dyslexia perform significantly
worse than typically developing children on measures of pho-
neme awareness, which requires the ability to make explicit
judgements about the sound structure of words. Unlike pho-
nological awareness, phonological processing (such as non-
word repetition, RAN, and paired-associate learning) requires
the use of speech and is free of metacognitive demands. Chil-
dren with dyslexia also perform poorly on measures of pho-
nological processing supporting the idea that they have
poorly specified phonological representations - difficulties
with which the brain codes phonology.
Further support for the phonological deficit hypothesis
comes from longitudinal studies which suggest that children's
early phonological skills and letter knowledge play a causal
Figure 3 Causes of dyslexia and reading comprehension impairment.
Dyslexia 735
role in their later reading development. Children at risk of
dyslexia, who go on to receive a diagnosis, display deficits in
phonological skills prior to reading instruction (at around age
4 years). These children also exhibit weaknesses in nonphono-
logical OL skills compared to children at risk of dyslexia who
do not go on to receive a diagnosis. However, mild deficits in
phonological skills are sometimes present in children at risk of
dyslexia but who do not fulfill diagnostic criteria. This suggests
that a phonological deficit could represent an 'endophenotype'
or risk marker of dyslexia that can be moderated by other risk
and protective factors, such that some of the 'at risk' children
develop dyslexia while others compensate and do not succumb
to reading problems.
Two factors that may potentially impede phonological
development are speech perception and auditory processing
skills. Research has aimed to establish whether a deficit in
the perception of speech sounds, possibly underpinned by a
broader weakness in auditory processing, may cause the pho-
nological deficit in dyslexia (Figure 4). However, deficits in
auditory processing are not characteristic of all children with
dyslexia and although there is some evidence that children
with dyslexia have mild impairments in speech perception,
this may be restricted to those who have broader OL impair-
ments. Longitudinal studies are needed to elucidate the causal
relationships among these skills.
Alternative Causal Theories of Dyslexia
The phonological deficit hypothesis is successful in accounting
for the reading failure, the nonword reading deficit and the
weak spelling skills, characteristic of individuals with dyslexia.
However, some researchers have identified a range of difficul-
ties associated with dyslexia not easily accounted for by this
theory. Proponents of the double-deficit hypothesis of dys-
lexia, suggest that some children with dyslexia have impair-
ments in phonology, others have problems in temporal
processing and perceptual speed and a third group has impair-
ments in both.
The magnocellular deficit hypothesis suggests that indivi-
duals with dyslexia have impairments in processing rapidly
presented auditory and visual stimuli resulting from impair-
ments in the magnocellular pathway in the brain. However,
there is little evidence that these skills predict variations in
reading in the typical population and not all studies have
replicated findings of magnocellular processing deficits in
Time
Figure 4 Figure showing the possible causal role of an auditory
perception deficit in the development of reading problems.
dyslexia; arguably such deficits may be associated with broader
neurological abnormality rather than playing a critical role in
the determination of reading difficulty.
Other researchers argue that the apparent link between
reading disability and timing deficits is associated with abnor-
mal cerebellar activation and functioning; this has been termed
the 'automatization deficit' or the 'cerebellar deficit' hypothe-
sis. This theory suggests that children with dyslexia have pro-
blems automatizing cognitive/motor skills and impaired time
perception - difficulties that originate from mild cerebellar
dysfunction. In support of this, the cerebellum has been impli-
cated in motor activities, balance, skill automatization and
adaptive learning, and in perceptual tasks that require the
precise representation of temporal information. However, it
is possible that cerebellar/ automaticity impairments may be
associated with comorbid conditions such as ADHD or devel-
opmental coordination disorder. Consistent with this hypoth-
esis, balance deficits have been found to be associated with
ADHD rather than dyslexia.
Cognitive Theories of Reading Comprehension Impairment
Reading comprehension is a complex skill and consequently
there are many potential routes for failure. Knowledge of word
meanings is critical for comprehension and poor comprehen-
ders have been reported to have difficulties on a range of tasks
that require access to a knowledge of vocabulary ( semantic
deficit hypothesis). Poor comprehenders are often slower and
less accurate when naming pictures with low-frequency names
and are poorer at making synonym judgements (e.g., when
asked to decide whether two words have a similar meaning)
particularly for more abstract words. Impairments in semantic
processing can also compromise some aspects of word recog-
nition in poor comprehenders. For example, they tend to be
poor at reading irregular and low-frequency words although
they are able to read regular and high-frequency words. It has
been argued that irregular and low-frequency words are harder
to read using the phonological and orthographic skills alone
and they require greater input from the 'semantic pathway' to
word recognition. Semantic processing impairments in poor
comprehenders should be considered in the context of
broader weaknesses in OL skills including listening compre-
hension, receptive and expressive vocabulary, oral expression,
narrative production, and grammar though phonological
skills are normal.
Beyond OL weaknesses, poor comprehenders also have
text level difficulties. Successful comprehension relies on
making inferences that go beyond what is written in the text
to construct an adequate and coherent representation. Infer-
ences are either 'cohesive' (i.e., they are necessary for compre-
hension and support text integration) or 'elaborative' (i.e., they
embellish information provided in the text). Poor comprehen-
ders have difficulties inferring specific meanings of nouns from
sentence context (e.g., inferring that 'fish' is most likely a
'shark' in the following sentence: 'The fish frightened the swim-
mer'). Poor comprehenders have also been shown to generate
fewer constructive inferences that require the reader to inte-
grate information from two different sources in text. How-
ever, it remains possible that poor inference-making in poor
comprehension is due to impairments in accessing relevant
736 Dyslexia
knowledge, maintaining motivation and a drive to read for
meaning, or due to a processing limitation (e.g., poor working
memory or the failure to inhibit irrelevant information).
Reading comprehension also relies on metacognitive strate-
gies such as comprehension monitoring (the ability to regulate
reading to the demands of the task and engage with the com-
prehension process). For example, if an individual realizes that
their comprehension is inadequate, they can take appropriate
steps, such as slowing their reading speed or rereading to rem-
edy the situation. There is evidence to suggest that poor com-
prehenders have weak metacognitive knowledge and/or are less
good at applying this knowledge during comprehension.
Considering the complex nature of reading comprehen-
sion, and the range of associated deficits, it seems likely that
multiple causal factors will explain reading comprehension
impairment. Poor comprehenders are a heterogeneous group
and no clear cognitive profile has emerged from the extant
research. Although many poor comprehenders have weak-
nesses in OL, a substantial proportion of the population do
not show such weaknesses and some poor comprehenders
show average or even good performance on measures of verbal
working memory, comprehension monitoring, and inference
making.
Biological Bases
It has been known for many years that reading skills are herita-
ble and therefore that reading difficulties run in families. Sev-
eral 'family-risk' studies have suggested that between 30% and
50% of children whose parents have dyslexia, will also develop
the disorder. However, families share genes and also environ-
ments so findings of familial aggregation of RDs alone are not
sufficient to demonstrate that reading skills are heritable. Twin
studies that compare the reading skills of monozygotic twins
(who share 100% of their genes) with dizygotic twins (who
share on average 50% of their genes) provide a more sensitive
design for separating genetic and environmental influences.
Such studies suggest that the concordance rates for dyslexia
are higher for monozygotic twins than for dizygotic twins,
confirming the genetic liability for dyslexia.
More recent research on the heritability of reading and its
difficulties has moved away from the use of a categorical 'diag-
nosis' of dyslexia to define reading as a continuous trait.
Within this view, quantitative genetic analyses confirm that
reading skills are significantly heritable across the distribution,
as are the skills which underlie them, including phonological
abilities and other language skills. Moreover, studies with
a longitudinal perspective have shown that there is shared
genetic variance between preschool phonological awareness
and early measures of reading, and also new sources of genetic
variance in reading which emerge after reading instruction
(perhaps contributing to orthographic learning). Findings
such as these illustrate that reading is underpinned by a range
of more primitive cognitive skills, many of which are also
heritable. However it needs to be borne in mind that genes
act through environments and individual differences in any
trait, and reading is no exception, also depend on environ-
mental factors. A recent behavior-genetic analysis assessed
the genetic and environmental influences on both initial
performance and growth in reading skills. While there were
significant genetic and environmental contributions to initial
reading level, some of the environmental influences on growth
in decoding ability were independent of initial performance. It
can be concluded that new environmental influences operate
on reading as it develops, over and above those that determine
initial performance; it is plausible that these are mediated by
the home and school environment and/or the child's personal
motivation to read.
Turning to reading comprehension, similar analyses, using
the quantitative approach, have shown that shared genetic
influences on word recognition and reading comprehension
are partly independent of shared genetic influences on listen-
ing and reading comprehension. Such findings, which are in
line with the simple view of reading, support the position that
individual difference in word recognition and reading compre-
hension are partly underpinned by different genetic etiologies.
Findings of family and twin studies showing the heritability
of RDs set the stage for molecular genetic approaches which
aim to identify genes that may be associated with the disorder.
Dyslexia, in common with other developmental cognitive dis-
orders is genetically heterogeneous and the risk of inheriting
dyslexia depends on the combined influence of many genes
as well as environmental influences. At the time of writing, a
number of gene markers and several candidate genes (on chro-
mosomes 6, 15, and 18) have consistently been associated with
dyslexia; moreover the same genes appear to operate across the
normal distribution underpinning component reading skills.
To date, the genes associated with individual differences in
word-level reading are different from those associated with
the language impairments implicated in reading comprehen-
sion impairments but further research is needed to investigate
these associations.
Genetic differences between good and poor readers are
likely to have an impact on the neurodevelopment in children
with dyslexia. Indeed there is considerable evidence that dys-
lexia is associated with structural and functional abnormalities
in various systems in the left hemisphere of the brain, particu-
larly structures within the temporal lobe which are involved in
language and learning to read. More specifically, studies have
shown disrupted temporoparietal brain responses during pho-
nological tasks and less left prefrontal brain response during
tasks that require the rapid processing of auditory stimuli in
both adults and children with dyslexia. However, in interpret-
ing such findings it is important to be aware that becoming
literate has an impact on brain structure and function, so some
of the effects observed may be a consequence rather than a
cause of dyslexia.
Findings from neuroimaging studies also provide some
support for alternative, sensory, theories of dyslexia. Support
for the magnocellular (visual) theory of dyslexia comes from
studies which have found abnormalities in the magnocellular
layers of the lateral and medial geniculate nuclei, in individuals
with dyslexia. Additionally, in support of the cerebellar deficit
hypothesis of dyslexia, studies have shown that the cerebellum,
an area of the brain that plays an important role in motor
control, is also implicated in cognitive processes involved in
reading and verbal memory. Furthermore, the cerebellum has
been found to be underactive in adults with dyslexia during
performance of a motor sequencing task. However, it is unclear
Dyslexia 737
how deficits in cerebellar function can cause the phonological
deficit which underpins dyslexia. Furthermore, the specificity
of impairments in the cerebellum to dyslexia is questionable
given that the cerebellum is also implicated in disorders such
as ADHD.
Far less attention has been devoted to understanding the
functional neuroanatomy of comprehension than of phonol-
ogy and word reading. However, studies using positron emis-
sion tomography (PET) have revealed increases in blood flow
to the left and right temporal poles during reading when stories
are compared with incomprehensible stories. Such regions
may therefore be involved in linking propositions to build a
mental representation of the text, and the anterior medial
parietal/posterior cingulate cortex has been associated with
linking incoming information with prior knowledge. Further-
more, neuropsychological studies have reported deficits fol-
lowing right brain damage in narrative comprehension and
semantic integration. Only one study to date has investigated
neuroanatomical abnormalities in poor comprehenders. This
study revealed that reading comprehension impairment is
associated with smaller cerebral volume and less leftward
asymmetry of the planum temporal, plana, and cerebellar
anterior lobe.
Environment
A range of environmental factors affect the development of
language and literacy. Significant effects of social and eco-
nomic circumstances typically associated with parental educa-
tion level, affect the language skills that children bring to the
task of learning and such differences may be compounded by
factors such as the amount of literacy related activities in the
home. Studies of home literacy environment have shown that
direct reading instruction in the home, focusing on letter
knowledge and print-related concepts, have positive effects
on children's development of decoding skills whereas language
activities surrounding shared book reading primarily affect
reading comprehension. All of these factors can be assumed
to affect the behavioral manifestation of dyslexia in the early
school years, and after school entry, different types of reading
instruction will also exert an influence on the extent to which
an early risk may translate into an identifiable RD. Beyond
reading instruction, it has been demonstrated that the amount
of reading practice a child undertakes affects reading develop-
ment such that good readers, who are more likely to enjoy
reading, practice more and gain further benefits (Matthew
effect). Contrary to this, for children with dyslexia, declining
levels of motivation may feed into low levels of print exposure
and a downward spiral of poor reading, low levels of literacy
and educational achievement (a powerful example of gene-
environment correlation is that children who inherit a risk of
reading difficulty select an environment not conducive to
reading development).
At the level of the 'macroenvironmenf the effects of culture
will also impinge on the child's emergent literacy skills.
Amongst these, a potent example is the language of instruction.
Languages vary in the transparency or consistency of the
orthography-phonology mappings as well as in the accessibility
of different phonological units. There is now a large body of
evidence showing that learning to read in an alphabetic system
is easier in more transparent languages and hence it has been
predicted that the behavioral manifestation of dyslexia varies
among languages. For example, problems of reading accuracy
are the primary sign of dyslexia in English whereas in Finnish
which is highly consistent, the main manifestation is a problem
of reading fluency. Notwithstanding this, the cognitive deficits
underpinning dyslexia appear to be universal among alphabetic
readers. Burgeoning research with readers of logographic (e.g.,
Chinese) and, alphasyllabic (e.g., Korean) languages indicates
that the predictors of individual differences in reading are
somewhat different and include morphological awareness.
Against this backdrop, the predictors of dyslexia may be differ-
ent in these languages.
Development
Clinical observation reveals that, although dyslexia is a persis-
tent disorder across the life-span, the primary signs of dyslexia
change with development, and longitudinal studies have
begun to elucidate the developmental trajectory of dyslexia
(Table 1).
Comorbidities
Comorbidity refers to the tendency for two or more develop-
mental disorders to occur together. Dyslexia frequently co-occurs
with difficulties in aspects of language, motor coordination,
mathematics, attention, and organization but these difficulties
Table 1 Developmental changes in the cognitive and behavioral
features of dyslexia
Developmental
phase
Behavioral signs of dyslexia
Cognitive features
consistent across
time
Preschool
Delayed or problematic
Poor verbal
speech
short-term memory
Poor expressive language
Slow naming of
familiar objects
and colors
Difficulty learning letters
Slow speed of verbal
processing
Early school
Poor letter sound
years
knowledge
Poor phoneme awareness
Poor decoding skills
Idiosyncratic spelling
Problems copying
Middle school
Slow reading
years
Poor decoding when faced
with new words
Poor spelling that is
phonetic or nonphonetic
Adolescence
Poor reading fluency
and
Slow speed of writing
adulthood
Poor organization and
expression in written work
Poor spelling
738 Dyslexia
by themselves are not thought to be markers of dyslexia. Rather,
they are likely to reflect the comorbidity between dyslexia
and other developmental disorders such as specific language
impairment (SLI), dyspraxia/developmental coordination dis-
order, dyscalculia, and ADHD. Both dyslexia and SLI have pho-
nological impairments at their core however children with SLI
also have additional difficulties with vocabulary and grammar.
It is debated whether the overlap between these two disorders
truly reflects comorbidity or whether SLI is a more severe form or
an earlier manifestation of dyslexia.
Difficulty with reading comprehension has been reported
in a number of different clinical disorders, including early onset
hydrocephalus, autism, nonverbal learning disorder, SLI, Turn-
er's syndrome, ADHD, pragmatic language impairment, and
William's syndrome. However, it is likely that the comprehen-
sion difficulties in these disorders have different manifesta-
tions. As with dyslexia it is tempting to draw parallels between
poor comprehenders and children with SLI. As outlined above,
poor comprehenders have weaknesses in spoken language sim-
ilar to children with SLI and there is a raised incidence of poor
reading comprehension in SLI. However, poor comprehenders
have clear strengths in phonological processing, which is an
area of deficit for many children with SLI. Given the reported
weaknesses in attention and working memory in poor compre-
henders there is also speculation as to whether poor compre-
henders have some form of ADHD but this remains to be
formally tested.
Although the extant research precludes strong conclusions,
Figure 5 illustrates the putative overlap of dyslexia and reading
comprehension impairment (PC) with ADHD and SLI.
Interventions
Interventions for dyslexia have been in use for many years and
there are accepted guidelines for good practice. However, it is
only recently that the effects of interventions have been eval-
uated rigorously using controlled designs.
In the ideal scenario of a virtuous circle, causal theories of
dyslexia lead directly to principled interventions and the effi-
cacy of these interventions in turn provide a powerful test of
the causal theories. Thus, starting from the premise that poor
readers have phonological deficits, if training in phonological
skills leads to improvements in reading skills, then this pro-
vides evidence for the causal theory. Findings from a large
number of studies now show that boosting phonological skills
is an effective intervention for poor readers provided it is
combined with training in letter-sound knowledge and linking
Figure 5 Possible comorbidities with dyslexia and reading
comprehension impairment.
the two. Arguably such interventions aim to improve the func-
tioning of the phonological pathway in the triangle model.
However, it is important to bear in mind that not all children
with dyslexia respond to such effective interventions to the
same degree. Current research is focusing on the characteris-
tics of these 'treatment resistors' and assessing the validity
of 'response to intervention' as a measure of the severity of
dyslexia.
Meanwhile, an unfortunate consequence of the poor re-
sponse of some children is to encourage parents to seek alterna-
tive therapies for dyslexia. Such interventions, such as the use
of colored filters or of motor programs, have a poor evidence-
base and are likely to impact comorbid features, such as visual
stress or dyspraxia, rather than the RD per se.
Very few studies have evaluated support for children with
reading comprehension impairment. Encouraging children to
use representational and transformational drawings to visualize
stories and answers to comprehension questions in their minds
has been shown to improve reading comprehension in poor
comprehenders compared with training that involves reading
and question answering. Studies that have trained children's
ability to make inferences have also been shown to be effective.
Such training has involved asking children to pick out words
from passages and explain how they contributed to the overall
meaning of sentences or stories, and asking them to generate
questions and predictions about the text. However, many of the
studies that have supported this position have methodological
weaknesses that render the results difficult to interpret.
Moving beyond text-level training, a recent randomized
controlled trial has evaluated the effectiveness of three multi-
componential training programs on reading comprehension
in poor comprehenders aged 8-9 years. The training programs
were designed to promote text level processing (TC), OL skills,
or a combination of the two (COM). The progress of children
in the intervention groups was compared against the progress
of a waiting control group. All intervention groups made sig-
nificant improvements in reading comprehension relative to
the control group immediately following training. Further-
more, 1 1 months after training, gains increased for the OL
group beyond the TC and COM groups. The OL and COM
groups also made significant progress in expressive vocabulary
and this was a mediator for these groups' improved reading
comprehension. These results provide strong support for a
causal relationship between OL and reading comprehension
and suggest that specific reading comprehension difficulties
reflect (at least partly) underlying OL weaknesses.
Progress has been made in recent years in understanding
both the nature of the difficulties experienced by children with
reading difficulties and the best methods for supporting their
development. Future research must endeavor to identify the
factors that influence response to intervention and to replicate
and extend previous findings in educationally realistic contexts
in order to determine their practicality outside research settings.
Future Directions
The field of research on RDs is rapidly expanding to encom-
pass studies of dyslexia in different languages, its genetic foun-
dations and neural correlates; against this backdrop, there is a
Dyslexia 739
growing evidence base regarding interventions. There is likely
to be major expansion in research on the neuroscience of
dyslexia in the coming decade, including investigation of the
neural markers of response to intervention. For this research
agenda to succeed it will be important to continue the
emerging trend away from considering dyslexia as a categorical
diagnosis to viewing it as a the culmination of a number of
different risk factors (or endophenotypes) with no clear cut
offs. Essentially then the threshold for diagnosis will become
free to vary according to the language, age and stage of learn-
ing, and the extent to which the difficulties affect everyday
functioning in the classroom or workplace. Arguably, progress
in the understanding of endophenotypes will depend upon the
findings of longitudinal studies, which investigate how risk
and protective factors conspire to produce or compensate for
the risk of dyslexia. In turn, such studies should improve our
understanding of both the comorbidities of dyslexia and the
broader phenotype and will be essential to progress in genetics
and neuroimaging. In contrast to dyslexia, much less is known
about the many causes of reading comprehension impairment
(of which dyslexia is of course one). Here the field is wide open
for longitudinal studies examining its nature and causes.
Finally, research on RDs and their interrelationships with
other developmental neurocognitive disorders is critical to
developing effective interventions. At the present time we
know a lot about interventions to develop decoding in the
early stages and a little about reading comprehension interven-
tions. However, the evidence base is lacking for those with
dyslexia who have moderately well developed skills but still
struggle to achieve their potential and for those with broader
language impairments. Since reading is fundamental to educa-
tional achievement, adult employment and mental capital,
this research agenda is well worth pursuing.
See also: Attention Deficit Hyperactivity Disorder; Autism and
Pervasive Developmental Disorders; Language Development;
Phonetics; Reading and Phonological Processing; Sentence
Processing.
Further Reading
Bishop DVM (2009) Genes, cognition and communication: Insights from
neurodevelopmental disorders. Annals of the New York Academy of Sciences
1156: 1-18.
Bishop DVM and Snowling MJ (2004) Developmental dyslexia and specific
language impairment: Same or different? Psychological Bulletin
130: 858-888.
Cain K (201 0) Reading Development and Difficulties. Oxford: BPS Blackwell.
Clarke PJ, Snowling MJ, Truelove E, and Hulme C (2010) Ameliorating children’s
reading comprehension difficulties: A randomised controlled trial. Psychological
Science 21: 1106-1116.
Gough PB and Tumner WE (1986) Decoding, reading and reading disability. Remedial
and Special Education 7: 6-10.
Hulme C and Snowling M (2009) Developmental Disorders of Language Learning and
Cognition. UK: Wiley-Blackwell.
Morton J and Frith U (1995) Causal modelling: A structural approach to developmental
psychopathology. In: Cicchetti D and Cohen DJ (eds.) Manual of Developmental
Psychopathology, pp. 357-390. New York: Wiley.
Muter V and Likierman H (2009) A Parents Guide to Dyslexia, Dyspraxia and Other
Learning Difficulties. London: Random House.
Pennington BF and Bishop DVM (2009) Relations among speech, language, and
reading disorders. Annual Review of Psychology 60: 283-306.
Rutter M, Bishop D, and Pine D, et al. (eds.) (2008) Ruttets Child and Adolescent
Psychiatry. Oxford: Blackwell.
Rutter M, Caspi A, Fergusson D, et al. (2004) Sex differences in developmental reading
disability: New findings from 4 epidemiological studies. Journal of the American
Medical Association 291: 2007-2012.
Schlaggar BL and Church JA (2009) Functional neuroimaging insights into the
developmental of skilled reading. Current Directions in Psychological Science
18: 21-26.
Snowling MJ (2009) Changing concepts of dyslexia: Nature, treatment and
co-morbidity. Journal of Child Psychology and Psychiatry November.
Virtual issue on-line http://www.wiley.com/bw/vi.asp?ref=0021-9630&site=1.
Snowling MJ and Hulme C (eds.) (2005) The Science of Reading: A Blackwell
Handbook, p. 661. Oxford: Blackwell.
Snowling MJ and Stackhouse J (eds.) (2006) Dyslexia, Speech and Language:
A Practitioners Handbook, 2nd edn. London: Wiley.
Relevant Websites
http ://www. interdys . 0 rg./FAQwhat is. htm - what is dyslexia.
http://publications.dcsf.gov.uk/ - Identifying and teaching children and young people
with dyslexia and literacy difficulties. An independent report from Sir Jim Rose to
the Secretary of State for Children, Schools and Families.
Encyclopedia of
Human
Behavior
Second Edition
Volume Two E-0
Encyclopedia of
HUMAN BEHAVIOR
SECOND EDITION
Volume 2
E-0
EDITORIAL ADVISORY BOARD
David M Buss
University of Texas, Austin, Texas, USA
Richard B Buxton
University of California, San Diego, California, USA
Nicholas Christenfeld
University of California, San Diego, California, USA
Orrin Devinsky
New York University, New York, New York, USA
Albert M Galaburda
Harvard Medical School, Boston, Massachusetts, USA
William Hirstein
Elmhurst College, Elmhurst, Illinois, USA
William G Iacono
University of Minnesota, Minneapolis, Minnesota, USA
John T Jost
New York University, New York, New York, USA
Jerome Kagan
Harvard University, Boston, Massachusetts, USA
Leif E O Kennair
Norwegian University of Science and Technology,
Trondheim, Norivay
David E Presti
University of California, Berkeley, California, USA
Encyclopedia of
HUMAN BEHAVIOR
SECOND EDITION
EDITOR-IN-CHIEF
V. S. RAMACHANDRAN
University of California, San Diego, California
Volume 2
E-0
ELSEVIER
AMSTERDAM BOSTON HEIDELBERG LONDON NEW YORK
PARIS SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY
Academic Press is an imprint of Elsevier
OXFORD
TOKYO
ACADEMIC
PRESS
Academic Press is an imprint of Elsevier
32 Jamestown Road, London NW1 7BY, UK
30 Corporate Drive, Suite 400, Burlington, MA 01803, USA
525 B Street, Suite 1900, San Diego, CA 92101-4495, USA
Second edition 2012
Copyright © 2012 Elsevier, Inc. All rights reserved.
The following articles are US Government works in the public domain and are not subject to copyright:
Creativity
Glial Cells
Military Psychology
Neuroimaging of Dementia
No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic,
mechanical, photocopying, recording or otherwise without the prior written permission of the publisher.
Permissions may be sought from Elsevier's Science & Technology Rights department in Oxford, UK: phone (+44) (0) 1865 843830;
fax (+44) (0) 1865 853333; email: permissions@elsevier.com
Alternatively you can submit your request online by visiting the Elsevier website at http://elsevier.com/locate/permissions and selecting
Obtaining permission to use Elsevier material.
Notice
No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability,
negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Control Number: 2012930155
ISBN: 978-0-12-375000-6
For information on all Elsevier publications
visit our website at elsevierdirect.com
Printed and bound in the United States of America
11 12 13 14 10 9 8 7 6 5 4 3 2 1
Working together to grow
libraries in developing countries
www.elsevier.com | www.bookaid.org | www.sabre.org
Editorial : Mica Haley, Anna Gebicka, Kate Mittell
Production : Justin Taylor
ELSEVIER
BOOK AID
International
Sabre Foundation
ABOUT THE EDITOR-IN-CHIEF
V.S. Ramachandran, MD, PhD., is Director of the Center for Brain and Cognition and professor with the Psychology
Department and the Neurosciences Program at the University of California, San Diego, and Adjunct Professor of
Biology at the Salk Institute. Ramachandran trained as a physician and obtained an MD from Stanley Medical
College and subsequently a PhD from Trinity College at the University of Cambridge, where he was elected a Senior
Rouse Ball Scholar. Ramachandran's early research was on visual perception, but he is best known for his work in
neurology.
He has received many honors and awards, including a fellowship from All Souls College, Oxford, an honorary
doctorate from Connecticut College, a gold medal from the Australian National University, the Ariens Kappers
Medal from the Royal Netherlands Academy of Sciences for landmark contributions in neuroscience, and the
presidential lecture award from the American Academy of Neurology. He is also a fellow of the Neurosciences Institute in La Jolla and a
fellow of the Institute for Advanced Studies in Behavioral Sciences at Stanford. He was invited by the BBC to give the Reith Lectures on 'The
Emerging Mind' in 2003 and is the first physician/ experimental psychologist to be given this honor since the series was begun by Bertrand
Russell in 1 949 - these lectures were subsequently published as A Brief Tour of Human Consciousness : From Impostor Poodles to Purple Numbers.
In 1995, he gave the Decade of the Brain Lecture at the 25th annual (silver jubilee) meeting of the Society for Neuroscience and more
recently, the inaugural keynote lecture at the Decade of the Brain conference held by NIMH at the Library of Congress and a public lecture at
the Getty Museum in Los Angeles. He also gave the first Hans Lucas Teuber lecture at MIT, the D.O. Hebb lecture at McGill, the Rudel-Moses
lecture at Columbia, the Dorcas Cumming (inaugural keynote) lecture at Cold Spring Harbor, the Raymond Adams neurology grand
rounds at Massachusetts General Hospital, Harvard, and the Jonas Salk memorial lecture, Salk Institute.
Ramachandran is a trustee for the San Diego Museum of Art and has lectured widely on art, visual perception, and the brain.
Ramachandran has published over 120 papers in scientific journals (including three invited review articles in Scientific American), coauthor
(with Sandra Blakesleee) of Phantoms in the Brain that has been translated into eight languages and formed the basis for a two-part series on
Channel Four TV in the UK and a one-hour PBS special in the United States. His work is featured frequently in the major news media
including BBC and PBS. Newsweek magazine recently named him a member of The Century Club,' one of the "hundred most prominent
people to watch in the next century. "
V
This page intentionally left blank
ABOUT THE EDITORIAL ADVISORY BOARD
David M. Buss received his BA from the University of Texas at Austin and his PhD from the
University of California at Berkeley. He served in professorial positions at Harvard University, the
University of Michigan, and the University of Texas, where he is currently professor of psychology.
David Buss received the American Psychological Association (APA) Distinguished Scientific Award
for Early Career Contribution to Psychology, the APA G. Stanley Hall Award, and the APA Distin-
guished Scientist Lecturer Award. The University of Texas awarded Buss the President's Associates
Teaching Excellence Award. He served as President of the Human Behavior and Evolution Society
(HBES). He is currently Head of the Individual Differences and Evolutionary Psychology Area of the
Psychology Department at the University of Texas at Austin.
Buss's books include The Evolution of Desire: Strategies of Hitman Mating (Basic Books; translated
into 11 languages); Evolutionary Psychology: The Neiv Science of the Mind (Allyn & Bacon); The
Dangerous Passion: Why Jealousy is as Necessary as Love and Sex (Free Press; translated into 13 languages);
The Murderer Next Door: Why the Mind is Designed to Kill (Penguin; translated into 12 languages); The
Handbook of Evolutionary Psychology (Wiley), for which he served as editor; W?ty Women Have Sex (Holt;
co-authored with Cindy Meston; translated into 16 languages); and The Evolution of Personality and
Individual Differences (Oxford University Press). Buss has more than 250 scientific publications to his
credit.
Buss has extensive cross-cultural research collaborations and lectures widely within the United
States and abroad. His primary research interests include the psychology of sex differences, human
sexuality, mating strategies, conflict between the sexes, why people kill, warfare, terrorism, stalking, and the psychology of prestige, status,
and reputation.
Richard B. Buxton received his BS (1976) and PhD (1981) degrees in physics from the
Massachusetts Institute of Technology, followed by a postdoc in biomedical imaging (PET and
MRI) at the Massachusetts General Hospital. He has been on the faculty of the Department of
Radiology at the University of California, San Diego, since 1990. Buxton's primary research is in
functional magnefic resonance imaging (fMRI), focused on understanding the connections between
neural activity, blood flow, and energy metabolism in the human brain. His experimental work
combines arterial spin labeling (ASL) methods with blood oxygenation-level dependent (BOLD)
methods to estimate changes in brain oxygen metabolism in response to a stimulus or a drug. His
more theoretical work involves mathematical modeling of the imaging methods, the BOLD effect,
and oxygen transport from blood to tissue. Buxton is based at the Center for fMRI and served as the
center's founding Director (2000-2007). He has written a textbook on fMRI, now in its second
edition, published by Cambridge University Press (2009).
VII
viii About the Editorial Advisory Board
Nicholas Christenfeld received a bachelor's degree in psychology and social relations from Harvard,
and a doctorate in social psychology from Columbia, and joined the faculty at the University of
California - San Diego and has remained there for two decades. His research program comprises
multiple, changing, only partly overlapping areas. He has addressed such topics as why some people
say 'um' so often, whether babies look like their mothers or their fathers, how we choose which box
of cereal to buy, why a baseball season is ten times as long as a football season, which parts of the
month carry the greatest risk of untimely death, whether the support of a woman is better for one's
blood pressure than that of a man, what sort of music might be useful in stress reduction, whether
some initials extend and some shorten the lives of their bearers, if people who live in, or even just
visit, New York City are at risk of heart attacks, and whether story spoilers do spoil stories. Currently,
Christenfeld is also exploring the role of forgiveness in poststress cardiovascular recovery, the impact
of crying on hormones, and the nature of a sense of humor. An overarching theme, if one exists,
might be the empirical exploration of the everyday phenomena of the world.
Orrin Devinsky is professor of neurology, neurosurgery, and psychiatry at New York University
(NYU). He directs the NYU Epilepsy Center and St. Barnabas Institute of Neurology. His research on
epilepsy includes phenomic-genomic relations, sudden unexpected death in epilepsy, surgical
therapies, new medicines and devices, quality of life, cognition and behavior, and neuroinflamma-
tion. His research on behavioral neurology includes hyperfamiliarity, delusions, autonomic nervous
function, and anterior cingulate cortex functions. He founded the organization Finding A Cure for
Epilepsy and Seizures and cofounded epilepsy.com and the Epilepsy Therapy Project. He serves on
the boards of these organizations and has served on the board of the American Epilepsy Society and
Epilepsy Foundation. His other interests include the history of neuropsychiatry, evolution, anthro-
pology, and animal intelligence.
Albert M. Galaburda is a cognitive neurologist and neuroscientist, specializing in developmental
cognitive disorders in adults and children. He received his medical degree from Boston University in
1971, trained in neurology at Harvard Medical School from 1973 to 1976, and began his research on
»|f“ dyslexia in 1979, when with neuropathologist Thomas Kemper, he reported for the first time minor
cortical malformations in the brain of an adult dyslexic who had died in an accident. Following that
original report, he published several other cases with similar malformations and launched a research
program modeling these malformations in rodent models, which continues to this date.
He uses RNAi and transgenic technologies in his present research to induce cortical malformations
h in rat and mouse brains, which in turn are tested anatomically, molecularly, and behaviorally in
collaboration with Glenn Rosen, Joseph LoTurco, and Holly Fitch, respectively. This latest research
I has shown that manipulation of dyslexia candidate gene homolog's I rodent embryos results in
7 molecular interference with neuronal migration, cell autonomous and noncell autonomous effects
on cortical development that mimic the changes seen in dyslexic brains, and behavioral changes
affecting auditory perception. Ex vivo imaging of the brains of genetically manipulated animals
discloses changes in cortical circuits affecting corticocortical and corticothalamic relationships.
Galaburda is presently the Emily Fisher Landau Professor of Neurology at Harvard Medical
School, and Head of the Cognitive Neurology Unit at Beth Israel Deaconess Medical Center in Boston.
He has published more than 150 scholarly articles on various topics of cognitive anatomy, cognitive
neurology, and dyslexia, as well as several books on related topics. Galaburda has received the Pattison Prize in Neuroscience, the Neuronal
Plasticity Prize from the IPSEN Foundation of France, Scientist of the Year from the Association for Children with Learning Disabilities
(now, Learning Disorders Association), and has been elected to membership in the American Neurological Association. He has delivered
the Rita Rudel Memorial Lecture, the Norman Geschwind Memorial Lecture, the Sylvio Conte Decade of the Brain Symposium Lecture, and
the Curt von Euler Memorial Lecture, among others. His research is funded by the Eunice Kennedy Shrive National Institute of Child Health
& Human Development. His teaching is funded by the National Institute of Neurological Disorders and Stroke.
About the Editorial Advisory Board ix
William Hirstein is Professor and Chair of the Philosophy Department at Elmhurst College,
Elmhurst, IL, USA. He received his PhD from the University of California, Davis, in 1994. He is
the author of several books, including On the Churchlands (Wadsworth Publishing, 2004), Brain
Fiction: Self-Deception and the Riddle of Confabulation (MIT Press, 2005), and Mindmelding : Conscious-
ness, Neuroscience, and the Mind's Privacy (Oxford University Press, 2012). His other interests include
autism, sociopathy, brain laterality, sense of self, and the misidentification syndromes.
William G. Iacono is Regents Professor of Psychology, Psychiatry, Law, and Neuroscience; Distin-
guished McKnight University Professor; Adjunct Professor of Child Development; and Codirector of
the Minnesota Center for Twin & Family Research at the University of Minnesota.
He studies the development of common mental disorders, focusing on the genetic liability for
substance abuse and antisocial behavior. He is past president of the Society for Psychophysiological
Research (SPR) . He has also received the Distinguished Contributions to Psychophysiology Award
from SPR for his lifetime contributions to this field. In addition, he has been awarded the Distin-
guished Scientist Award from the Society for a Science of Clinical Psychology. His research is funded
by grants from the National Institute of Mental Health, National Institute on Drug Abuse, and
National Institute on Alcoholism and Alcohol Abuse. He currently holds a National Institute of
Health MERIT (Method to Extend Research in Time) Award for his research achievement. He has
published over 350 scientific articles.
John T. Jost is Professor of Psychology at New York University. His research, which addresses
stereotyping, prejudice, political ideology, and system justification theory, has appeared in top
scientific journals and received national and international media attention. He has published over
90 articles and chapters and coedited four books: The Psychology of Legitimacy (2001), Perspectivism in
Social Psychology (2004), Political Psychology: Key Readings (2004), and Social and Psychological Bases of
Ideology and System Justification (2011). Jost has received numerous honors, including the Gordon
Allport Intergroup Relations Prize (three times), Erik Erikson Award for Early Career Research
Achievement in Political Psychology, International Society for Self and Identity Early Career
Award, Society for Personality and Social Psychology Theoretical Innovation Prize, Society of
Experimental Social Psychology Career Trajectory Award, and the Morton Deutsch Award for
Distinguished Scholarly and Practical Contributions to Social Justice.
x About the Editorial Advisory Board
Jerome Kagan is Emeritus Professor of Psychology at Harvard University. His research over the past
57 years has centered on the cognitive and emotional development of infants and children. During
the last 30 years, he and his colleagues have concentrated on the temperamental biases called high
and low reactivity in infants. He is a fellow of the American Academy of Arts and Sciences and a
member of the Institute of Medicine in the National Academy of Sciences. He has received distin-
guished scientist awards from the American Psychological Association, the Society for Research in
Child Development, and the Child-Mind Institute. Kagan received his PhD from Yale University in
1954 and taught at the Ohio State University and Antioch College before joining the Harvard faculty
in 1964.
Leif Edward Ottesen Kennair, PhD, is Head of Department at the Department of Psychology,
Norwegian University of Science and Technology, Trondheim, Norway. He graduated as a clinical
psychologist from the University of Bergen, 1999, and is a specialist of clinical adult psychology
(Norwegian Psychology Association), and a qualified cognitive therapist and supervisor of cogni-
tive-behavioral therapy (Norwegian Cognitive Therapy Association). He has worked as a clinical
psychologist/ chief psychologist at the Nordfjord Psychiatric Centre and as a researcher at St. Olav's
Hospital, Trondheim. His research interests span from mainstream clinical psychology efficacy research
(generalized anxiety disorder, OCD) to evolutionary psychology (evolutionary psychopathology, the
evolutionary psychology of war, mate choice and sexual behavior). Kennair's PhD thesis was on the
topic of evolutionary clinical psychology. He has written an introduction to evolutionary psychology,
which has been published in Norwegian and Danish. He lectures on evolutionary psychology,
behavioral genetics, and personality psychology and is a much-used lecturer for training psychologists
and psychiatrists on topics such as cognitive-behavior therapy and treatment of anxiety and OCD.
David E. Presti is a neurobiologist and cognitive scientist at the University of California in Berkeley,
where he has taught in the Department of Molecular and Cell Biology for more than 20 years. For
more than a decade ( 1990-2000) he also worked in the treatment of addiction and of posttraumatic
stress disorder (PTSD) at the Department of Veterans Affairs Medical Center in San Francisco. His
areas of expertise include the chemistry of the human nervous system, the effects of drugs on the
brain and the mind, the treatment of addiction, and the scientific study of the mind and the
consciousness. He has doctorates in molecular biology and biophysics from the California Institute
of Technology and in clinical psychology from the University of Oregon. Since 2004, he has also
been teaching neuroscience to Tibetan monks in India in a program initiated by the Dalai Lama. His
primary research interest is the relation between mentality, consciousness, and brain physiology, the
so-called mind-body problem.
REVIEWERS
We would like to thank the following for their valuable assistance in peer-reviewing the articles within the Encyclopedia of Human Behavior:
Mike Aamodt
maamodt@radford.edu
Donelson Forsyth
dforsyth@richmond.edu
Josh Ackerman
joshack@mit.edu
Willem E. Frankenhuis
wfrankenhuis @ gmail . com
Patrick Ackles
packles@elmhurst.edu
Robert Freidin
freidin@princeton.edu
Robert Allan
allanr@lafayette.edu
Alan Fridlund
fridlund@psych.ucsb.edu
Lisa Aspinwall
lisa.aspinwall@psych.utah.edu
Luciano Gasser
luciano.gasser@phz.ch
Elizabeth Austin
elizabeth.austin@ed.ac.uk
Itimar Gati
msgati@mscc.huji.ac.il
Efrain Azmitia
efrain.azmitia@nyu.edu
David Geary
gearyd@missouri.edu
James Beggan
james.beggan@louisville.edu
William Gerin
bgerinx@gmail.com
Stephen Benning
stephen.d.benning@vanderbilt.edu
Jennifer Gillis
jmg0001@auburn.edu
Anna Berti
annamaria.berti@unito.it
David Cleaves
david.gleaves@canterbury.ac.nz
Prasad Bestha
durgabestha@creighton.edu
Dan Grodner
dgrodnel @swarthmore.edu
David Bjorklund
dbjorklu@fau.edu
Edward Hagen
ehagen@gmail.com
Herbert Bless
bless @uni-mannheim. de
Roger Hagen
roger.hagen@svt.ntnu.no
Jason Bocarro
j ason_bocarro @ncsu.edu
James Hampton
j.a.hampton@city.ac.uk
Robert Bomstein
bornstein@adelphi.edu
Curt Haugtvedt
haugtved @fisher.osu.edu
Jos Bosch
j.a.bosch@bham.ac.uk
Marie Helweg-Larsen
helwegm @ dickinson. edu
Bruce Bridgeman
bruceb@ucsc.edu
Ralph Hood
ralph-hood@utc.edu
Richard Brislin
rbrislin@hawaii.edu
John Hunsley
john.hunsley@uottawa.ca
Lllrike Buhlmann
ulrike.buhlmann@hu-berlin.de
William Ickes
ickes@uta.edu
William Bukowski
william. bukowski@concordia.ca
Nancy Jackson
nancy- j ackson @ uiowa .edu
Josh Burk
jabur2@wm.edu
Rebecca Jackson
bjackson@pgsp.edu
Michael Burton
mlburton@uci.edu
Kali Johnson
klj03@fsu.edu
Shawn Charlton
srcharlton@mac.com
Michael Kavsek
kavsek@uni-bonn.de
Nishimura Chiaki
Susan Kemper
skemper@ku.edu
Jason Clark
jasonanthonyclark@gmail.com
Monica Harris Kern
harris@pop.uky.edu
Taya Cohen
tcohen@cmu.edu
Ashley King
akingl@email.arizona.edu
Michael Cole
mcole@ucsd.edu
Gerjo Kok
g.kok@maastrichtuniversity.nl
Dennis R. Combs
dcombs@uttyler.edu
Nate Kornell
nate.kornell@williams.edu
Karen Cook
kcook@stanford.edu
Megan Kozak
mkozak@pace.edu
Joel Cooper
jcoops@princeton.edu
Howard Krishner
hkirshner@aol.com
Katja Riiter Corcoran
katja.corcoran@uni-koeln.de
Peter Kuppens
peter, kuppens @ psy .kuleuven .be
Alita Cousins
cousinsa@easternct.edu
David Kutzik
dkutzik@aol.com
Arthur Cropley
acropley@dodo.com.au
Peter Lamal
palamal@uncc.edu
Walter Crozier
r.crozier@uea.ac.uk
Richard Lapan
lapan@educ.umass.edu
Martin Daly
daly@mcmaster.ca
Daniel Lapsley
dlapslel@nd.edu
Maayan Davidov
mdavidov@mscc.huji.ac.il
Steve Lee
stevelee@psych.ucla.edu
Hasker Davis
hdavis@uccs.edu
Antoon Leenaars
draalee@sympatico.ca
Susanne Denham
sdenham@gmu.edu
Daniel S. Levine
levine@uta.edu
Brent Donnellan
donnel59@msu.edu
Debra Lieberman
debra@miami.edu
David Dozois
ddozois@uwo.ca
Scott Liening
sliening@mail. utexas . edu
Steve Duck
steve-duck@uiowa.edu
Scott Lilienfeld
slilien@emory.edu
Roger Dunn
rdunn@sunstroke.sdsu.edu
Wolfgang Linden
wlinden@psych.ubc.ca
Paul Eastwick
eastwick@tamu.edu
John Loehlin
loehlin@psy.utexas.edu
David W. Evans
dwevans @bucknell.edu
Brian Lowery
lowery_brian @ gsb . Stanford .edu
Gary W. Evans
gwe 1 @ Cornell . edu
Mark Lumley
mlumley@wayne.edu
Terry Falcomata
falcomata@austin.utexas.edu
Robert Maiden
fmaiden@alfred.edu
Maggie Faulkner
mfaulkner@towson.edu
Greg Maio
maio@cardiff.ac.uk
Anwei Feng
eds808@bangor.ac.uk
Robert Mair
rgm@unh.edu
Gereon Fink
g.r.fmk@fz-juelich.de
Gaurav Mathur
gaurav.mathur@gallaudet.edu
XI
xii Reviewers
Susan Matt
smatt@weber.edu
Nicolas Rothen
beat. meier@psy.unibe. ch
Michael D. Matthews
mike.matthews@usma.edu
Charles Sanislow
csanislow@wesleyan.edu
lim Mcgough
jmcgough@mednet.ucla.edu
David Sarwer
dsarwer@mail.med.upenn.edu
David Mcneill
dmcneill @ uchicago . edu
Geoffrey Schoenbaum
schoenbg@schoenbaumlab.org
Elizabeth Meadows
meadolea@cmich.edu
David Schroeder
dave@uark.edu
Jeff Miller
millerjeff@duq.edu
Donald Shankweiler
donald.shankweiler@uconn.edu
Paul Mills
pmills@ucsd.edu
David Share
dshare@constmct.haifa.ac.il
Ian Mitroff
imitroff@alliant.edu
Katrina L. Sifferd
sifferdk@elmhurst.edu
Evan Murray
edmurray@partners.org
Barry Smith
bdsmith@psyc.umd.edu
lordi Navarra
jordi.navarra@gmail.com
Peter Smith
p.smith@gold.ac.uk
Bernard Nijstad
b.a.nijstad@mg.nl
John Smythies
smythies@psy.ucsd.edu
Kathleen O'Connor
kmo8@cornell.edu
Hugo Spiers
h.spiers@ucl.ac.uk
Christin Ogle
christin.ogle@duke.edu
Jerry Suls
jerry-suls@uiowa.edu
Tom Ollendick
tho@vt.edu
Stephan Swinnen
stephan.swinnen@faber.kuleuven.be
Dick Olson
rolson@psych.colorado.edu
Ross Thompson
rathompson@ucdavis.edu
Robert O'Neill
roneill@ed.utah.edu
Geoffrey Thorpe
geoffrey_thorpe@umit.maine.edu
Joel Paris
joel.paris@mcgill.ca
Bmce Thyer
bthyer@fsu.edu
Gilbert Parra
gparra@memphis.edu
Elliot Tucker-Drob
tuckerdrob@psy.utexas.edu
Ty Partridge
tpartrid@wayne.edu
Brent Turvey
bturvey@corpus-deliai.com
Del Paulhus
dpaulhus@psych.ubc.ca
Michael Vanyukov
mmv@pitt.edu
Keith Payne
payne@unc.edu
Joel Voss
joelvoss@illinois.edu
Reinhard Pekrun
pekmn@lmu.de
Mark v an Vugt
m.van.vugt@psy.vu.nl
Carin Perilloux
perilloux@mail.utexas.edu
Rich Walker
walkerr@wssu.edu
John Christopher Perry
jchristopher.perry@mcgill.ca
Colleen Ward
colleen.ward@vuw.ac.nz
Chris Peterson
chrispet@umich.edu
Duane Watson
dgwatson@gmail.com
Elizabeth Pomery
elizabeth.pomery@yale.edu
William Watson
william_watson@urmc.rochester.edu
Radek Ptak
radek.ptak@hcuge.ch
John Welte
welte @ ria. buffalo .edu
Kisha Radliff
radliff. 2 @ osu. edu
Robb Wilier
willer@berkeley.edu
Gabriel Radvansky
gabriel.a.radvansky. 1 @nd.edu
Gwen Wittenbaum
gwittenb @ msu . edu
Ranga Ramanujam
ranga @ owen . vanderb ilt.edu
Steve Wonderlich
swonderlich@nrifargo.com
Sara Rix
srix@aarp.org
Jenny Yiend
jenny.yiend@kcl.ac.uk
Richard Roberts
rroberts@ets.org
Tom Zentall
zentall@email.uky.edu
Kelly Rohan
kelly . rohan @ uvm . edu
CONTRIBUTORS
P C G Adam
The University of New South Wales, Sydney, NSW,
Australia
M Allemand
University of Zurich, Zurich, Switzerland
M W Allen
California State Polytechnic University, Pomona, CA,
USA
S Allison
University of Richmond, Richmond, VA, USA
A al-Nowaihi
University of Leicester, Leicester, UK
S M Andersen
New York University, New York, NY, USA
K M Antshel
State University of New York - Upstate Medical
University, New York, NY, USA
L Asarian
University of Zurich, Zurich, Switzerland
J R Averill
University of Massachusetts, Amherst, MA, USA
E C Azmitia
New York University, New York, NY, USA
B Bailer
Temple University, Philadelphia, PA, USA
J M Bailey
Northwestern University, Evanston, IL, USA
D S Bailis
University of Manitoba, Winnipeg, MB, Canada
M D Baker
East Carolina University, Greenville, NC, USA
J D Baldwin
University of California at Santa Barbara, Santa Barbara,
CA, USA
J I Baldwin
University of California at Santa Barbara, Santa Barbara,
CA, USA
J Barden
Howard University, Washington, DC, USA
E E Bartolini
Wesleyan University, Middletown, CT, USA
P Bartolomeo
Hopital Salpetriere, Paris, France; Catholic University,
Milan, Italy
M R Basso
University of Tulsa, Tulsa, OK, USA
A Baum '
The University of Texas at Arlington, Arlington, TX, USA
1 B Becker
Zina Pitcher Place, Ann Arbor, MI, USA
1 Beggan
University of Louisville, Louisville, KY, USA
D C Beidel
University of Central Florida, Orlando, FL, USA
D R Beike
University of Arkansas, Fayetteville, AR, USA
S Bell
University of Saskatchewan, Saskatoon, SK, Canada
S Bensmaia
University of Chicago, Chicago, IL, USA
L Bensman
Hilbert College, Hamburg, NY, USA
G G Berntson
Ohio State University, Columbus, OH, USA
B Berry
University of Nebraska - Lincoln, Lincoln, NE, USA
1 Deceased.
Xiii
xiv Contributors
M E Berryhill
University of Nevada, Reno, NV, USA
A Berti
University of Torino, Torino, Italy
D P Bestha
Carolinas Healthcare System, Charlotte, NC, USA
A M Betz
Florida Institute of Technology, Melbourne, FL, USA
R S Bhatt
University of Kentucky, Lexington, KY, USA
T Bjorgvinsson
McLean Hospital, Belmont, MA, USA
C J Black
University of Arizona, Tucson, AZ, USA
S Blum
University of California, Irvine, CA, USA
L Bobadilla
Western Carolina University, Cullowhee, NC, USA
P C Borchgrevink
St. Olavs University Hospital, Trondheim, Norivay;
Norwegian University of Science and Technology,
Trondheim, Norway
E Borgida
University of Minnesota, Minneapolis, MN, USA
M J Boyle
La Salle University, Philadelphia, PA, USA
M M Bradley
University of Florida, Gainesville, FL, USA
B Bridgeman
University of California, Santa Cruz, CA, USA
S F Brosnan
Georgia State University, Atlanta, GA, USA
M Brow
University of California, Irvine, CA, USA
K R Brown
Houston OCD Program, Houston, TX, USA
R T Brown
Wayne State University, Detroit, MI, USA
D Bub
University of Victoria, Victoria, BC, Canada
C L Buchanan
The Children’s Hospital of Philadelphia, Philadelphia,
PA, USA
K A Bujarski
Dartmouth Medical College, Lebanon, NH, USA
P Bull
University of York, Heslington, York, UK
D C Burdick
The Richard Stockton College of New Jersey, Galloway,
NJ, USA
M M Burke
Vanderbilt University, Nashville, TN, USA
N L Burke
University of South Florida, Tampa, FL, USA
J Burrus
Center for Academic and Workplace Readiness and
Success, Princeton, NJ, USA
S A Burt
Michigan State University, East Lansing, MI, USA
B J Bushman
The Ohio State University, Columbus, OH, USA; VU
University, Amsterdam, The Netherlands
L M Bylsma
University of South Florida, Tampa, FL, USA
J T Cacioppo
University of Chicago, Chicago, IL, USA
D D Cahn
State University of New York at New Paltz, New Paltz,
NY, USA
E H Callahan
Institute for Child Development, SUNY at Binghamton,
Binghamton, NY, USA
M C Capozzoli
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
E Cardena
Lund University, Lund, Sweden
P J Carnevale
University of Southern California, Los Angeles, CA, USA
P Case
Saint Andrews Presbyterian College, Laurinburg, NC,
USA
S E Cassin
Sunnybrook Health Sciences Centre and University of
Toronto, Toronto, ON, Canada
L Castro
University of Iowa, Iowa City, I A, USA
E P Charles
Pennsylvania State University, Altoona, PA, USA
S R Charlton
University of Central Arkansas, Conway, AR, USA
Contributors xv
V Chauhan
New School for Social Research, New York, NY, USA
E Chemerinski
Mount Sinai School of Medicine, New York, NY, USA;
fames J. Peters Veterans Affairs Medical Center, Bronx,
NY, USA
C F Chick
Cornell University, Ithaca, NY, USA
J G Chipperfield
University of Manitoba, Winnipeg, MB, Canada
S D Christman
University of Toledo, Toledo, OH, USA
T Chung
Pittsburgh Adolescent Alcohol Research Center,
Pittsburgh, PA, USA
L A Clark
University of Notre Dame, Notre Dame, IN, USA
E G Clary
Kutztown University, Kutztown, PA, USA
A Collins
University of Nebraska - Lincoln, Lincoln, NE, USA
P J Colombo
Tulane University, New Orleans, LA, USA
D R Combs
University of Texas at Tyler, Tyler, TX, USA
D M Condon
Northwestern University, Evanston, IL, USA
M Conrad
University of Illinois, Chicago, IL, USA
K S Cook
Stanford University, Stanford, CA, USA
P R Corlett
Yale University, New Haven, CT, USA
H Coskun
Abant Izzet Baysal University, Golkoy Bolu, Turkey
K Cote
Brock University, St. Catharines, ON, Canada
N Cowan
University of Missouri, Columbia, MO, USA
J Craner
University of Maine, Orono, ME, USA
M E Crawley
Tulane University, New Orleans, LA, USA
T D Crespi
The University of Hartford, West Hartford, CT, USA
A Culebras
Upstate Medical University, Syracuse, NY, USA;
Community General Hospital, Syracuse, NY, USA
C L Culpepper
University of Tennessee, Knoxville, TN, USA
] A Cummings
Zina Pitcher Place, Ann Arbor, MI, USA
R C Curtis
Adelphi University and W.A. White Institute, New York,
NY, USA
V Curtis
London School of Hygiene and Tropical Medicine,
London, UK
B P Daly
Drexel University, Philadelphia, PA, USA
J M Daly
Temple University, Philadelphia, PA, USA
M P Daly
Temple University, Philadelphia, PA, USA
J E Davidson
Lewis & Clark College, Portland, OR, USA
H P Davis
University of Colorado, Colorado Springs, CO, USA
A T DeMarco
Temple University, Philadelphia, PA, USA
M D'Esposito
University of California at Berkeley, Berkeley, CA, USA
J B F de Wit
The University of New South Wales, Sydney, NSW,
Australia
L de-Wit
University of Leuven (KU. Leuven), Leuven, Belgium
S Dhami
University of Leicester, Leicester, UK
C Di Dio
University of Parma, Parma, Italy
K S Dobson
University of Calgary, Calgary, AB, Canada
J Domingues
University of Connecticut, Storrs, CT, USA
M B Donnellan
Michigan State University, East Lansing, MI, USA
R L Doty
University of Pennsylvania School of Medicine,
Philadelphia, PA, USA
xvi Contributors
A L Dougall
The University of Texas at Arlington, Arlington, TX, USA
V Drago
IRCCS Centro San Giovanni di Dio Fatebenefratelli,
Brescia, Italy
S Duck
University of Iowa, Iowa City, I A, USA
J Duntley
The Richard Stockton College of New Jersey, Pomona, NJ,
USA
A B Durawa
Warsaw School of Social Sciences and Humanities,
Wroclaw, Poland
B S East Jr.
Tulane University, New Orleans, LA, USA
R J Edelmann
University of Roehampton, London, UK
D M Eichen
Temple University, Philadelphia, PA, USA
B J Ellis
University of Arizona, Tucson, AZ, USA
E K Englander
Bridgewater State University, Bridgewater, MA, USA
J T Enns
University of British Columbia, Vancouver, BC, Canada
A M Epp
University of Calgary, Calgary, AB, Canada
A Erdemir
Vanderbilt University, Nashville, TN, USA
T Ertelt
University of North Dakota, Fargo, ND, USA
EMR Escobar
York University, Toronto, ON, Canada
J Ey e
Lehigh University, Bethlehem, PA, USA
H J Eysenck'
Institute of Psychiatry, London, UK
S B G Eysenck
Institute of Psychiatry, London, UK
M Fabbri-Destro
Universita di Parma and IIT Brain Center for Motor and
Social Cognition, Parma, Italy
^Deceased.
A Fang
Boston University, Boston, MA, USA
E Fantino
University of California, San Diego, La Jolla, CA, USA
S V Faraone
State University of New York - Upstate Medical
University, New York, NY, USA
B Fehr
University of Winnipeg, Winnipeg, MB, Canada
A Fenigstein
Kenyon College, Gambier, OH, USA
R D Fields
National Institutes of Health (NICHD), Bethesda, MD,
USA
A J Figueredo
University of Arizona, Tucson, AZ, USA
D Fiset
Universite du Quebec en Outaouais, Gatineau, QC,
Canada
P Fisher
University of Liverpool, Liverpool, UK
W A Fisher
University of Western Ontario, London, ON, Canada
E B Foa
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
K A Force
Ryerson University, Toronto, ON, Canada
E A Fors
St. Olavs University Hospital, Trondheim, Norway;
Norwegian University of Science and Technology,
Trondheim, Norivay
J L Freeman
Forensic Solutions LLC, Sitka, AK, USA
A Freitas-Magalhaes
University Fernando Pessoa Health Sciences School,
Porto, Portugal
H S Friedman
University of California, Riverside, CA, USA
G O Gabbard
Baylor College of Medicine, Houston, TX, USA
A Gallace
Universita di Milano-Bicocca, Milan, Italy
M W Gallagher
University of Kansas, Lawrence, KS, USA
Contributors xvii
V Gallese
University of Parma, Parma, Italy
E Gambrill
University of California, Berkeley, CA, USA
G Ganis
University of Plymouth, Plymouth, UK; Harvard Medical
School, Boston, MA, USA
C Garbacz
Flora, MS, USA
R A Garcia
University of Arizona, Tucson, AZ, USA
Y Gazes
Columbia University Medical Center, New York, NY,
USA
D C Geary
University of Missouri, Columbia, MO, USA
N Geary
Institute of Food, Nutrition and Health, Schwerzenbach,
Switzerland
K Gee
University of Sheffield, Sheffield, UK
M Gelfand
University of Maryland, College Park, MD, USA
D Gentner
Northwestern University, Evanston, IL, USA
A Gerbasi
California State University, Northridge, CA, USA
W Gerin
The Pennsylvania State University, Pennsylvania, PA,
USA
F X Gibbons
Dartmouth College, Hanover, NH, USA
R Gifford
University of Victoria, Victoria, BC, Canada
A L Gilchrist
University of Missouri, Columbia, MO, USA
) M Gillis
Auburn University, Auburn, AL, USA
G Gillund
College of Wooster, Wooster, OH, USA
E L Glisky
University of Arizona, Tucson, AZ, USA
VGloy
Institute of Food, Nutrition and Health, Schwerzenbach,
Switzerland
D A Goldman
Cornell University, Ithaca, NY, USA
E B Goldsmith
Florida State University, Tallahassee, FL, USA
D C Gooch
University of York, York, UK
M Goodman
University of California, Davis, CA, USA
R Goodvin
Western Washington University, Bellingham, WA, USA
SCC Graham
Friedrich-Alexander University, Erlangen, Germany
J E Grant
University of Minnesota Medical Center, Minneapolis,
MN, USA
S L Gray
University of Tennessee, Knoxville, TN, USA
B Greyson
University of Virginia, Charlottesville, VA, USA
M M Griffin
Vanderbilt University, Nashville, TN, USA
S Grossberg
Boston University, Boston, MA, USA
R Gupta
University of Iowa, Iowa City, IA, USA
C A Gutner
Boston University, Boston, MA, USA
R Hagen
Norwegian University of Science and Technology,
Trondheim, Norway
P W Halligan
Cardiff University, Cardiff, UK
P A Hancock
University of Central Florida, Orlando, FL, USA
F Happe
King's College, London, UK
E Harmon-Jones
Texas A&M University, College Station, TX, USA
C R Harris
University of California, San Diego, CA, USA
J H Harvey
The University of Iowa, Iowa City, IA, USA
E Hatfield
University of Hawaii at Manoa, Honolulu, HI, USA
xviii Contributors
B Hayslip Jr.
University of North Texas, Denton, TX, USA
J A Headley
The University of Akron, Akron, OH, USA
K M Heilman
University of Florida, Gainesville, FL, USA
K A Henderson
North Carolina State University, Raleigh, NC, USA
L M Henderson
University of York, York, UK
A R Hill
Tulane University, New Orleans, LA, USA
O Hjemdal
Norwegian University of Science and Technology,
Trondheim, Norway
R M Hodapp
Vanderbilt University, Nashville, TN, USA
S Hoeffler
Vanderbilt University, Nashville, TN, USA
C Hoelscher
University of Freiburg, Freiburg, Germany
S G Hofmann
Boston University, Boston, MA, USA
A E Holmes
Baylor University, Waco, TX, USA
R W Hood Jr.
University of Tennessee at Chattanooga, Chattanooga,
TN, USA
V Hoorens
Katholieke Universiteit Leuven, Leuven, Belgium
S Horning
University of Colorado, Colorado Springs, CO, USA
C Hunt
University of Minnesota, Minneapolis, MN, USA
E R Igou
University of Limerick, Co. Limerick, Ireland
J Z Ingber
Fordham University, New York, NY, USA
R Ingram
University of Kansas, Lawrence, KS, USA
W J Jacobs
University of Arizona, Tucson, AZ, USA
L E James
University of Colorado at Colorado Springs, Colorado
Springs, CO, USA
B L Janzen
University of Saskatchewan, Saskatoon, SK, Canada
J A Jimenez
University of California, San Diego, La Jolla, CA, USA
J Johnson
York University, Toronto, ON, Canada
P K Jonason
University of South Alabama, Mobile, AL, USA
D Jones
University of Utah, Salt Lake City, UT, USA
K Jones
University of Nevada, Reno, NV, USA
A C Jordan
University of Arizona, Tucson, AZ, USA
R A Josephs
The University of Texas at Austin, Austin, TX, USA
K W Jung
Asan Medical Center, Seoul, South Korea
J Kagan
Harvard University, Cambridge, MA, USA
S J Karau
Southern Illinois University, Carbondale, IL, USA
F R Kardes
University of Cincinnati, Cincinnati, OH, USA
J D Kassel
University of Illinois, Chicago, IL, USA
A Kayser
University of California at San Francisco, Emeryville, CA,
USA
AKeil
University of Florida, Gainesville, FL, USA
E F Kelly
University of Virginia, Charlottesville, VA, USA
E W Kelly
University of Virginia, Charlottesville, VA, USA
I W Kelly
University of Saskatchewan, Saskatoon, SK, Canada
P C Kendall
Temple University, Philadelphia, PA, USA
R V Kendrick
University of Tennessee, Knoxville, TN, USA
LEO Kennair
Department of Psychology, Norwegian University of
Science and Technology, Trondheim, Norway
Contributors xix
D T Kenrick
Arizona State University, Tempe, AZ, USA
J Kenworthy
The University of Texas at Arlington, Arlington, TX, USA
M L Kern
University of California, Riverside, CA, USA
) F Kihlstrom
University of California, Berkeley, CA, USA
D R Kille
University of Waterloo, Waterloo, ON, Canada
Y K Kim
University of Southern California, Los Angeles, CA, USA
Y Y Kim
University of Oklahoma, Norman, OK, USA
F A A Kingdom
McGill University, Montreal, QC, Canada
H S Kirshner
Vanderbilt University School of Medicine, Nashville, TN,
USA
S Kitayama
University of Michigan, Ann Arbor, MI, USA
A M Klahr
Michigan State University, East Lansing, MI, USA
S J Ko
San Diego State University, San Diego, CA, USA
T Kohut
University of Western Ontario, London, ON, Canada
A Kordik
Friedrich-Alexander University, Erlangen, Germany
J K Kreuz
The University of Akron, Akron, OH, USA
R F Krueger
University of Minnesota, Minneapolis, MN, USA
K Krysinska
University of Queensland, Brisbane, QLD, Australia
R Kurzban
University of Pennsylvania, Philadelphia, PA, USA
D Laible
Lehigh University, Bethlehem, PA, USA
Y Lampl
Wolfson Medical Center, Holon, Israel; Tel Aviv
University, Holon, Israel
D K Lapsley
University of Notre Dame, Notre Dame, IN, USA
H Lasnik
The University of Maryland, College Park, MD, USA
B Latzko
University of Leipzig, Leipzig, Germany
J H Laurence
Temple University, Philadelphia, PA, USA
H Lavine
State University of New York - Stony Brook, Stony Brook,
NY, USA
M J Lawler
University of South Dakota, Vermillion, SD, USA
E-K Lee
University of Illinois, Urbana-Champaign, IL, USA
B T Legesse
McLean Hospital, Belmont, MA, USA
A M Legg
University of California, Riverside, CA, USA
R Lehmann
University of Zurich, Zurich, Switzerland
A F Lemieux
Emory University, Atlanta, GA, USA; Purchase College -
State University of New York, Purchase, NY, USA; and
National Consortium for the Study of Terrorism and
Responses to Terrorism, University of Maryland, College
Park, MD, USA
J G Lenz
Florida State University, Tallahassee, EL, USA
S J Lepore
Temple University, Philadelphia, PA, USA
D Lester
The Richard Stockton College ofNeiv Jersey, Pomona, NJ,
USA
A LeUnes
Texas A&M University, College Station, TX, USA
D S Levine
The University of Texas at Arlington, Arlington, TX, USA
N P Li
Singapore Management University, Singapore
YJ Li
Arizona State University, Tempe, AZ, USA
S H Liening
The University of Texas at Austin, Austin, TX, USA
S O Lilienfeld
Emory University, Atlanta, GA, USA
xx Contributors
D A Lishner
University of Wisconsin - Oshkosh, Oshkosh, WI, USA
J A Litman
University of South Florida, St. Petersburg, FL, USA
J H Liu
Victoria University of Wellington, Wellington,
New Zealand
K Lorduy
The University of Texas at Arlington, Arlington, TX, USA
A Luszczynska
University of Colorado, Colorado Springs, CO, USA;
Warsaw School of Social Sciences and Humanities,
Wroclaw, Poland
S J Lynn
Binghamton University, Binghamton, NY, USA
C N Macrae
University of Aberdeen, Aberdeen, UK
V Madaan
University of Virginia Health System, Charlottesville, VA,
USA
D Maestripieri
The University of Chicago, Chicago, IL, USA
R G Mair
University of New Hampshire, Durham, NH, USA
T Malti
University of Toronto Mississauga, Mississauga, ON,
Canada
L Manfredi
University of Chicago, Chicago, IL, USA
J Marino
University of North Dakota, Fargo, ND, USA
P F Martelli
University of California, Berkeley, CA, USA
D M Marx
San Diego State University, San Diego, CA, USA
J C Masdeu
New York Medical College, Valhalla, New York, NY,
USA
R Mata
University of Basel, Basel, Switzerland
K J Mathewson
McMaster University, Hamilton, ON, Canada
G Matthews
University of Cincinnati, Cincinnati, OH, USA
M D Matthews
United States Military Academy, West Point, NY, USA
R E Mayer
University of California, Santa Barbara, CA, USA
K C McCulloch
Idaho State University, Pocatello, ID, USA
K L McDonald
University of Alabama, Tuscaloosa, AL, USA
M McGrath
Unity Health System, Rochester, NY, USA
R E McGrath
Fairleigh Dickinson University, Teaneck, NJ, USA
W A McKim
Memorial University of Newfoundland, St. John's, NL,
Canada
D W McNeil
West Virginia University, Morgantown, WV, USA
E A Meadows
Central Michigan University, Mount Pleasant, MI, USA
P H Mehta
The University of Oregon, Eugene, OR, USA
F Mesa
University of Central Florida, Orlando, FL, USA
R G Meyer
University of Louisville, Louisville, KY, USA
W-U Meyer
University of Bielefeld, Bielefeld, Germany
J G Miller
New School for Social Research, New York, NY, USA
P J Mills
University of California, San Diego, La Jolla, CA, USA
N Minniti
Temple University, Philadelphia, PA, USA
V Miskovic
McMaster University, Hamilton, ON, Canada
J E Mitchell
University of North Dakota, Fargo, ND, USA
CMok
Palo Alto University, Palo Alto, CA, USA
T A Morgan
University of Iowa, Iowa City, I A, USA
A Morin
Mount Royal University, Calgaiy, AB, Canada
Contributors xxi
E D Murray
McLean Hospital, Belmont, MA, USA
M P Mychailyszyn
Temple University, Philadelphia, PA, USA
S-J Myung
Asan Medical Center, Seoul, South Korea
C T Nagoshi
Arizona State University, Tempe, AZ, USA
K Naragon-Gainey
University of Iowa, Iowa City, I A, USA
G Narasimham
Vanderbilt University, Nashville, TN, USA
M Natishyn
The University of Texas at Arlington, Arlington, TX, USA
N Nayak
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
M Neppi-Modona
University of Torino, Torino, Italy
K Neville
Northwestern University, Chicago, IL, USA
A B Newberg
University of Pennsylvania, Philadelphia, PA, USA
M Niepel
University of Bielefeld, Bielefeld, Germany
M M Nieves
University of Central Florida, Orlando, FL, USA
M Nikolas
University of Washington School of Medicine/Seattle
Children's Hospital, Seattle, WA, USA
G J Norman
Ohio State University, Columbus, OH, USA
P L Nunez
Tulane University, New Orleans, LA, USA
S Oberle
University of Colorado at Colorado Springs, Colorado
Springs, CO, USA
E H O'Brien
University of Michigan, Ann Arbor, MI, USA
P J O'Brien
Alberta Health Services, Calgary, AB, Canada
B L Odlaug
University of Minnesota Medical Center, Minneapolis,
MN, USA
L A O'Leary
RHAM High School, Hebron, CT, USA
M A Olson
University of Tennessee, Knoxville, TN, USA
J Omarzu
Loras College, Dubuque, IA, USA
S H Osipow
The Ohio State University, Columbus, OH, USA
W F Overton
Temple University, Philadelphia, PA, USA
M E Owens
University of Central Florida, Orlando, FL, USA
D J Packer
Lehigh University, Bethlehem, PA, USA
K S Page
University of North Texas, Denton, TX, USA
V Panaite
University of South Florida, Tampa, FL, USA
S T Pardo
Cornell University, Ithaca, NY, USA
J Paris
McGill University, Montreal, QC, Canada;
Institute of Community and Family Psychiatry,
Montreal, QC, Canada
J Park
University of Michigan, Ann Arbor, MI, USA
A Parker
Gothenburg University, Gothenburg, Sweden
C Parkinson
Dartmouth College, Hanover, MA, USA
A E Parrish
Georgia State University, Atlanta, GA, USA
T Partridge
Wayne State University, Detroit, MI, USA
J Pascual-Leone
York University, Toronto, ON, Canada
M L Patterson
University of Missouri, St. Louis, MO, USA
J H Patton
Baylor University, Waco, TX, USA
P B Paulus
The University of Texas at Arlington, Arlington, TX, USA
A D Pellegrini
University of Minnesota /Twin Cities Campus,
Minneapolis, MN, USA
xxii Contributors
A N Perry
Zina Pitcher Place, Ann Arbor, MI, USA
R P Perry
University of Manitoba, Winnipeg, MB, Canada
D Peterson
University of Nevada, Reno, NV, USA
G W Peterson
Florida State University, Tallahassee, FL, USA
J D Pettigrew
University of Queensland, St Lucia, QLD, Australia
R E Petty
Ohio State University, Columbus, OH, USA
D Paez
Universidad del Pais Vasco, San Sebastian, Spain
M V Pezzo
University of South Florida, St. Petersburg, FL, USA
B O Ploog
College of Staten Island, City University of New York,
Staten Island, NY, USA
E A Pomery
Yale University, New Haven, CT, USA
M E Portman
Private Practice, Jerusalem, Israel
S S Posavac
Vanderbilt University, Nashville, TN, USA
F C Power
University of Notre Dame, Notre Dame, IN, USA
M B Powers
Center for the Treatment and Study of Anxiety,
Philadelphia, PA, USA
D E Presti
University of California, Berkeley, CA, USA
B El Price
McLean Hospital, Belmont, MA, USA
D Pusch
Sheldon Chumir Centre, Calgary, AB, Canada
J E Pyers
Wellesley College, Wellesley, MA, USA
S Quadflieg
Catholic University of Louvain, Louvain-La-Neuve,
Belgium
P C Quinn
University of Delaware, Newark, DE, USA
J S Rabin
Towson University, Towson, MD, USA
M Rabinowitz
Fordham University, New York, NY, USA
E Race
Boston University School of Medicine, Boston, MA, USA
F Rachal
Baylor College of Medicine, Houston, TX, USA
El Rachlin
Stony Brook University, Stony Brook, NY, USA
G A Radvansky
University of Notre Dame, Notre Dame, IN, USA
A J Raines
Columbia University, New York, NY, USA
J W Ransom
University of Arkansas, Fayetteville, AR, USA
R L Rapson
University of Hawaii at Manoa, Honolulu, HI, USA
N A Rector
Sunnybrook Health Sciences Centre and University of
Toronto, Toronto, ON, Canada
S Reicher
University of St. Andrews, Fife, UK
R Reisenzein
University of Greifswald, Greifswald, Germany
W Revel le
Northwestern University, Evanston, IL, USA
V F Reyna
Cornell University, Ithaca, NY, USA
R B Ricco
California State University, San Bernardino, CA, USA
B J Ries
Minnesota State University, Mankato, MN, USA
J J Rieser
Vanderbilt University, Nashville, TN, USA
R R Riffle
Wright State University, Dayton, OH, USA
J H Riskind
George Mason University, Fairfax, VA, USA
G Rizzolatti
Universita di Parma and IIT Brain Center for Motor and
Social Cognition, Parma, Italy
El S Roane
SUNY Upstate Medical University, Syracuse, NY, USA
K El Roberts
University of California, Berkeley, CA, USA
Contributors xxiii
R D Roberts
Center for Academic and Workplace Readiness and
Success, Philadelphia, PA, USA
R W Robins
University of California, Davis, CA, USA
J L Rodriguez
Central Michigan University, Mount Pleasant, MI, USA
R G Romanczyk
Institute for Child Development, SUNY at Binghamton,
Binghamton, NY, USA
R Rosenthal
University of California, Riverside, CA, USA
M Roth
Auburn University, Auburn, AL, USA
P Roussos
Mount Sinai School of Medicine, New York, NY, USA
A G Rosch
Friedrich-Alexander University, Erlangen, Germany
K H Rubin
University of Maryland, College Park, MD, USA
M A Runco
University of Georgia, Athens, GA, USA
S W Russell
Northwestern University, Chicago, IL, USA
D H Saklofske
University of Western Ontario, London, ON, Canada
C Salmon
University of Redlands, Redlands, CA, USA
C A Sanislow
Wesleyan University, Middletown, CT, USA
B A Sarb
Western Washington University, Bellingham, WA, USA
S A Saribay
Bogazici University, Bebek, Istanbul, Turkey
A Scarpa
Virginia Tech, Blacksburg, VA, USA
L M Schaefer
University of South Florida, Tampa, FL, USA
H E Schendan
Plymouth University, Plymouth, UK
G L Schlomer
University of Arizona, Tucson, AZ, USA
L A Schmidt
McMaster University, Hamilton, ON, Canada
T R Schneider
Wright State University, Dayton, OH, USA
A Schnider
Hopitaux Universitaires de Geneve, Geneva, Switzerland
S M Schueller
University of California, San Francisco, CA, USA
O C Schultheiss
Friedrich-Alexander University, Erlangen, Germany
J A Sefcek
Kent State University, Ashtabula, OH, USA
R A Sevcik
Georgia State University, Atlanta, GA, USA
M Sevlever
Auburn University, Auburn, AL, USA
D Shankweiler
Haskins Laboratories, New Haven, CT, USA; University
of Connecticut, Storrs, CT, USA
D Shannahoff-Khalsa
University of California San Diego, La Jolla, CA, USA
K J Sher
The University of Missouri and the Midwest Alcoholism
Research Center, Columbia, MO, USA
T F Shipley
Temple University, Philadelphia, PA, USA
L J Siever
Mount Sinai School of Medicine, New York, NY, USA;
James J. Peters Veterans Affairs Medical Center, Bronx,
NY, USA
K L Sifferd
Elmhurst College, Elmhurst, IL, USA
S T Sigmon
University of Maine, Orono, ME, USA
R C Silver
University of California, Irvine, CA, USA
J Simner
University of Edinburgh, Edinburgh, UK
M Simonescu
State University of New York - Upstate Medical
University, New York, NY, USA
J D Sinnott
Towson University, Towson, MD, USA
J J Skowronski
Northern Illinois University, DeKalb, IL, USA
J C Smith
Roosevelt University, Chicago, IL, USA
xxiv Contributors
L Smith
Northwestern University, Evanston, IL, USA
P K Smith
University of California, San Diego, CA, USA
S F Smith
Vanderbilt University, Nashville, TN, USA
M Smithson
Australian National University, Canberra, ACT,
Australia
P Snell
Bridgewater State University, Bridgewater, MA, USA
M J Snowling
University of York, York, UK
KVSobel
University of Central Arkansas, Conway, AR, USA
A Soldan
University of Massachusetts - Dartmouth, North
Dartmouth, MA, USA
R Sommer
University of California, Davis, CA, USA
S C South
Purdue University, West Lafayette, IN, USA
C Spence
University of Oxford, Oxford, UK
C Spencer
University of Sheffield, Sheffield, UK
M R Sperling
Jefferson Medical College, Philadelphia, PA, USA
H J Spiers
University College London, London, UK
B Spinath
Heidelberg University, Heidelberg, Germany
B Spring
Northwestern University, Chicago, IL, USA
R Srinivasan
University of California, Irvine, CA, USA
S Stack
Wayne State University, Detroit, MI, USA
M A Stanley
Veterans Affairs Health Services Research and
Development Center of Excellence, Baylor College of
Medicine, Houston, TX, USA
P N Stearns
George Mason University, Fairfax, VA, USA
G A Stefanatos
Temple University, Philadelphia, PA, USA
K Steffen
University of North Dakota, Fargo, ND, USA
R M Stelmack
University of Ottawa, Ottawa, ON, Canada
Y Stern
Columbia University Medical Center, New York, NY,
USA
R J Sternberg
Oklahoma State University, Stillwater, OK, USA
H L Sterns
The University of Akron, Akron, OH, USA
T L Stewart
University of Manitoba, Winnipeg, MB, Canada
P C Stey
University of Notre Dame, Notre Dame, IN, USA
T C Stiles
St. Olavs University Hospital, Trondheim, Norway;
Norwegian University of Science and Technology,
Trondheim, Norway
M K Stimmler
University of California, Berkeley, CA, USA
R Stinchfield
University of Minnesota Medical School, Minneapolis,
MN, USA
M L Stock
The George Washington University, Washington, DC,
USA
E L Stocks
University of Texas at Tyler, Tyler, TX, USA
S Stolarz-Fantino
University of California, San Diego, La Jolla, CA, USA
A Strasser
Technical University, Munich, Germany
M Stroebe
Utrecht University, Utrecht, The Netherlands
W Stroebe
Utrecht University, Utrecht, The Netherlands
G L Strosser
Southern Utah University, Cedar City, UT, USA
A A Stukas
La Trobe University, Bundoora, VIC, Australia
J N Swanson
The University of Texas at Arlington, Arlington, TX, USA
Contributors xxv
S M Swearer
University of Nebraska - Lincoln, Lincoln, NE, USA
K Sweeny
University of California, Riverside, CA, USA
S P Swinnen
Katholieke Universiteit Leuven, Leuven, Belgium
E B Talbot
University of South Dakota, Vermillion, SD, USA
A K Tamplin
University of Notre Dame, Notre Dame, IN, USA
R Tanoue
University of Nevada, Reno, NV, USA
J Taylor
Florida State University, Tallahassee, FL, USA
M Thayer
San Diego State University, San Diego, CA, USA
J K Thompson
University of South Florida, Tampa, FL, USA
L A Thompson
New Mexico State University, Las Cruces, NM, USA
R A Thompson
University of California, Davis, CA, USA
G L Thorpe
University of Maine, Orono, ME, USA
S J Tobin
University of Queensland, St Lucia, QLD, Australia
D Tranel
University of Ioiva, Iowa City, I A, USA
K Treadwell
University of Connecticut, Storrs, CT, USA
L A Trevino
The University of Texas at Arlington, Arlington, TX, USA
H C Triandis
University of Illinois, Champaign, IL, USA
B Tuller
National Science Foundation, Arlington, VA, USA
C L Turk
Washburn University, Topeka, KS, USA
B E Turvey
Forensic Solutions LLC, Sitka, AK, USA
M L Uhelski
The University of Texas at Arlington, Arlington, TX, USA
G E Vaillant
Massachusetts General Hospital, Boston, MA, USA
K A Valentine
Singapore Management University, Singapore
F van Dongen
University of Limerick, Co. Limerick, Ireland
WAP van Tilburg
University of Limerick, Co. Limerick, Ireland
A M Vargovich
West Virginia University, Morgantown, WV, USA
M Verfaellie
Boston University School of Medicine, Boston, MA, USA
A Verges
The University of Missouri and the Midwest Alcoholism
Research Center, Columbia, MO, USA
E A Virmani
Center for Child and Family Studies, Sausalito, CA, USA
K M von Ranson
University of Calgary, Calgary, AB, Canada
J Wagemans
University of Leuven (K.U. Leuven), Leuven, Belgium
J Waguspack
University of Texas at Tyler, Tyler, TX, USA
W R Walker
Winston-Salem State University, Winston-Salem, NC,
USA
G L Wallace
King's College, London, UK
P Wallace
The Johns Hopkins University, Baltimore, MD, USA
E A Wasserman
University of Iowa, Iowa City, IA, USA
S F Waters
University of California, Davis, CA, USA
D Watson
University of Notre Dame, Notre Dame, IN, USA
D G Watson
University of Illinois, Urbana-Champaign, IL, USA
M R Watson
University of British Columbia, Vancouver, BC, Canada
W H Watson
University of Rochester Medical Center, Rochester, NY,
USA
C A Weaver
Baylor University, Waco, TX, USA
C M Weaver
Palo Alto University, Palo Alto, CA, USA
xxvi Contributors
L Wei
University of London, London, UK
M M Weissman
Columbia University, New York, NY, USA; New York
State Psychiatric Institute, New York, NY, USA
D P Welsh
University of Tennessee, Knoxville, TN, USA
C Westenbroek
Zina Pitcher Place, Ann Arbor, Ml, USA
T Wheatley
Dartmouth College, Hanover, MA, USA
J L Whitaker
The Ohio State University, Columbus, OH, USA
T A Widiger
University of Kentucky, Lexington, KY, USA
0 Wilder
University of South Florida, St. Petersburg, FL, USA
1 Wilhelmsen
Haraldsplass Deaconal Hospital, Institute of Medicine,
University of Bergen, Bergen, Norway
A Wilke
Clarkson University, Potsdam, NY, USA
D R Wilson
Creighton University, Omaha, NE, USA
L Wilson
Virginia Tech, Blacksburg, VA, USA
J Wilt
Northwestern University, Evanston, IL, USA
R Winograd
The University of Missouri and the Midwest Alcoholism
Research Center, Columbia, MO, USA
K C Winters
University of Minnesota Medical School, Minneapolis,
MN, USA
J C Wise
Oglethorpe University, Atlanta, GA, USA
J V Wood
University of Waterloo, Waterloo, ON, Canada
F J Yammarino
State University of New York at Binghamton,
Binghamton, NY, USA
D-H Yang
Asan Medical Center, Seoul, South Korea
K L Yoon
University of Maine, Orono, ME, USA
F M Zaromb
Center for Validity Research, Princeton, NJ, USA
M J Zawadzki
The Pennsylvania State University, Pennsylvania, PA,
USA
E C Zoloth
Wesleyan University, Middletown, CT, USA
HOW TO USE THE ENCYCLOPEDIA
The Encyclopedia of Human Behavior is intended for
use by students, research professionals, and interested
others. Articles have been chosen to reflect major dis-
ciplines in the study of human behavior, common
topics of research by professionals in this domain,
and areas of public interest and concern. Each article
serves as a comprehensive overview of a given area,
providing both breadth of coverage for students and
depth of coverage for research professionals. We have
designed the encyclopedia with the following features
for maximum accessibility for all readers.
Articles in the encyclopedia are arranged alphabet-
ically by subject. Complete tables of contents appear
in all volumes. The index is located in Volume 3.
Because the reader's topic of interest may be listed
under a broader article title, we encourage use of the
index for access to a subject area, rather than use of
the table of contents alone.
Each article contains a glossary, cross-references,
and a further reading list. The outline allows a quick
scan of the major ideas discussed within each article.
The glossary contains terms that may be unfamiliar to
the reader, with each term defined in the context of its
use in that article. Thus, a term may appear in the
glossary of another article defined in a slightly differ-
ent manner, or with a subtle nuance specific to that
article. For clarity, we have allowed these differences
to remain so that terms are defined relative to the
context of each article.
Each article has been cross-referenced to other
related articles in the encyclopedia. Cross-references
will always appear at the end of an article. Where
multiple cross-references apply to an article, the cross-
references will be listed alphabetically. We encourage
readers to use the cross-references to locate other arti-
cles in the encyclopedia that will provide more detailed
information about a particular subject.
The further reading section lists recent secondary
sources to aid the reader in locating more detailed or
technical information. Review and research articles
that are considered to be of primary importance to
the understanding of a given subject area are also
listed. The further reading lists are not intended to
provide a full reference listing of all the material
covered in the context of a given article, but are
provided as guides.
XXVII
This page intentionally left blank
PREFACE
The social scientists have a long way to go to catch up, but they may be up to the most important scientific business of
all, if and when they finally get to the right questions. Our behavior toward each other is the strangest, most
unpredictable, and almost entirely unaccountable of all the phenomena with which we are obliged to live.
Lewis Thomas
Psychology, the study of the human mind, has made many rapid strides during the past four decades.
There is now, more than ever before, a real need for a standard reference work covering all aspects of human
behavior. The Encyclopedia of Human Behavior is the most up-to-date and comprehensive collection of reviews
currently available. The essays will be of interest not only to clinical and experimental psychologists but also to
students in fields such as psychiatry, neuroscience, philosophy, cognitive science, and medicine. Indeed, given
the enormous range of topics covered, no one interested in human nature can fail to find something of interest
in each volume. The format of the volumes lends itself just as readily to casual perusal as it does to serious
inquiry.
My colleagues and I are often asked questions such as: What is the superego? What is repression? How
reliable is eyewitness testimony? How much sleep do we need? What do we know of the psychology of laughter,
language, cruelty, or politics? Or of love, cunning, and deceit? We have all experienced the frustration of not
being able to answer such questions or find the answers quickly without recourse to extensive library research.
This encyclopedia should prove to be an invaluable resource in such situations. Also, students of psychology
and related health professions will find this collection of articles useful as a starting point when they embark on
new research projects dealing with specific aspects of human behavior.
The study of human behavior is an enterprise that covers an enormous variety of subjects, ranging from the
minutiae of neurophysiology to such familiar but poorly understood topics as Freudian psychology. Psychology
is a science that is still very much in its infancy even though it has had a very long history, almost as long as
that of physics and biology. Anyone interested in the history of ideas should be puzzled by the differences
between advances in biology and advances in psychology. The progress in biology has been characterized by
a number of landmark discoveries, each of which resulted in a breakthrough in understanding, for example,
the discoveries of cells, Mendel's laws of heredity, chromosomes, mutations, and most recently DNA and the
genetic code. Psychology, however, has until recently been characterized by an embarrassingly long sequence of
'theories,' each of which was really nothing more than a passing fad that rarely outlived the person who
proposed it. I have always found this contrast to be quite remarkable and can think of no simple explanation
for it other than the fact that human behavior is inherently more complex, quixotic, and difficult to fathom.
Fortunately, the picture has changed radically over the past four decades, particularly in psychiatry and
cognitive neuroscience. There are two reasons for this change: First, there has been a growing dissatisfaction
with metaphorical explanations (Peter Medawar calls explanations of this kind "analgesics," for "they dull the
ache of incomprehension without removing the cause'') and a healthy trend toward replacing them with more
mechanistic explanations. Second, progress has been aided by the advent of several new technological innova-
tions for studying the structure and function of the human brain. This encyclopedia covers as many of these
recent advances as possible within a three-volume set.
My own experience is mainly in neuropsychology, medicine, and visual science, and I am therefore indebted
to the editorial advisory board, as they selected the authors in other areas and saw each essay through the long
process of peer review, revisions, and copy-editing. Most of the entries are by acknowledged experts in the field.
XXIX
xxx Preface
Given the nature and scope of this enterprise, some degree of overlap among the essays was not only inevitable
but also desirable since our goal was to ensure that each article was a self-contained summary of one specific
aspect of human behavior. Given the space limitations, each author was encouraged to provide an overview
and convey the general flavor of an area of research rather than attempt an exhaustive review. The result is a very
stimulating and informative collection of essays.
I have no doubt that this work will prove useful to specialists. If it also succeeds in kindling a spark of interest
in some aspect of human behavior among undergraduate and graduate students, then our efforts will have been
amply rewarded, for no enterprise is more important to the future of our species than an understanding of
human nature in all its diverse manifestations.
V S Ramachandran
CONTENTS
About the Editor-in-Chief
V
About the Editorial Advisory Board
vii
Reviewers
xi
Contributors
xiii
How to Use the Encyclopedia
xxvii
Preface
XXTX
VOLUME 1
Academic Achievement
B Spinath
1
Addictions and Adolescence
K C Winters, T Chung, R Stinchfield, I D Kassel, and M Conrad
9
Adolescence
S L Gray, C L Culpepper, and D P Welsh
22
Adrenal Glands
L A Trevino, M L Uhelski, A L Dougall, and A Baum
30
Aggression
B J Bushman and E H O'Brien
37
Aging and Cognition
S Horning and H P Davis
44
Aging and the Brain
P 1 Colombo, M E Crawley, B S East Jr., and A R Hill
53
Agnosia (including Prosopagnosia and Anosognosia)
A Berti and M Neppi-Modona
60
Agoraphobia and Panic Disorder
G L Thorpe, S T Sigmon, and K L Yoon
68
Agraphia and Alexia
D Fiset and D Bub
76
Alcohol: Psychosocial Effects
K J Sher, R Winograd, and A Verges
84
Alexithymia
V Panaite and L M Bylsma
92
Altruism and Helping Behavior
A A Stukas and E G Clary
100
XXXI
xxxii Contents
Alzheimer's Disease
Y Gazes, A Soldan, and Y Stern
108
Amnesia
/ F Kihlstrom and E L Glisky
116
Amnesia and the Brain
E Race and M Verfaellie
125
Analogical Reasoning
D Centner and L Smith
130
Anger
1 R Averill
137
Animal Cognition
L Castro and E A Wasserman
145
Antisocial and Narcissistic Personality Disorder
L Bobadilla and j Taylor
154
Anxiety and Fear
D W McNeil, A M Vargovich, B 1 Ries, and C L Turk
161
Anxiety Disorders
1 L Rodriguez and E A Meadows
169
Aphasia
H S Kirshner
177
Appetite
N Geary
187
Asperger's Syndrome and Nonverbal Learning Disorder
1 M Gillis, M Sevlever, and M Roth
199
Associative Learning
E Fantino and S Stolarz-Fantino
204
Attention
C Spence
211
Attention Deficit Hyperactivity Disorder
M Simonescu, K M Antshel, and S V Faraone
218
Attitude Change
R E Petty
224
Attitude Formation
M A Olson and R V Kendrick
230
Attribution
S / Tobin
236
Autism and Pervasive Developmental Disorders
R G Romanczyk and E FI Callahan
243
Autobiographical Remembering and the Self
] J Skowronski and W R Walker
251
Avoidant Personality Disorder
C A Sanislow, E E Bartolini, and E C Zoloth
257
Behavior Analysis
H S Roane and A M Betz
267
Behavior Genetics of Personality
S C South and R F Krueger
274
Contents xxxiii
Behavior Measurement in Psychobiological Research 280
G I Norman, G G Bemtson, and J T Cacioppo
Behavioral Economics 288
S Dhami and A al-Nowaihi
Behavioral Genetics 301
C T Nagoshi
Behavioral Medicine 311
7 A Jimenez and P J Mills
Behavioral Pharmacology 319
W A McKim
The Behavior-Genetics of Intelligence 326
M A Runco
Big Five Model and Personality Disorders 330
T A Widiger
Bilingualism and Multilingualism 338
L Wei
Biofeedback Therapy 344
K W Jung, D-H Yang, and S-J Myung
Bipolar Disorder 348
/ D Pettigrew
Birth Order, Effect on Personality, and Behavior 353
C Salmon
Body Dysmorphic Disorder: A Review 360
/ E Grant and B L Odlaug
Body Image 365
N L Burke, L M Schaefer, and J K Thompson
Borderline Personality Disorder 372
7 Paris
The Brain 377
R G Mair
Brain and Behavior Relationships 386
B T Legesse, B H Price, and E D Murray
Brain Chemicals: Global Projections of Ancient Aromatic Neuro transmitters 396
E C Azmitia
Brainwashing and Totalitarian Influence 406
S M Andersen and S A Saribay
Bulimia Nervosa 413
7 E Mitchell, K Steffen, T Ertelt, and J Marino
Bullying 417
S M Swearer, A Collins, and B Berry
Caffeine 423
W Revelle, D M Condon, and J Wilt
Career Development 430
S H Osipoiu
Catecholamines and Behavior 434
L A Trevino, K Lorduy, M Natishyn, A L Dougall, and A Baum
xxxiv Contents
Central Auditory Processing Disorders
G A Stefanatos and A T DeMarco
441
Central Nervous System
B P Daly, D M Eichen, B Bailer, R T Brown, and C L Buchanan
454
Child Abuse
M J Lawler and E B Talbot
460
Childhood Mental Disorders
A Scarpa and L Wilson
467
Chunking
A L Gilchrist and N Cowan
476
Classical Conditioning
B O Ploog
484
The Clinical and Cognitive Psychology of Conflict
A Parker
492
Clinical Assessment
E Gambrill
499
Clinical Psychology: An Information Processing Approach
P Fisher
510
Cognition and Personality
G Mattheivs
517
Cognitive Behavior Therapy
R Hagen and O Hjemdal
525
Cognitive Bias
A Wilke and R Mata
531
Cognitive Disorders
M E Berryhill, D Peterson, K Jones, and R Tanoue
536
Cognitive Dissonance Theory
E Harmon-Jones
543
Comparative Primate Psychology
D Maestripieri
550
Competition
S H Liening, P H Mehta, and R A Josephs
556
Confabulation and Reality Filtering
A Schnider
563
Conflict Communication
D D Cahn
572
Conformity and Obedience
D J Packer
580
Consumer Psychology
S S Posavac, F R Kardes, and S Hoeffler
589
Coping
S Blum, M Brow, and R C Silver
596
Creative and Imaginative Thinking
M A Runco
602
Creativity
V Drago and K M Heilman
606
Contents xxxv
Crisis Management 618
A M Legg and K Sweeny
Cross-Cultural Adaptation 623
V Y Kim
Crowd Psychology 63 1
S Reicher
Crowding: Effects on Health and Behavior 638
S J Lepore
Cultural Psychology 644
/ G Miller and V Chauhan
Decision Making (Individuals) 651
C F Chick, S T Pardo, V F Reyna, and D A Goldman
Defense Mechanisms 659
G E Vaillant
Delusions 667
P R Corlett
Dependent Personality (Disorder) 674
T A Morgan and L A Clark
Depression 682
R Ingram
Depth Perception 690
M R Watson and I T Enns
Developmental Psychopathology 697
A M Klahr, S A Burt, and M Nikolas
Disgust 702
V Curtis
Dissociative Disorders 710
E Cardeha
Divorce 716
S Stack
Drugs, the Brain, and Behavior 724
D E Presti
Dyslexia 732
M I Snowling, D C Gooch, and L M Henderson
VOLUME 2
Eating Disorder Not Otherwise Specified 1
K M von Ranson
Ecological Psychology 7
E P Charles and R Sommer
Electroconvulsive Therapy and Transcranial Magnetic Stimulation 13
V Madaan, D P Bestha, and D R Wilson
Electroencephalography 15
R Srinivasan and P L Nunez
Embarrassment and Blushing 24
R I Edelmann
xxxvi Contents
Empathy
E L Stocks and D A Lishner
32
Empirical Challenges to Conventional Mind-Brain Theory
E F Kelly and E W Kelly
38
Environmental Cognition
C Spencer and K Gee
46
Environmental Psychology
R Gifford
54
Episodic and Semantic Systems of Autobiographical Memory
D R Beike and J W Ransom
61
Episodic Memory
G Gillund
68
Equity Theory
E Hatfield, R L Rapson, and L Bensman
73
Event-Related Potentials (ERPs)
M M Bradley and A Keil
79
Evidence-Based Practice
B Spring, K Neville, and S W Russell
86
Evolutionary Clinical Psychology
LEO Kennair
94
Evolutionary Developmental Psychology
A C Jordan, G L Schlomer, and B J Ellis
103
Evolutionary Personality Psychology
A / Figueredo, J A Sefcek, C J Black, R A Garcia, and W J Jacobs
111
Evolutionary Psychology
R Kurzban
118
Evolutionary Social Psychology
D T Kenrick and Y J Li
125
Exercise and Sport Psychology
A LeUnes
132
Expectation
V Hoorens
142
Extraversion-Introversion
D H Saklofske, H J Eysenck, S B G Eysenck, R M Stelmack, and W Revelle
150
Eye Movements
B Bridgeman
160
Eyewitness Identification
M McGrath
167
Facial Expression of Emotion
A Freitas-Magalhdes
173
Family Systems
W H Watson
184
Forensic Psychology in Contemporary Society
T D Crespi and L A O'Leary
194
Free Will
H Rachlin
199
Contents xxxvii
Friendship 205
B Fehr
Generalized Anxiety Disorder 215
M E Portman, J H Riskind, and N A Rector
Genius, Eminence, and Giftedness 221
/ E Davidson
Gerotechnology 228
D C Burdick
Gestalt Psychology and the Development of Perceptual Organization 238
P C Quinn and R S Bhatt
Gestures 248
PBull
Glial Cells 255
R D Fields
Grandparenthood 261
B Hayslip Jr. and K S Page
Grief and Bereavement 268
M Stroebe and W Stroebe
Group Dynamics 276
P B Paulus, J Kenworthy, and FI Coskun
Hallucinations 283
K A Bujarski and M R Sperling
Handedness 290
S D Christman
Hippocampal Formation 297
H J Spiers
History of Film and Music 305
K A Force
Histrionic Personality Disorder 312
S F Smith and S O Lilienfeld
HIV/AIDS: The Role of Behavior and the Social Environment in a Global Pandemic 316
J B F de Wit and P C G Adam
Homeostasis 324
L Asarian, V Gloy, and N Geary
Homicide 334
/ Duntley
Hope and Optimism 342
D S Bailis and J G Chipperfield
Hormones and Behavior 350
7 B Becker, J A Cummings, A N Perry, and C Westenbroek
Human Factors/Ergonomics 358
P A Hancock
Human Intelligence 364
R J Sternberg
Human Mating 371
P K Jonason, K A Valentine, and N P Li
xxxviii Contents
Hypnosis 378
S / Lynn
Hypochondriasis or Health Anxiety 385
I Wilhelmsen
Id, Ego, and Superego 393
D K Lapsley and P C Stey
Implicit Memory 400
H E Schendan
Impression Formation 410
C N Macrae and S Quadflieg
Individual Differences in Temperament: Definition, Measurement, and Outcomes 418
K I Mathewson, V Miskovic, and L A Schmidt
Individualism 426
I Park and S Kitayama
Inner Speech 436
A Morin
Intellectual Disabilities 444
M M Griffin, M M Burke, and R M Hodapp
Intention 452
C Parkinson and T Wheatley
Internet Behavior 458
P Wallace
Interpersonal Perception and Communication 465
J Omarzu and J H Harvey
Interpersonal Psychotherapy 472
A 1 Raines and M M Weissman
Jealousy 479
P N Stearns
Judgment 487
S R Charlton and K V Sobel
Jury Psychology 495
B E Turvey and J L Freeman
Kinship 503
D Jones
Language Development 511
J C Wise and R A Sevcik
Leadership 517
F / Yammarino
Learned Helplessness 525
/ N Swanson, A L Dougall, and A Baum
Leisure 531
K A Henderson
Logic: Development of Logical Operations 538
J Pascual-Leone, E MR Escobar, and J Johnson
Love and Intimacy 550
E Hatfield, L Bensman, and R L Rapson
Contents xxxix
Marital Dysfunction
557
S C South
Mate Selection
K A Valentine and N P Li
564
Media Influence on Behavior
B J Bushman and J L Whitaker
571
Meditation: The Science and the Art
D Shannahoff-Khalsa
576
Memory
G A Radvansky and A K Tamplin
585
Memory, Neural Substrates
R Gupta and D Tranel
593
Mental Imagery
601
G Ganis
Military Psychology
M D Matthews and J H Laurence
608
The Mind-Body Problem
615
D E Presti
The Mirror Mechanism
G Rizzolatti and M Fabbri-Destro
622
Molecular Genetics and Human Behavior
T Partridge
630
Moral Development
F C Power
637
Moral Emotions
T Malti and B Latzko
644
Motivation
O C Schidtheiss, A Strasser, A G Rosch, A Kordik, and S C C Graham
650
Motor Control
S P Swinnen
657
Near-Death Experiences
B Greyson
669
Negotiation
P 1 Carnevale and Y K Kim
677
Neuroesthetics: The Body in Esthetic Experience
V Gallese and C Di Dio
687
Neuroethics
K L Sifferd
694
Neuroexecutive Function
D S Levine
701
Neuroimaging of Dementia
J C Masdeu
707
Neurotechnologies
A Kayser and M D'Esposito
718
Neuro theology
A B Newberg
724
xl Contents
Nonverbal Communication 731
M L Patterson
Obsessive-Compulsive Disorder 739
K R Brown, M A Stanley, and T Bjorgvinsson
Obsessive- Compulsive Personality Disorder 746
M E Owens and D C Beidel
Operant Conditioning 749
E Fantino and S Stolarz-Fantino
Organizational Behavior 757
P F Martelli, M K Stimmler, and K H Roberts
Our Cognitive Map 765
S Bell
VOLUME 3
Pain 1
Y Lampl
Paranoia 14
A Fenigstein
Parenting 21
D Laible and / Eye
Parent-Offspring Conflict 28
M Goodman, S F Waters, and R A Thompson
Peer Relationships and Influence in Childhood 34
K L McDonald and K H Rubin
Perceived Control 42
/ G Chipperfield, R P Perry, and T L Stewart
Perceptual Development 49
L A Thompson and G L Strosser
Personal Relationships in Everyday Life 57
S Duck
Personality Assessment 61
R E McGrath
Personality Development 68
M B Donnellan and R W Robins
Personality Development and Aging 74
M Allemand and R Lehmann
Personality Disorders 82
C M Weaver, C Mok, and R G Meyer
Personality, Structure 90
K Naragon-Gainey and D Watson
Persuasion 96
J Barden and R E Petty
Phobias 103
S E Cassin, J H Riskind, and N A Rector
Phonetics 109
P Case and B Tuller
Contents xli
Planning
M V Pezzo, 1 A Litman, and O Wilder
119
Play
A D Pellegrini
126
Pornography
T Kohut and W A Fisher
132
Positive Psychology
S M Schueller
140
Posttraumatic Stress Disorder
M B Powers, N Nayak, M C Capozzoli, and E B Foa
148
Preference fudgments (Individuals)
E R Igou, F van Dongen, and WAP van Tilburg
153
Prejudice, Discrimination, and Stereotypes (Racial Bias)
D M Marx and S 1 Ko
160
Premenstrual Syndrome (PMS)
S T Sigmon, J Craner, K L Yoon, and G L Thorpe
167
Primate Cognition
A E Parrish and S F Brosnan
174
Problem Solving
R E Mayer
181
Psychodynamic Psychotherapy: Theory and Practice
G O Gabbard and F Rachal
187
Psychological Predictors of Heart Disease
H S Friedman and M L Kern
195
Psychology and Religion
R W Hood Jr.
201
Psychology of Reading
C A Weaver and A E Holmes
209
Psychology, Science, and Astrology
I W Kelly, B L Janzen, and D H Saklofske
218
Psychopathology: Diagnosis, Assessment, and Classification
A M Epp, K S Dobson, and D Pusch
225
Psychophysics
F A A Kingdom
234
Psychotherapy
D Pusch, K S Dobson, and P J O'Brien
240
Reading and Phonological Processing
D Shankweiler
249
Reasoning
R B Ricco and W F Overton
257
Retirement
H L Sterns, J A Headley, and J K Kreuz
265
Risk-Compensating Behavior
M Thayer and C Garbacz
271
Risk-Taking Behavior (Young Male Syndrome)
M D Baker
276
xlii Contents
Savant Syndrome
F Happe and G L Wallace
281
Schizoid and Schizotypal Personality Disorder
P Roussos, E Chemerinski, and L J Siever
286
Schizophrenia
D R Combs , / Waguspack, and M R Basso
295
Selective Mutism
M M Nieves, F Mesa, and D C Beidel
302
Self-Defeating Behaviors
R C Curtis
307
Self-Efficacy
M W Gallagher
314
Self-Esteem
D R Kille and J V Wood
321
Self-Fulfilling Prophecy
R Rosenthal
328
Self-Regulatory Skills and Behavior Change
A Luszczynska and A B Durauia
336
Self-Relaxation
J C Smith
343
Semantic Memory
H E Schendan
350
Sensation Seeking
J H Patton
359
Sense of Smell
R L Doty
366
Sense of Taste (Effect on Behavior)
B P Daly, M P Daly, N Minniti, and J M Daly
373
The Sense of Touch
S Bensmaia and L Manfredi
379
Sentence Processing
E-K Lee and D G Watson
387
Separation Anxiety
M P Mychailyszyn, P C Kendall, ] Domingues, and K Treadwell
396
Sex Differences
D C Geary
403
Sex Roles
j D Sinnott and J S Rabin
411
Sexual Behavior
J D Baldwin and J I Baldwin
418
Sign Languages
J E Pyers
425
Sleep, Biological Rhythms, and Performance
K Cote
435
Sleep Disorders
A Culebras
442
Contents xliii
Social Anxiety Disorder
S G Hofmann, C A Gutner, and A Fang
450
Social Cognition
C Hunt, E Borgida, and H Lavine
456
Social Comparison
E A Pomery, F X Gibbons, and M L Stock
463
Social Development (Attachment, Imprinting)
R Goodvin and B A Sarb
470
Social Exchange
K S Cook and A Gerbasi
479
Social Loafing (and Facilitation)
S J Karau
486
Social Support
S J Lepore
493
Social Values (Influence on Behavior)
7 Beggan and S Allison
497
Socioemotional Development
R A Thompson and E A Virmani
504
Somatoform Disorders
E A Fors, T C Stiles, and P C Borchgrevink
512
Spatial Orientation
7 7 Rieser, G Narasimham, and A Erdemir
519
Spatial Perception
M E Beriyhill, C Hoelscher, and T F Shipley
525
Stress and Blood Pressure Dysregulation
W Gerin and M J Zawadzki
531
Stress and Illness
T R Schneider and R R Riffle
536
Studying
M Rabinowitz and j Z Ingber
540
Subjective Culture
H C Triandis and M Gelfand
544
Subliminal Perception
P K Smith and K C McCulloch
551
Suicide
D Lester and K Krysinska
558
Surprise
R Reisenzein, W-U Meyer, and M Niepel
564
Synesthesia
7 Simner
571
Syntax
77 Lasnik
578
Temperament and Individual Differences
7 Kagan
589
Terrorism
A F Lemieux and M J Boyle
595
xliv Contents
Test Behavior
604
F M Zaromb, J Bumis, and R D Roberts
Tickling
611
C R Harris
Touch
616
C Spence and A Gallace
Uncertainty
621
M Smithson
Violence
629
E K Englander and P Snell
Visual Motion Perception
637
S Grossberg
Visual Neglect
652
P W Halligan and P Bartolomeo
Visual Perception
665
L de-Wit and J Wagemans
Vocational Choice
672
G W Peterson and I G Lem,
War
683
/ H Liu, M W Allen, and D Paez
Word Retrieval
690
L E James and S Oberle
Work Efficiency and Motivation
697
E B Goldsmith
Index
703
Eating Disorder Not Otherwise Specified
K M von Ranson, University of Calgary, Calgary, AB, Canada
© 2012 Elsevier Inc. All rights reserved.
Glossary
Amenorrhea In women and postpubertal girls, loss of three
consecutive menstrual periods.
Anorexia nervosa An eating disorder in which one is
significantly underweight and exhibits fear of gaining
weight, body image disturbance, and amenorrhea.
It is explicit or inferred that the weight loss is purposeful.
Binge eating Episodes of uncontrolled eating of a larger
amount of food than is usual under the circumstances,
within a 2-h period of time.
Bulimia nervosa An eating disorder in which one
experiences recurrent episodes of binge eating and
inappropriate compensatory behavior, and one's
self-evaluation is unduly influenced by one's shape and
weight.
Compensatory behavior Inappropriate behaviors
undertaken with the intent to lose or avoid gaining
weight. Encompasses both purging and nonpurging
behaviors.
DSM An abbreviation for the Diagnostic and Statistical
Manual of Mental Disorders, which provides definitions
of mental disorder categories. It is sometimes referred to
as the 'psychiatrist's Bible.' The DSM is currently in its
fourth edition (DSM-IV) and is used by mental health
professionals who are primarily located in North America.
Eating disorder A persistent, severe disturbance in eating
behavior that is not better explained by a medical disorder or
another psychological disorder. Typically also involves
pathological attitudes related to eating, shape, and/or weight.
Nonpurging behavior Inappropriate compensatory
behaviors, such as excessive exercise or fasting for a day or
more, that do not involve purging.
Not otherwise specified (NOS) In DSM-IV, this residual
category is reserved for disorders of clinical severity that do
not fit other, more specific categories. For eating disorders,
the specific categories are anorexia nervosa and bulimia
nervosa.
Purging Compensatory behaviors such as self-induced
vomiting or misuse of laxatives, diuretics, or enemas, which
are undertaken to induce immediate or rapid weight loss or
to prevent weight gain.
Residual A type of diagnostic category composed of
symptom patterns of clinical severity that are left over and
not otherwise accounted for by existing diagnoses.
Subsyndromal Symptoms exhibited that do not
meet a diagnostic threshold for severity or frequency.
For example, subsyndromal bulimia nervosa might
involve binge eating on average once a week for 3 months,
in contrast to full-threshold bulimia nervosa in which binge
eating occurs at least two times a week for 3 months.
Far from being fixed entities, definitions of mental disorders
may change over time, as research findings and clinical
observations continue to impact experts' understanding of
these problems. In particular, an evolution in conceptualiza-
tion has profoundly affected eating disorders, which have
become relatively common and have sprung to the attention
of mental health researchers and practitioners and the general
public only since the final quarter of the twentieth century.
Among eating disorders, this evolving conceptualization is
particularly relevant for the residual diagnostic category of
eating disorder not otherwise specified (EDNOS).
What is EDNOS? In a nutshell, an EDNOS diagnosis
describes clinically significant eating disorders that do not
satisfy the criteria for either of the two, officially recognized
eating disorders: anorexia nervosa and bulimia nervosa. The
EDNOS category spans binge-eating disorder, subsyndromal
variants of anorexia nervosa or bulimia nervosa, and a range of
other symptom patterns.
A Brief History of EDNOS
Like it or not, our understanding of mental disorders relies to a
great extent on our nosology, or classification system. Classifi-
cation systems provide a lens through which psychopathology
is construed, a framework within which mental disorders are
studied and treated, and a critical means of communication
among professionals about the problems they encounter
in clinical practice and research. The classification system
used in North America is summarized in a text called the
Diagnostic and Statistical Manual of Mental Disorders
(DSM), which is published and updated periodically by the
American Psychiatric Association. It describes mental disorders
and provides criteria for their diagnosis, and is currently in
its fourth edition (i.e., DSM-IV). A revision of the descriptive
material, but not the diagnostic criteria, was published in 2000
(i.e., DSM-IV-TR); for simplicity, however, only DSM-IV is
referred to here.
The residual diagnostic category of EDNOS was first intro-
duced in the 1987 revised version of the DSM (i.e., DSM-III-R).
In the 1980 edition of the DSM (i.e., DSM-III), a comparable
residual category was included under a different name: atypical
eating disorder. Despite the name change, in both cases this
residual diagnosis was intended to be used when an individual
experienced clinically significant eating-disorder symptoms
that did not meet specific criteria for either anorexia nervosa
or bulimia nervosa. In the current DSM system, not otherwise
specified (NOS) diagnoses are used to describe atypical, pre-
sumably uncommon variants within a diagnostic category.
NOS categories exist for all major mental disorder categories
1
2 Eating Disorder Not Otherwise Specified
in the DSM. The DSM-IV retained the EDNOS definition used
in DSM-III-R but added more examples of symptom patterns
eligible for an EDNOS diagnosis.
Current Definition of EDNOS
EDNOS encompasses diverse, clinically significant eating dis-
turbances. Six specific examples of subtypes of EDNOS were
provided in DSM-IV, although other, unspecified variants of
clinical severity - in other words, syndromes that cause distress,
impairment in functioning, or both - may be included as well.
Roughly speaking, the EDNOS category can be divided into the
following three parts, which do not necessarily reflect their
relative prevalence in the population at large. The first domain
is subsyndromal anorexia nervosa and bulimia nervosa. The
second domain is binge-eating disorder, included in DSM-IV as
a provisional diagnosis that merited further study. The third
domain includes patterns of eating disturbances distinct from
the officially recognized eating disorders (anorexia nervosa and
bulimia nervosa) and other forms of EDNOS already described.
As a residual category, EDNOS is primarily a diagnosis of
exclusion. In other words, to date it has been defined largely by
what it is not (i.e., anorexia nervosa or bulimia nervosa) rather
than what it is. To understand the diverse patterns of symptoms
described by this diagnostic category, first it is essential to
understand how the core eating disorders of anorexia nervosa
and bulimia nervosa are defined.
Subsyndromal Anorexia Nervosa and Bulimia Nervosa
In DSM-IV, anorexia nervosa is described as the refusal to
maintain a minimally normal body weight, accompanied by
the following symptoms:
• intense fear of gaining weight;
• distortions in the experience and significance of body
weight and shape, or undue influence of body weight or
shape on self-evaluation, or denial of the seriousness of the
current low body weight;
• amenorrhea, in girls who have started menstruating.
The weight threshold guideline provided for a diagnosis of
anorexia nervosa - <85% of expected body weight for age and
height, or a body mass index of 17.5kgm~ 2 - tends to be
strictly observed, providing a sharp cut-off at a given point in
time between those whose weight meets this criterion and
those whose weight does not. Two subtypes of anorexia ner-
vosa are recognized: restricting type, characterized by no regu-
lar binge-eating or purging behavior, and binge-eating/purging
type, characterized by regular binge-eating or purging behav-
ior. Note that the primary difference between the binge-eating/
purging type of anorexia nervosa and bulimia nervosa is
weight: by definition, one must be significantly underweight
to qualify for a diagnosis of anorexia nervosa.
DSM-IV lists two examples of EDNOS that constitute devia-
tions from the prototypical presentation of anorexia nervosa.
Many individuals with anorexic-like problems are not deemed
eligible for a diagnosis of anorexia nervosa because, although
underweight, they do not quite meet the weight criterion (i.e.,
their weight is 85-90% of ideal body weight), or, alternatively,
because, if female, they continue to menstruate. Currently,
individuals with either of these anorexic-like eating disorders
would be diagnosed with EDNOS rather than anorexia nervosa.
In DSM-IV, bulimia nervosa is defined as involving binge-
eating and compensatory behaviors to influence weight that
occur, on average, at least twice weekly for 3 months, along
with self-evaluation that is disproportionately influenced by
one's body weight and shape. By definition, an episode of
bulimia nervosa may not occur exclusively during an episode
of anorexia nervosa. A third example of EDNOS that is listed
in DSM-IV describes symptoms that fall short of the prototype
for bulimia nervosa. This example describes those with sub-
syndromal bulimic-like symptoms: in other words, those who
binge eat and use compensatory behaviors regularly, yet less
often than twice weekly for 3 months or for a shorter duration
than 3 months.
Binge-Eating Disorder
Although formally still considered a specific subtype of EDNOS,
binge-eating disorder was introduced as a provisional eating-
disorder diagnosis in DSM-IV, with the aim of furthering
research to ascertain the validity of the diagnosis. The inclusion
of diagnostic criteria for binge-eating disorder in DSM-IV has
indeed facilitated a subsequent proliferation of research on this
syndrome. Since its publication, more research by far has been
conducted on binge-eating disorder than on any of the other
specific examples of EDNOS listed in DSM-IV. Research on
binge-eating disorder has aimed to determine, among other
things, its severity, effective treatments, its distinctiveness from
bulimia nervosa, and its utility as a diagnostic category.
According to the DSM-IV's provisional criteria, binge-eating
disorder entails recurrent, distressing episodes of binge eating,
as described earlier. In binge-eating disorder, binge-eating epi-
sodes must not be accompanied by regular compensatory
behavior such as that seen in bulimia nervosa, that is, purging,
fasting, or excessive exercise. Binge eating must occur on at
least 2 days a week for 6 months, on average. In addition,
binge-eating episodes must be accompanied by at least three
of the following five features:
• eating more rapidly than normal;
• eating until one feels uncomfortably full;
• eating large amounts of food when not hungry;
• eating alone due to embarrassment or guilt;
• eating amounts that lead one to feel disgusted, depressed,
or very guilty.
Finally, binge-eating disorder may not occur exclusively
during a period in which one has either anorexia nervosa or
bulimia nervosa.
Atypical Symptom Patterns
Other eating disturbances that are distinct from anorexia
nervosa, bulimia nervosa, their subsyndromal variants, and
binge-eating disorder also fall into the EDNOS diagnostic cate-
gory. The following two examples are presented in DSM-IV as
additional subtypes of EDNOS.
Purging disorder
The fifth example of EDNOS has been referred to variously as
purging disorder, subjective bulimia nervosa, compensatory
Eating Disorder Not Otherwise Specified 3
eating disorder, and EDNOS-purging only. It involves the reg-
ular use of inappropriate compensatory behaviors among nor-
mal-weight individuals after eating small amounts of food,
accompanied by intense fear of gaining weight or becoming
fat. Although the topic has begun receiving intensive study by
researchers, the precise boundaries of this problem relative to
anorexia nervosa and bulimia nervosa have not yet been
widely agreed upon, and its validity as a unique diagnostic
category remains under debate. Most experts believe that
insufficient evidence has accrued so far to merit the designa-
tion of purging disorder as a diagnosis in DSM. A few studies
suggest that purging disorder may be as common as anorexia
nervosa and bulimia nervosa and that it involves significant
distress and impairment, although the findings have not been
entirely consistent.
Chewing and spitting food
The sixth example of EDNOS included in DSM-IV is the recur-
rent chewing and spitting out of large amounts of food without
swallowing. Little systematic study has addressed this problem,
but it may be common among those with eating disorders, and
may be associated with food restriction rather than binge
eating. It is unclear to what degree chewing and spitting occurs
among people who do not have other eating disorders.
Other atypical EDNOS symptom patterns
Several other atypical eating disturbances that have been
described in the empirical literature are not listed as examples
of EDNOS in DSM-IV, yet may still qualify for an EDNOS diag-
nosis if they are of clinical severity and not due to an underlying
medical illness. These problems include, but are not limited to,
non-fat phobic anorexia nervosa (a self-starvation syndrome that
is not associated with a fear of fatness) and night-eating syn-
drome (overeating at night, including after having fallen asleep).
Our knowledge of these symptom patterns tends to be limited.
Epidemiology and Significance of EDNOS
Given the initial, common sense view that problems described
by NOS categories are inherently rare and less severe than their
officially recognized counterparts, several myths about EDNOS
have been hard to shake, despite the accumulation of research
findings contradicting them. First, eating disorders have been
recognized by the scientific community as significant, notewor-
thy conditions only since approximately the late 1970s. In the
years immediately following the initial inclusion of eating
disorders in DSM-III in 1980, almost all eating disorder related
clinical and research attention focused on anorexia nervosa
and bulimia nervosa, which were widely believed to be the
most common and severe forms of eating disturbance.
Subsequent research has demonstrated that ~40-60% of
those seeking treatment for an eating disorder do not, in fact,
have anorexia nervosa or bulimia nervosa. Instead, they merit
an EDNOS diagnosis. Similarly high rates of EDNOS have also
emerged from studies of nonclinical samples. These consistent
findings indicate that, far from being a rarity, as currently
defined, EDNOS actually is the most common eating-disorder
diagnosis. Studies of the prevalence of eating disorders suggest
that as many as 12% of young women experience some type of
eating disorder, including EDNOS. However, because EDNOS
is not a unitary phenomenon and because the diagnosis lacks
inclusion criteria, it is difficult to estimate population preva-
lence of EDNOS in a precise manner.
Second, EDNOS is often mistakenly believed to affect
mainly, or only, Caucasian girls and young women. However,
EDNOS affects a wide range of people, probably in part because
it includes such a diversity of different symptom patterns. It
affects men, ethnic minorities, athletes in aesthetic sports such
as ballet and gymnastics, young children, and the elderly.
Third, there is a widespread misperception among mental
health professionals, other health care professionals, third-party
payers, and the general public about the meaning and import of
a residual eating-disorder diagnosis. Although EDNOS is often
believed to describe eating disturbances that are subclinical and
therefore less severe, research has shown that the consequences
or impact of EDNOS symptoms are often no less severe or
problematic than in anorexia nervosa or bulimia nervosa. That
is, people with EDNOS tend to experience a degree of general
psychopathology, distress, and impairment that is on a par with
anorexia nervosa and bulimia nervosa. Thus, the EDNOS diag-
nosis describes serious, problematic mental disorders, not
merely atypical or minor variants of eating disorders.
Binge-Eating Disorder
As noted above, the inclusion of specific, provisional diagnostic
criteria for binge-eating disorder in DSM-IV sparked a great deal
of research related to it, including its epidemiology and signifi-
cance. The diagnosis of binge-eating disorder affects ~3-6 % of
adults in their lifetimes, which is at least twice as common as
diagnoses of bulimia nervosa (1-3% of women) or anorexia
nervosa (0.5% of women). Binge-eating disorder is more com-
mon among obese individuals, with some estimates indicating
that as many as 30% of obese individuals have binge-eating
disorder. Those with binge-eating disorder also typically exhibit
poor psychosocial functioning and poor social adjustment and
often experience other mental disorders as well.
Binge-eating disorder is the most prevalent form of EDNOS.
The gender ratio for binge-eating disorder is approximately two
males to three females, indicating that it affects a much greater
proportion of males than does anorexia nervosa or bulimia
nervosa, which are estimated to affect about ten times as many
females as males. The average duration of illness for binge-
eating disorder among those studied is at least 1 4 years, suggest-
ing that it is a chronic and reasonably stable disorder, similar to
bulimia nervosa. Family studies demonstrate that it is transmit-
ted separately from anorexia nervosa and bulimia nervosa.
Taken as a whole, the body of research conducted on binge-
eating disorder demonstrates that this syndrome is distinct
from anorexia nervosa and bulimia nervosa on a number of
key dimensions, and that it poses a significant problem for
those it affects. Given its severity, it is likely that binge-eating
disorder will be added to future versions of the DSM.
Assessment of EDNOS
As with all eating disorders, careful, rigorous assessment is
critical for obtaining accurate estimates of the prevalence of
the component symptoms as well as symptom patterns of
4 Eating Disorder Not Otherwise Specified
EDNOS. Because the lay person's understanding of binge eating
is broader than the scientifically accepted definition, rates of
binge eating tend to be higher when assessed by self-report
rather than by diagnostic interview. As a result, self-reports of
binge eating should be interpreted cautiously. Rather, semi-
structured interviews administered by trained individuals,
who apply standardized definitions to criteria that are often
not clear-cut and who may ferret out answers to often compli-
cated questions, are considered the gold standard for diagnosis.
Semistructured interviews commonly used to assess eating-
disorder symptoms and diagnoses are the Eating Disorder
Examination and the Structured Interview for DSM-IV Disor-
ders. A substantial investment of time and expertise is required
for training in the proper conduct of these interviews.
Self-report questionnaires are often used in assessing eating-
disorder symptoms and associated constructs, such as body
image. Self-report questionnaires may be used to screen for or
to more fully describe one's eating-disorder symptoms. The
choice of questionnaire should depend on the constructs one
wishes to assess. Some frequently used questionnaires of eating-
disorder symptoms include the Eating Disorder Examination
Questionnaire, Eating Disorder Inventory, Eating Attitudes Test,
Multiaxial Assessment of Eating Disorder Symptoms, Stirling
Eating Disorder Scales, Eating Disorder Questionnaire, DSM-IV
Diagnostic Questionnaire, Minnesota Eating Behavior Survey,
and Questionnaire on Eating and Weight Patterns-Revised.
Treatment of EDNOS
To date, very few researchers have attempted to design and
study treatments addressing the broad assortment of problems
comprising EDNOS. The single exception has been transdiag-
nostic cognitive-behavioral therapy for eating disorders, which
represents a recent, promising approach for ameliorating a
variety of eating-disorder symptoms including EDNOS as
well as bulimia nervosa, although it excludes individuals
whose body mass index falls below 17.5 kg m -2 . This treat-
ment has a focused form, targeting eating-disorder symptoms
alone, and a broad form, also targeting mood intolerance, clini-
cal perfectionism, low self-esteem, or interpersonal difficulties.
By contrast, research focused on treatments for binge-eating
disorder has flourished. Cognitive-behavior therapy, includ-
ing individual and group formats as well as self-help versions,
has proved to be the most effective treatment for symptoms of
binge-eating disorder thus far. There is evidence that interper-
sonal psychotherapy may be similarly effective in reducing
binge-eating symptoms. Other forms of psychotherapy that
show promise in treating binge-eating disorder are dialectical
behavior therapy, behavioral weight loss programs, and moti-
vational interviewing. Certain medications have also demon-
strated short-term efficacy in reducing binge-eating frequency
and inducing modest weight loss among individuals with
binge-eating disorder.
Problems with the EDNOS Diagnosis
The fact that a single EDNOS diagnosis describes a hodgepodge
of eating problems poses several substantial problems that
affect our understanding of eating disorders as a whole.
As noted earlier, residual categories are usually assumed to
apply only to the minority of individuals who experience
atypical manifestations of a type of problem. In that light,
EDNOS presents a problem because it encompasses numerous
symptom patterns, none of which constitutes a majority of the
category. People with an EDNOS diagnosis may experience
entirely distinct symptoms and are bound together only
loosely through their shared characterization as experiencing
eating pathology. The breadth of symptom patterns encom-
passed by the EDNOS category hampers efforts to identify
epidemiology, course, prevention strategies, and effective treat-
ments for these problems. Researchers and clinicians need a
common language to advance knowledge, and an EDNOS
diagnosis describes too many symptom patterns to make it
meaningful. The heterogeneity of this diagnosis has been a
major impediment to our accumulation of knowledge about,
and treatment of, eating disorders.
The lack of inclusion criteria for EDNOS means that there is
no clear demarcation between EDNOS and eating disturbances
that may be unusual but are not pathological. Clinical judg-
ment, which is inherently subjective, is required to make this
determination, and the necessity of using clinical judgment has
the effect of impeding reliability, or agreement across experts
and, therefore, communication. In short, the basic problem is
that the syndromes and symptoms we look for inevitably affect
what we find. For example, we thought previously that binge-
eating disorder was uncommon and not very problematic, but
systematic research has indicated that the opposite is true. This
pattern may prove to be evident with other forms of EDNOS,
such as purging disorder.
It is important to note that an eating disorder diagnosis
describes a person's symptoms at a single point in time (or
for a fairly short period of time) and does not consider the
longer-term trajectory or course of the symptoms. Thus, an
EDNOS diagnosis may describe a person who is on the way
toward developing, or recovering from, anorexia nervosa or
bulimia nervosa, as well as a person whose eating-disorder
symptoms are not in flux. EDNOS is associated with symptoms
that vary from mild to severe, in part according to one's symp-
tom course and the phase of the disorder being currently
experienced. Longitudinal research has also shown that diag-
nostic migration, or movement among diagnostic categories,
including EDNOS, is common among those affected with
eating disorders. This movement is not random, but shows
trends over time from more restrictive behaviors toward more
dyscontrolled ones, such as binge eating and purging beha-
viors, as well as trends toward recovery for a substantial num-
ber of those affected.
Certain time-related elements of binge-eating disorder cri-
teria have been criticized by experts, including the appropriate-
ness of the 2-h maximum duration for each binge-eating
episode, the frequency of 2 days a week of binge eating, and
the 6-month duration for the symptom pattern. Specifically,
questions have been raised about the utility of such strict
inclusion criteria when those with symptoms slightly less
severe than these have been shown to be indistinguishable
from those with full-threshold binge-eating disorder. As a
result, some or all of these time-related elements may be
revised if binge-eating disorder receives full disorder status in
a future edition of DSM.
Eating Disorder Not Otherwise Specified 5
As the reader may have already noticed, there has been little
discussion of children or adolescents thus far. Indeed, a large
proportion of eating disorders related research is conducted
with individuals university-aged and older; much less research
is available on adolescents and even less on children. As eating
disorders are most common among those who have reached
puberty, and affect large numbers of adolescents in particular,
the lack of data on younger sufferers may be largely attributable
to the greater number of challenges inherent in studying
children.
Clinical presentations of children and adolescents with
eating disorders may differ from those described in DSM-IV
in part because of their stage of development, and the conse-
quences of delaying an eating-disorder diagnosis may be
severe. In particular, experts from the Workgroup for Classifi-
cation of Eating Disorders in Children and Adolescents have
argued for the use of lower, more sensitive thresholds of sever-
ity as diagnostic boundaries in children and adolescents, for
example, lower frequency of purging behaviors, and significant
declines from growth curves for determining whether a child or
adolescent satisfies the weight criterion for anorexia nervosa.
These experts have also recommended that behavioral indica-
tors of fear of gaining weight and body image disturbance
should be allowed in lieu of self-reported symptoms, and
have pointed out that requiring that certain symptoms be
articulated by the child or adolescent may be developmen-
tally inappropriate. Furthermore, these experts have recom-
mended that multiple informants, including parents, be used
in determining eating-disorder symptom patterns. If adopted,
these alterations to DSM criteria would reduce the number
of EDNOS cases and simultaneously increase the number of
officially recognized eating disorders that are seen in children
and adolescents.
Current Trends
At the time of writing in early 2010, proposed criteria for the
next edition of the DSM (i.e., DSM-5) had been recently
published online and a call for experts' feedback gone out.
These draft criteria are subject to change before the DSM-5
publication, but the proposed changes suggest that concerns
over the excessive breadth and heterogeneity of EDNOS
will probably yield a tightening of its criteria by creating a
new diagnostic category and shifting the boundaries of the
other eating disorders. Specifically, binge-eating disorder may
achieve the status of a full-fledged eating disorder comparable
to anorexia nervosa and bulimia nervosa, and criteria for
both anorexia nervosa and bulimia nervosa may be loosened
to accommodate those with anorexic-like and bulimic-like
syndromes. Anorexia nervosa may no longer require amenor-
rhea for diagnosis, and bulimia nervosa and binge-eating
disorder both may require binge eating only once a week,
on average, for 3 months. Altogether, these changes would
diminish the number of individuals captured within the
EDNOS category, allowing researchers to focus on and better
characterize the remaining symptom patterns. Efforts are
under way to identify and examine the remaining symptom
patterns that make up EDNOS, such as atypical anorexia ner-
vosa, atypical bulimia nervosa, atypical binge eating disorder,
purging disorder and night-eating syndrome. In DSM-5, the
category of Eating Disorders may also be broadened and
renamed, Feeding and Eating Conditions Note Elsewhere Classified.
Summary and Conclusions
EDNOS is a residual diagnosis for those clinically significant
eating disorders that do not fit the criteria for either of the two
officially recognized eating disorders, anorexia nervosa and
bulimia nervosa. EDNOS spans symptom patterns that include
binge-eating disorder and subsyndromal variants of anorexia
nervosa or bulimia nervosa, as well as other problematic eating
disturbances, such as purging disorder. Its symptoms may
include food restriction, binge eating, compensatory behaviors,
preoccupation with weight and shape, and body image distur-
bance. Studies of clinical and community samples have demon-
strated that EDNOS is the most common eating-disorder
diagnosis. Furthermore, EDNOS tends to be just as severe, per-
sistent, and problematic as anorexia nervosa or bulimia nervosa.
The most common subtype of EDNOS is binge-eating disor-
der, which has been well characterized and affects at least twice
the number of people as do anorexia nervosa and bulimia
nervosa together. Binge-eating disorder affects men as well as
women, children as well as the elderly, athletes in aesthetic
sports, and people of diverse cultural backgrounds and ethni-
cities. It is often a chronic problem that increases the risk of
weight gain and obesity. Cognitive-behavior therapy is consid-
ered the first-line treatment for binge-eating disorder, although
other treatment approaches also have shown promise.
The heterogeneous nature of the EDNOS diagnosis has
proved to be unwieldy, and has slowed efforts to further
our understanding of eating disorders as a whole, as well as
to identify effective treatments. Consideration is being given to
diminishing the scope of this diagnostic category in future
editions of our nosological scheme, the DSM, by broadening
criteria for the diagnoses of anorexia nervosa and bulimia
nervosa, and by designating binge-eating disorder as an offi-
cially recognized, specific eating disorder. Taking these steps
would enable experts to isolate and study in depth the atypical
symptom patterns that remain within the EDNOS category,
and would help bring EDNOS into line with the other residual
diagnostic categories for mental disorders that are recognized
in the DSM.
See also: Body Image; Bulimia Nervosa.
Further Reading
American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental
Disorders, 4th edn. Washington, DC: Author.
Bravender T, Bryant-Waugh R, Herzog D, et al. (2010) Classification of eating
disturbance in children and adolescents: Proposed changes for the DSM-V.
European Eating Disorders Review 18: 79-89.
Cassin SE, von Ranson KM, Heng K, Brar J, and Wojtowicz AE (2008) Adapted
motivational interviewing for women with binge eating disorder: A randomized
controlled trial. Psychology of Addictive Behaviors 22: 417-425.
Fairburn CG and Bohn K (2005) Eating disorder NOS (EDNOS): An example of the
troublesome ‘not otherwise specified 1 (NOS) category in DSM-IV. Behaviour
Research and Therapy 43: 691-701 .
6 Eating Disorder Not Otherwise Specified
Fairburn CG, Cooper Z, Doll HA, et al. (2009) Transdiagnostic cognitive-behavioral
therapy for eating disorders: A two-site trial with 60-week follow-up. American
Journal of Psychiatry 166: 311-319.
Grilo CM and Mitchell JE (eds.) (2010) The Treatment of Eating Disorders: A Clinical
Handbook. New York, NY: Guilford Press.
Hudson Jl, Hiripi E, Pope HG Jr., and Kessler RC (2007) The prevalence and correlates
of eating disorders in the National Comorbidity Survey Replication. Biological
Psychiatry 61 : 348-358.
Machado PP, Machado BC, Goncalves S, et al. (2007) The prevalence of eating
disorders not otherwise specified. International Journal of Eating Disorders
40: 212-217.
Mitchell JE and Peterson CB (eds.) (2005) Assessment of Eating Disorders. New York:
Guilford Press.
Stice E, Marti C, Shaw H, and Jaconis M (2009) An 8-year longitudinal study of the
natural history of threshold, subthreshold, and partial eating disorders from a
community sample of adolescents. Journal of Abnormal Psychology 1 1 8: 587-597.
Strlegel-Moore RH and Franko DL (2008) Should binge eating disorder be included In
the DSM-V? A critical review of the state of the evidence. Annual Review of Clinical
Psychology 4: 305-324.
Thomas JJ, Vartanian LR, and Brownell KD (2009) The relationship between eating
disorder not otherwise specified (EDN0S) and officially recognized eating disorders:
Meta-analysis and implications for DSM. Psychological Bulletin 135: 407-433.
von Ranson KM (2008) Personality and eating disorders. In: Wonderlich S, Mitchell JE,
de Zwaan M, and Steiger H (eds.) Annual Review of Eating Disorders, Part 2,
pp. 84-96. Oxford, UK: Radcliffe.
von Ranson KM and Robinson KE (2006) Who is providing what type of psychotherapy
to eating disorder clients? A survey. International Journal of Eating Disorders
39: 27-34.
Wonderlich SA, Joiner TE Jr., Keel PK, Williamson DA, and Crosby RD (2007)
Eating disorder diagnoses: Empirical approaches to classification. American
Psychologist 62: 167-180.
Relevant Websites
http://www.aedweb.org/ - Academy for Eating Disorders.
http://www.b-eat.co.uk/Home - Beating Eating Disorders (UK).
http://win.niddk.nih.gov/publications/binge.htm - Binge Eating Disorder (USA).
http://www.dsm5.org/Pages/Default.aspx - DSM-V (US).
http://www.nimh.nih.gov/health/topics/eating-disorders/index.shtml - Eating Disorders
information (USA).
http://www.feast-ed.org/ - Families Empowered and Supporting Treatment of Eating
Disorders.
http://www.maudsleyparents.org/ - Maudsley Parents - A site for parents of eating
disordered children.
http://www.nami.org/- National Alliance on Mental Illness (USA).
http://www.anad.org/ - National Association of Anorexia Nervosa and Associated
Disorders (USA).
http://www.nationaleatingdisorders.org/ - National Eating Disorders
Association (USA).
http://www.nedic.ca/ - National Eating Disorder Information Centre (Canada).
http://www.nice.org.uk/-National Institute for Health and Clinical
Excellence (UK).
http://www.something-fishy.org/ - Something Fishy Website on Eating
Disorders.
Ecological Psychology
E P Charles, Pennsylvania State University, Altoona, PA, USA
R Sommer, University of California, Davis, CA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Activities of perception Self-generated movements that
expose organisms to patterns of ambient energy, especially
when particular movements make it possible to perceive
things that would otherwise be missed.
Affordance The relationship between a given organism and
aspects of its environment that make specific actions
(or outcomes by means of action) possible.
Behavior setting A bounded, self-regulated, and
ordered system composed of replaceable human
and physical components that interact in a synchronized
fashion to carry out the activities expected to take
place in the setting.
Higher-order invariants Patterns in ambient energy that
specify environmental properties.
Perceptual system All parts of an organism involved
in actively looking, smelling, touching, etc. This
is contrasted with the more limited parts of
the organism studied for their role in passive
sensation.
Ecology in the Biological Sciences
Ecology, derived from a Greek word meaning 'household,'
is the branch of biology that studies the relationships
between organisms and their surroundings, especially with
regard to the adaptations of organisms to environments and
the resulting population distributions. It also studies invasion
and succession among communities and the gradual replace-
ment of one species by another that is more suited to a changed
environment.
Ecological study of a species requires a description of
its habitat and the species' relationship to that habitat, its
social organization, and its relationships to other species,
emphasizing functional adaptations. Because of its broad
scope, the term has been introduced into the social and behav-
ioral sciences and is used in different ways, often in combina-
tion with other terms as in social ecology, human ecology,
ecological psychology, and ecopsychology. All these fields
describe how organisms are affected by their surroundings
and how they, in turn, contribute to changes in those sur-
roundings through their behavior. Studies of human popula-
tions from an ecological perspective are not as common as in
the case of other species. The great diversity of human habitats
and the ability of humans to adapt to different environments
as well as to change their environments are likely reasons for
the lesser influence of an ecological perspective in the behav-
ioral sciences.
Ecological Approaches in Psychology
The term ecological psychology is most commonly applied to
J. J. Gibson and E. J. Gibson's approach to studying perception
and to R. Barker's approach to studying social interactions.
H. Heft traces the origins of both approaches to William
James's radical empiricism. James was one of the early propo-
nents of incorporation of evolutionary thinking into psychol-
ogy. He advocated a psychology that did not neglect normal
everyday experience.
The term 'psychological ecology' was introduced by K. Lewin
in 1943 in an attempt to shift the emphasis in psychology
away from individual organisms and their mental processes to
organism-environment relationships. Coming out of gestalt
psychology, Lewin's usage of the term was primarily psycho-
logical, referring to a person's interpretations of the external
environment in terms of goals, barriers, and boundaries. Lewin
defined psychological ecology as the relationship between
psychological and nonpsychological factors. In approaching
a problem, a researcher examines nonpsychological data
first to find out how they determine the conditions of life
for an individual or group. Only after the nonpsychological
factors are known can research begin on the psychological
dimensions of a problem. A. Wicker describes how this
approach necessarily focuses on the ordinary environments
in which people spend most of their time. Wicker and
Sommer developed this strategy further in a paper docu-
menting psychology's lack of interest in common, mundane
behavior settings.
E. Brunswik, Lewin's contemporary and also from the gestalt
tradition, developed an ecological model of perception in an
attempt to widen the scope of research beyond the narrow
range of stimuli used in laboratory experiments. Brunswik
maintained that research on perception should employ stimuli
typical of real life situations in an approach called 'ecological
sampling.' According to Brunswik, the researcher's task is to
locate within the richness and variety of real-world stimuli
invariant patterns that are capable of representing a stable exter-
nal world and relevant to the functional adaptation of a species.
These approaches influenced two separate lines of research
and theory in psychology, both calling themselves 'ecological,'
one emphasizing perception and the other, behavioral adapta-
tions. Although the two lines of research share the ecological
perspective of examining functional adaptations of organisms
to their environments, they are concerned with different issues,
employ different methods, and publish their studies in differ-
ent journals for different audiences. The perception research is
most closely identified with J. J. Gibson and the behavioral
studies with R. Barker and his associates. Gibson's studies are
7
8 Ecological Psychology
almost entirely experimental and theory driven, while Barker's
work is largely observational and atheoretical. Because the two
approaches to ecological psychology are so different, they will
be described separately.
The terms 'ecological' and 'psychology' have also been
blended into ecopsychology, an applied field whose central tenet
is the reciprocal relationship between the personal well-being of
individuals and the well-being of the planet. This approach
gained recognition in a 1992 book and subsequent anthology
by T. Roszak. Ecopsychology uses examples of traditional
peoples who have evolved a harmonious relationship with
nature. Similar to E. O. Wilson's concept of biophilia, which
describes the innate bond between humans and other living
organisms, ecopsychology emphasizes positive emotions such
as caring, connectedness, and the healing powers of contact with
nature. The latter is the basis of ecotherapy, which is accumulat-
ing a research base. There is increasing research evidence on the
psychological benefits of nature contact, especially wilderness
experience and human-animal bonds. Separation of humans
from nature is viewed as responsible for environmental destruc-
tion as well as people's alienation and sense of incompleteness.
Restoration of the connection between humans and nature is
considered beneficial to both. The field has also brought psy-
chological principles into environmental education and pro-
moted sustainable lifestyles. Related intellectual movements
are Conservation Psychology and Deep Ecology.
Gibson’s Ecological Psychology
J. J. Gibson and E. J. Gibson's professional careers were dedi-
cated to explaining how organisms used perception to success-
fully match their behaviors to their circumstances. Early in
their partnership, the two agreed to focus on slightly different
aspects of the problem, J. J. Gibson focusing broadly on estab-
lished perceptual abilities, and E. J. Gibson focusing on per-
ceptual development. Each of their four independently
authored books appeared on a recent list of the 100 most
important publications in the history of cognitive psychology,
and they continue to influence the current generation of
researchers, though the more devoted enthusiasts remain a
bit marginalized. As ecological psychology is most closely asso-
ciated with J. J. Gibson's work, and he was more interested in
the theoretical novelty of their approach, the references to
'Gibson' below will be to him.
Gibson completed his graduate studies at Princeton in 1920
under the direction of E. B. Holt (1842-1910) who was a
protege of William James and a leader in the philosophical
movement known as New Realism. Thus, Gibson's work can
be seen as part of a continuing lineage of American philoso-
phy. It has hints of pragmatism, a general rejection of dualism,
and, in toto, is taken by its supporters as a successful explana-
tion for how direct perception can occur. The term 'direct' has
kindled a number of debates regarding Gibson's approach.
It draws contrast with 'indirect' theories of perception in
which some psychological intermediary, such as mental inter-
pretation or past memories, supports perceiving. Gibson con-
sidered perception as a type of relationship between the
organism and the environment; presaging modern systems
theory. Gibson's approach does not attribute the activity of
the organism-environment system to any single component
of the system (i.e., the brain), nor does it allow us to treat the
bidirectional causality of the system as if it were linear (making
it incorrect to imply that sensation causes perception). Because
much attention is drawn to the philosophical aspects of
Gibson's work, the methodological innovations developed in
tandem with the theory are often neglected. Gibson claimed
that the main purpose of empirical studies should be to dem-
onstrate something unexpected. To that end, his thinking often
dwelled on results that did not seem to fit within established
models and addressed research questions not easily answered
by established theories.
Gibson's thinking developed over time, and one could treat
his books as rough markers of stages in his career. Many of
Gibson's early-career innovations highlighted in The Perception
of the Visual World are now widely accepted. In that period, he
popularized the notion that people were sensitive to optic
flow, moving patterns of stimulation on the retina. He also
argued for the value of naturalistic studies of perception,
providing evidence that real-world perceptual accuracy is
often far better than what is suggested by constraining labora-
tory studies or research on picture perception. He also
provided some of the earliest studies using motion pictures in
psychological research. In contrast, many of Gibson's mid-
career innovations described in Senses Considered as Perceptual
Systems are still viewed with skepticism: At that stage, he argued
that we could study the ecology of energy (light rays, sounds
waves, etc.) to identify the environmental stmcture that an
organism's perceptual abilities evolved to monitor. His first
paper explicitly on that topic was titled ‘Ecological Optics,'
from which his approach acquired the term 'ecological.' He
also explained perception in terms of an almost whole-body
attunement, in which the brain could play a minor role. This
was largely motivated by an attempt to define perception in
a manner that was egalitarian with respect to organisms that
bear little resemblance to humans, but still perceive their
world (insects, squid, etc.). Many of Gibson's late-career inno-
vations, outlined in The Ecological Approach to Visual Perception,
have become popular, but not typically in connection with his
larger system of thinking. During this period, he emphasized
the concept of affordances, which he saw as opportunities for
action created by a match between environmental properties
and organism properties (i.e., given my hand shape, my coffee
mug affords picking up, whereas it does not afford the same to
a deer or an ant). He also claimed that his approach could
make obsolete many of the classic debates regarding percep-
tion. Most notably, he argued for 'direct perception' and laid
out a sophisticated argument that many supposedly complex
cognitive problems were solvable quite easily via perception.
The central tenets of Gibson's last approach are (1) percep-
tion is possible because of higher-order invariants in ambient
energy arrays that specify environmental properties, (2) per-
ception is an active process, and (3) organisms are specifically
sensitive to the actions afforded them by their surroundings.
If these points are treated as purely axiomatic, they can easily
become the type of 'very abstract' or 'purely theoretical' jargon
to which many research scientists are averse. However, if they
are treated as working hypotheses, then each tenet opens lines
of research that are grounded in past empirical observations
and can make clear predictions:
Ecological Psychology 9
1. Higher-order invariants : Perception is a process of discrimi-
nating between different types of situations. Patterns in
energy fluxes are created by aspects of the world and
some of the structure of the ambient energy specifies (is
specific to) the properties of the environment. As a working
hypothesis, this suggests that (1) we can identify the specific
patterns of energy to which people are sensitive, (2) some
pattern of energy exists that can be used to accurately make
whatever perceptual distinction we are interested in, and
(3) over the course of development, the specifying patterns
become the patterns to which we are sensitive. For example,
the optic parameter tau specifies time-to-contact when
velocity is constant. D. Lee and colleagues demonstrated
that gannet dives and pigeon landings are guided by
sensitivity to that higher-order invariant. Research has sub-
sequently shown that people exhibit direct sensitivity to
tau in tasks ranging from baseball catching to automobile
braking.
2. Active perception: Because the higher-order invariants are
spread out across time and space, sensitivity to them
requires movement. Thus, it can be argued that perception
is an active process, and that every part of our body
involved in the acts of perception can be considered part
of the perceptual system. As a working hypothesis, this
suggests that (1) people should perceive more accurately
when they are allowed to move freely, (2) not all ways of
moving will lead to better perception, and (3) over the
course of development, people's movements should come
to approach the ideal movements. For example, the length
of an object (such as a rod) is specified by its inertial tensor,
which can be detected only if the object is moved around in
multiple dimensions. M. Turvey and C. Carello have
demonstrated that people trying to determine the length
of a rod they cannot see are remarkably accurate in a multi-
plicity of conditions (varying from holding the rod to hav-
ing the rod taped to their backs), under a variety of
circumstances (even underwater). However, participants
are accurate only if they are allowed to move the rod, and
only if they are allowed to move the rod in the right ways -
ways that reveal the inertial tensor 'I.'
3. Affordances: In different situations, different actions are
possible. Thus, relative to any given organism, the world
can be described in terms of what it affords. The world is
made up of affordances, opportunities for action. Due to
evolutionary and developmental factors, organisms are
able to perceive these affordances. As a working hypothesis,
this suggests that (a) we can identify the physical proper-
ties of an organism and its environment that determine
which actions are possible, (b) we can identify patterns of
energy that specify many, if not all, of those affordances,
and (c) organisms should be (or should develop to be)
sensitive specifically to transitions between situations in
which actions are possible and situations in which they
are not. For example, K. Adolph and colleagues have
shown that infants rapidly learn to attend to the steepness
of slopes when determining whether or not to crawl down
a surface and that they accurately distinguish dangerous
from safe situations. Further, infants must learn these sen-
sitivities anew for different styles of locomotion, as they
develop.
Certainly, a theoretical or methodological commitment to any
one of these points does not necessitate a commitment to
the other two. However, those in the tradition of Gibson's eco-
logical psychology see them as intimately intertwined: Organ-
isms successfully match their behavior to the environment
because they leam to discriminate between complex patterns of
energy that specify which actions are possible. Perceptual sys-
tems attune to higher-order invariants that specify affordances.
Gibson's ecological approach thus offers an explanation
for the way organisms can accomplish many sophisticated
things through perceptual processes. This has led to a great
deal of controversy as it offers the potential for a broad
assault on traditional cognitive psychology. In the extreme,
some ecological psychologists argue in favor of the radical
position that perceptual explanations can be used to solve
all supposedly 'cognitive' tasks. For example, traditional expla-
nations of how people learn to drive rely on their remem-
brances of past driving experiences. However, ecological
explanations can explain the same changes in terms of
improved attunement to driving-relevant affordances, without
reference to memory of any kind. That is, if a person's per-
ceptual system is developed properly, he or she can directly
perceive how to drive. Problems like stopping before you
hit another car need not involve 'thinking' about how hard
to press the brake or 'computation' based on distance and veloc-
ity, nor need it involve other cognitive processes such as 'antici-
pation,' 'representation,' or 'simulation.' A simpler explanation
is that one can perceive what to do, probably through a real-time
coupling between brake pressure and optic invariants specifying
time to impact. While this style of explanation becomes more
awkward in certain circumstances, such as when dealing with
picture perception, abstract activities such as higher math, or
complex verbal interactions, those of the more radical persua-
sion believe that the awkwardness will disappear when we learn
to phrase the problems in a proper manner.
While Gibson's ecological psychology began as a some-
what narrow attempt to provide more sophisticated answers
to classic perceptual problems, it has developed into an
approach capable of making insightful contributions to many
fields of psychology. Methods developed by Gibson and neo-
Gibsonian researchers continue to lead to new discoveries in
an expanding array of experimental settings.
Barker’s Ecological Psychology
Barker was strongly influenced by the field theory of Lewin,
with whom he collaborated on several classic experiments in
child psychology. Barker subsequently became disenchanted
with experimentation, which he believed failed to provide
information about the frequency, duration, and complexity
of behavior as it occurred in actual life. While admiring
Lewin's choice to study problems from real situations, Barker
maintained that behavior should be studied without outside
manipulation or imposition of structure. Rather than the
study of contrived artificial conditions in the laboratory,
he advocated the study of behavior settings that already
existed, using methods that exerted as little influence as pos-
sible upon the situation being studied. Barker's method relied
almost exclusively upon nonreactive observations of people
1 0 Ecological Psychology
in their ordinary environments. With his associates H. Wright
and P. Gump, Barker initially called this approach psycho-
logical ecology, then ecological psychology, and subse-
quently, eco-behavioral science. The methodology drew
more heavily from natural sciences such as botany and geol-
ogy, which encourage documentation of processes as they
occur naturally, than from physical sciences such as physics
and chemistry, whose laboratory experiments isolate ele-
ments from their natural contexts.
Barker realized that psychology differed from other sciences
in having proceeded directly from theory to laboratory experi-
mentation without passing through a separate descriptive
phase. As a consequence, psychology generally lacks ways to
systematically describe, classify, and conduct research on
human behavior in its natural context. To remedy this defi-
ciency, Barker and Wright established in 1947 a field station in
Oskaloosa, Kansas, a small community with a population of
about 700. Field stations had been used previously in psychol-
ogy to study animal behavior, but not human behavior. The
goal of the field station was to bring to psychology the oppor-
tunities long available to biologists - easy access to the subject
matter of the science unaltered by the selection and processing
that occurs in the laboratory or clinic.
The small size of the community, and the fact that Barker
resided there, made it possible for him to study local behavior
settings for 25 years. As Barker had intended, the psychological
field station was an established organizational unit continuing
over time, whose staff included a cadre of long-term research-
ers and staff, with space for visiting researchers. It contained
facilities for collecting, preserving, and retrieving naturalistic
data; archives of primary data comparable to those collected
in other natural sciences over long periods of time, which will
be accessible to other researchers who can analyze them in
their own ways; and statistical methods appropriate for natu-
ralistic data, perhaps similar to those used in quantitative
botany and geology.
Observers from the field station ranged throughout the
town to record everyday activities of children. Sometimes a
researcher followed a single child for a designated period of
time, recording where the child went, who was present there,
and what the child did. At other times, a researcher remained
in a designated location and recorded which children visited
and their activities while they were there. Based on these
observations, Barker concluded that the behavior of a child
could be predicted more accurately from knowing the situation
the child was in than from knowing the child's individ-
ual characteristics. This laid the groundwork for studies
by S. Milgram, P. Zimbardo, and others demonstrating the
power of situations to influence behavior vis-a-vis personality
characteristics. Throughout the existence of the field station,
there were continuous attempts to keep the community
informed about the project's methods and goals. The approach
followed the naturalists' credo of studying phenomena with-
out changing them, although there is some question as to
whether, and if so, how the presence of professors and gradu-
ate student observers affected life in a small community. Sub-
sequently, Barker and P. Schoggen compared the results of
behavior setting surveys in Oskaloosa and Yoredale, England.
Barker maintained that the ecological environment has an
objective reality with observable geographical, physical, and
temporal attributes. He stated the agenda for ecological psy-
chology as a series of questions, such as 'What are environ-
ments like?'; 'How do environments select and shape the
behavior of their occupants?'; and 'What are the structural
dynamics of environments to which people must adapt?'
Barker's conception of ecological psychology rested on several
assumptions: (1) the systems of relationships that link indivi-
duals and groups with their social and physical environment are
complex; (2) such systems cannot be understood piecemeal;
(3) environment-behavior systems have properties that
develop and change over long periods of time; (4) change in
one part of the system is likely to affect the other parts of
the system; (5) the challenge of ecological psychology is to
obtain sufficient understanding about environmental systems
to be able to predict and control the effects of planned and
unplanned interventions.
The initial methodological problem of ecological psychol-
ogy, as Barker saw it, was the identification of the natural units
into which environments could be divided. These units, called
behavior settings, have replaceable human and physical com-
ponents that interact in a synchronized fashion to carry out an
ordered sequence of events, which constitutes the setting pro-
gram. Because they are discovered through observation, they
are in no way imposed or created by the experimenter. To the
lay observer, behavior settings such as a gas station or a lunch
counter are as objective as a building or a river. Units are
identified through a behavior setting survey - a comprehensive
inventory and description of all the public behavior settings in
a community or institution. Even though the occupants may
change, the behavior setting will persist. Behavior settings are
common phenomena in everyday life and lay people have no
difficulty recognizing them; ecological psychology is thus
linked to readily understood concepts. The same behavior
setting might provide different inputs to the same person as
that individual's behavior changes. Adding to Barker's con-
struct, Wicker maintains that behavior settings have develop-
mental cycles, including a beginning and an end, and are
linked to other settings, thus requiring a larger and more
complex range of investigations, including the effects of foun-
ders and successors upon setting functions.
Barker's ecological psychology tended to be atheoretical.
One exception is now called Staffing Theory (originally known
as Manning Theory). Comparison of the behavior of inhabi-
tants of underpopulated, adequately populated, and overpo-
pulated settings provides a test of the proposition that the
environment is an interdependent whole with predictable
effects on its human inhabitants. According to Staffing Theory,
when there are too few people to perform the necessary roles,
the number and range of pressures on each inhabitant
increases because the same pressure to perform necessary
roles will be distributed over fewer inhabitants. From this, it
follows that inhabitants of underpopulated settings, in com-
parison with those in adequately populated settings, are likely
to be more active within the setting and engage in a greater
variety of actions. Admission standards may be lowered, differ-
ences among applicants may be virtually ignored, and occu-
pants must work harder at more difficult tasks to keep the
setting going. Each occupant will be more valued and given
greater responsibility. Overpopulated settings encourage the
creation of mechanisms to control the number of applicants
Ecological Psychology 1 1
along with attempts to increase the capacity of the setting,
either through enlargement or through shifts to alternative
larger settings. Predictions based on population size have gen-
erally been confirmed in studies conducted in schools,
churches, and workplaces.
Barker's interpretation of ecological psychology as a descrip-
tive naturalistic behavioral science has been extended and
refined in recent years. Behavior setting surveys have been
conducted in a variety of locations, including schools,
churches, workplaces, retail shops, mental health centers, vir-
tual communities, and clinics for HIV patients. K. Fox has
linked Barker's eco-behavioral approach to social accounting
theories, showing how data from large-scale inventories of
behavior settings can reveal changes in the quality of commu-
nity life, thereby complementing other social and economic
indicators. J. Barker has extended the approach to political
science, demonstrating how residents can make behavior
setting inventories to obtain valid knowledge about commu-
nity activities in order to improve their situation. G. Kaminski
has focused attention on individual and group goal-setting in
the formulation of behavior setting programs. G. Taylor has
used an ecological perspective to study urban crime. E. Morris
attempts to integrate ecological psychology with Applied
Behavioral Analysis. Wicker has extended both the methods
and objectives of ecological psychology, and recommends a
combination of qualitative and quantitative approaches with
the latter used to study the purposive goal-directed behavior of
individuals. While Barker emphasized commonalities among
behavior settings of the same type, Wicker focuses on the
uniqueness of settings. As an alternative to laborious and
lengthy specimen records of individual behavior, Wicker uses
intensive longitudinal case studies. He also extends Barker's
nonintrusive descriptive methods to provide a framework for
positive interventions to improve community settings and to
develop new settings to meet new community needs.
Future of Ecological Psychology
Barker and Gibson shared a deep suspicion of laboratory
experiments that distorted the processes they were designed
to study. They preferred to study behavior in its natural context
rather than in the artificial conditions of the laboratory or
clinic. When they started their research careers, both men
found that the laboratory-based techniques they had learned
in graduate school were inadequate for encompassing the real-
world phenomena that interested them. Gibson realized that
elaborate methods for studying the perception of static stimuli
by a stationary observer in a darkened experimental room
would be useless in explaining pilot responses in a moving
aircraft. As for Barker, it meant recognizing that elaborate
developmental studies using conceptually loaded artificial
tasks did not provide insight into the everyday lives of children
at lunch, on the playground, and at church.
Present Status and Future of Gibson’s Approach
In many ways, the present and future of Gibson's approach
look far brighter than its past. There are still many who
view Gibson's work with suspicion, almost as a type of
mysticism, but typically their only exposure is to the philo-
sophical arguments associated with the approach. As the pub-
lic face of the field comes to include more and more in-depth
studies, the methodological advances offered become hard
to deny, and the appeal of the approach is broadened.
Thus, Gibson's approach continues to inspire a considerable
amount of research and theory. Work by ecological psy-
chology purists has managed to become at least slightly
more mainstream, aided by the growing popularity of
dynamic-systems approaches and by the success of ecological
psychologists in analyzing problems faced by people with
degenerative diseases that affect perception-action systems.
In addition, ecological psychology has begun to have tremen-
dous influence in fields such as kinesiology and sports
psychology. In addition, some of the tenets of ecological
psychology have proved to be appealing to researchers less
interested in the system as a whole. Most obviously, the
term 'affordance' has become commonplace in many areas
of psychology and computer science.
Much of this success can be attributed to dedicated adher-
ents at major universities. The International Society for Ecolog-
ical Psychology, though small, continues to grow. Its biannual
meeting, the International Conference on Perception and
Action, is well attended, as is the European Workshop of
Ecological Psychology. The society's publications include the
journal Ecological Psychology and a book series ‘Resources
for Ecological Psychology.' The current academic headquarters
of the field is the Center for the Ecological Study of Percep-
tion and Action at the University of Connecticut. Many past
and current leaders of the field taught at or graduated from
this program.
There are obstacles yet to be overcome. Ecological psychol-
ogy has yet to gain many of the trappings of more mainstream
approaches. There are few academic courses in the subject;
there are neither any textbooks nor other book-length surveys
attempting to codify a broad range of recent advances in the
field (except perhaps some published conference proceedings);
and universities rarely think they need an ecological psycholo-
gist on the staff to round out their departments. In the end,
it may be that the ecological approach is too far removed from
lay beliefs to flourish to the same degree as cognitive psychol-
ogy, but its influence continues to grow.
The long-term direction of any field is difficult to guess, but
some short-term trends in psychology should be predictable:
(1) While past work typically focused on object perception,
future work will forge links between the ecological approach
and social psychology. (2) While past work typically took
participants' motives for granted, future work, inspired largely
by the work of B. Shaw, will try to understand how intention-
ality affects the perceptual control of behavior. (3) While past
work focused on philosophical systems with which the ecolog-
ical approach clashed, future work will try to contextualize
ecological theory with respect to traditions with which it is
more compatible. (4) It is also clear that ecological psycholo-
gists will continue to be at the forefront of generating and
applying ever more sophisticated mathematical models of
behavior and behavior control, and will continue to be impor-
tant in creating and utilizing the equipment required to test
such models.
1 2 Ecological Psychology
Present Status and Future of Barker’s Approach
Barker's interest in real-world behavior in ordinary settings
has had a mixed reception in the behavioral sciences.
Although some cognitive researchers venture outside the labo-
ratory to study behavior in mundane settings, the majority
remain in their laboratories. Environmental psychologists
have been more open to leaving the laboratory, but their
methods typically involve paper-and-pencil tests, interviews,
and questionnaires. Few have adopted Barker's reliance on
naturalistic observation. Qualitative sociologists were never
wedded to the laboratory, are familiar with case study meth-
ods, and do not object to observing mundane settings. How-
ever, they have shown little interest in the measurement of
reliability or in quantification, both hallmarks of Barker's
approach. By the 1980s, Barker's ecological psychology was
described as an isolated and self-contained enterprise. Since
then, its marginalization has increased as the field station in
Oskaloosa closed without being replicated. Barker, his associ-
ates, and his students have retired from academic positions
without being replaced, and there is no journal or Ph.D. pro-
gram specifically devoted to Barker's type of eco-behavioral
research. The labor-intensive, costly, time-consuming surveys
that Barker advocated require decades rather than years: as
such, they do not find favor with grant-supporting agencies,
and publishers are uninterested in lengthy monographs that
devote hundreds of pages to mundane behaviors of ordi-
nary people. In addition, Barker's approach lacked an action
component, either to halt environmental degradation or to
remedy its effects and promote more sustainable settings and
communities.
While Barker's work at the Midwest Field Station remains
largely a footnote in developmental and social psychology,
there have been some changes that suggest a future for an
ecological approach in psychology. The effects of context
or situation upon behavior are widely recognized. Milgram's
obedience studies and Zimbardo's simulated prison research,
along with real-world examples such as the conduct of
American guards at Abu Ghraib prison, have demonstrated
the inadequacy of trait-based explanations of behavior relative
to the power of situations. The actions of two people in a single
situation (e.g., sitting in a dentist's chair or shopping in a
grocery store) will be more similar due to situational con-
straints than the action of a single individual in these two
different situations. This raises questions about the predic-
tive value of personality trait measures when information
about situational influences is lacking. While Barker remained
committed to the naturalist's hands-off style of research, future
researchers may want to explore the behavior setting model
not only to predict but also to change behavior. Whether
Barker's ecological psychology, which involved an adult
researcher following children throughout the day or sta-
tioned in a public location taking notes, is truly nonreactive,
is debatable. There is no question that some habituation to
a researcher's continued presence in a setting will occur, but
issues of residual reactivity remain.
Technical developments, such as the availability of small
inexpensive motion-activated video cameras to record behav-
ior, software programs for classifying and digitizing visual data,
and easy compact storage of highly detailed digital specimen
records, should make it easier to conduct eco-behavior studies.
No longer will rooms of file cabinets with paper notes be
required. Digitized material can be distributed online, circum-
venting publisher resistance to books and monographs contain-
ing essentially raw behavioral data, and making it easier for
investigators to access original data.
Ecological psychology is taught in many different academic
departments and institutionalized under headings such as
ecopsychology, ecotherapy, human ecology, and conservation
psychology, which have journals, training programs, and prac-
titioners. In its insistence on external validity, ecological psy-
chology has some resemblance to anthropological and
sociological fieldwork. However, there are major differences.
Most ethnographic fieldwork is qualitative, while Barker's
approach was entirely quantitative. Furthermore, Barker's eco-
logical psychology was nonevaluative and showed little interest
in changing, improving, or even questioning the settings studied
in terms of power relationships (e.g., What is the purpose of a
school system or of a school board?), which contrasts with a
more activist and critical social science. Nor does it view the
topics it studies within a larger cultural context. As a remedy,
D. Georgiou and D. Carspecken attempt to integrate Barber's
Behavior Setting Survey with Critical Qualitative Research.
Gibson drew his examples from the outside world, but he
and his students, when possible, tested their hypotheses in the
laboratory which is not often done in anthropology or sociol-
ogy. Gibson's and Barker's versions of ecological psychology
are both different from each other and from mainstream trends
in the social and behavioral sciences.
See also: Environmental Psychology.
Further Reading
Barker R and Gump P (1 964) Big School Small School. Stanford, CA: Stanford
University Press.
Gibson EJ (1 969) Principles of Perceptual Learning and Development. East Norwalk,
CT: Appleton-Century-Crofts.
Gibson EJ (2002) Perceiving the Affordances: A Portrait of Two Psychologists. Mahwah,
NJ: Lawrence Erlbaum Associates.
Gibson JJ (1950) The Perception of the Visual World. Boston, MA: Houghton Mifflin.
Gibson JJ (1966) The Senses Considered as Perceptual Systems. Boston, MA:
Houghton Mifflin.
Gibson JJ (1979) The Ecological Approach to Visual Perception. Boston, MA:
Houghton (Reprinted in 1986 by Lawrence Erlbaum Associates.)
Heft H (2001) Ecological Psychology in Context: James Gibson, Roger Barker, and
the Legacy of William James's Radical Empiricism. Mahwah, NJ: Erlbaum.
Wicker AW (2002) Ecological psychology: Historical contexts, current conception,
prospective directions. In: Bechtel RB and Churchman A (eds.) Handbook of
Environmental Psychology, 2nd edn. pp. 114-128. New York: Wiley.
Relevant Websites
http://ione.psy.uconn.edu/~cespaweb/index.html - Center for the Ecological Study
of Perception and Action (CESPA)
http://www.trincoll.edu/depts/ecopsyc/isep/- International Society for Ecological
Psychology
http://online.sagepub.com/cgi/searchresults?qbe=speab;37/3/295&journalcode=
speab&minscore=5000 - Sage Publications
Electroconvulsive Therapy and Transcranial Magnetic Stimulation
V Madaan, University of Virginia Health System, Charlottesville, VA, USA
D P Bestha, Carolinas Healthcare System, Charlotte, NC, USA
D R Wilson, Creighton University, Omaha, NE, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary somatic symptoms. ECT is primarily used for MDD that has
Dorsolateral prefrontal cortex (DLPFC) Frontal lobe area, not responded to antidepressant medications,
which coordinates executive functioning. Psychosis Characterized by delusions, hallucinations, and
Major depressive disorder (MDD) Clinical syndrome abnormal thought processes leading to disorganized speech
characterized by low mood, anhedonia, and multiple an d behavior.
Electroconvulsive Therapy
Electroconvulsive therapy (ECT) is the most effective somatic
treatment available in psychiatry today. Over seven decades, a
refinement in the delivery of electrical stimulus, use of anes-
thetic agents (usually a short-acting barbiturate), and muscle
relaxants (succinylcholine) have established ECT as a safe,
rapid, and effective treatment in clinical psychiatry. It is one of
the safest procedures under general anesthesia, with a reported
mortality rate of <0.002%. ECT is most commonly used for
severe or medication-resistant major depressive disorder
(MDD). Compared to antidepressant medications, ECT has a
quicker onset of action and a higher response rate. However, the
response rate varies from around 80-90% in patients who were
given ECT as a first line treatment or early on in the course of
illness to around 60% in patients who were resistant to treat-
ment with antidepressants. Often, it has been used in elderly
patients with MDD, especially those with psychotic features,
and also in pregnant patients in whom a quicker response is
needed and use of medications may be restricted. It is also a
valuable option for intractable psychosis, mania, and catatonia.
Furthermore, it has also been used effectively in refractory
Parkinson's disease, neuroleptic malignant syndrome, and
intractable epilepsy. Due to the higher and more rapid response
rates of ECT, it is trending toward an earlier placement in
treatment algorithms.
While the mechanism by which ECT delivers its therapeutic
benefits is unclear, etiological theories range from neurogenesis
(with an increase in brain-derived neurotrophic factors), altera-
tions in regional cerebral blood flow, modulation of numerous
neurotransmitters and hormones, and the anticonvulsant
effects of ECT. A comprehensive pre-ECT work up involves
taking a detailed history and performing a thorough physical
examination as well as screening blood tests, such as complete
blood count and comprehensive metabolic profile. Additional
tests such as an electrocardiogram, chest radiography, or com-
puterized tomogram of the head may be ordered if clinically
relevant. Baseline cognitive functioning should be assessed.
ECT has been performed with both unilateral and bilateral
electrode placement. The most common electrode placements
involve bilateral temporal or unilateral (nondominant) place-
ment with electrodes in temporal and vertex positions. Bilateral
placement is often more efficacious but also more prone to
adverse effects, notably pertaining to confusion and memory.
With the evolution of ECT, there has also been a significant shift
in using brief and ultra-brief pulse-wave stimulation rather than
prolonged sine-wave stimulation, since the brief pulse wave is
less likely to produce cognitive adverse effects. Though there is
no consensus on stimulus dosing, doses that have usually been
found to be effective are 1.5-2. 5 times above the seizure thresh-
old for bilateral and 2. 5-6. 8 times above the threshold for
unilateral placement. The duration of a seizure does not correlate
with ECT outcome. A course of ECT is usually 6-12 in number,
with ECTs delivered twice or thrice weekly. There is no consensus
on continuation and maintenance ECT to prevent relapse and
recurrence but this is often offered to patients who are resistant to
medications and have responded well to ECT. Usually, antide-
pressants and antipsychotic medications are continued but it is
advisable to taper anticonvulsants and benzodiazepines since
they increase seizure threshold. Lithium is also best avoided as
it can contribute to confusion and prolonged seizures. There is
no absolute contraindication for ECT. Relative contraindica-
tions of ECT include recent myocardial infarction, arrhythmias,
decompensated congestive heart failure, and intracranial hyper-
tension. Common adverse effects of ECT include transient eleva-
tion in blood pressure, irregular heart rate, headache, and muscle
aches. Rarely, prolonged seizures and even status epilepticus can
occur although these typically can be quickly diagnosed and
treated given constant EEG monitoring. A spectrum of cognitive
impairments can occur from use of ECT. These include acute
confusion due to the actual seizure, which resolves within min-
utes to hours, and anterograde amnesia, which resolves in about
1-3 weeks after ECT discontinuation. Another cognitive adverse
effect, retrograde amnesia, involves episodic memory from
about 1-3 months prior to the commencement of ECT. Unilat-
eral electrode placement with brief and ultra-brief pulse-wave
stimulus delivery has improved the tolerance of ECT and mini-
mized some of the above mentioned cognitive impairments.
Subjective reports of persistent or severe memory impairments
are not confirmed when assessed by available testing.
Transcranial Magnetic Stimulation
Transcranial magnetic stimulation (TMS) is a neuromodula-
tory technique that involves passing of brief current through a
13
14 Electroconvulsive Therapy and Transcranial Magnetic Stimulation
stimulating coil placed over a particular area of the scalp to
induce a magnetic field. This magnetic field, in turn, induces a
low grade electrical current in underlying brain region. TMS
has been used for research purposes in the field of neuros-
ciences for more than two decades, is now approved by the
US Food and Drug Administration (FDA) for refractory depres-
sion, and continues to be tested for further clinical use in a
variety of neurological and psychiatric disorders. The use of
current with a frequency < 1 FFz is referred to as low frequency or
slow and has inhibitory effects. Current above 1 Hz is termed as
high frequency or fast and has excitatory effects. Delivery of a
series of pulses or a train of stimulation is termed as repetitive
TMS (rTMS). This is believed to have longer-lasting effects on
neurotransmission. rTMS is being tested for both research and
clinical purposes involving neuropsychiatric disorders. Single
pulse TMS has been used to map motor cortex output and
paired pulse TMS has been used to identify interactions involv-
ing various cortical areas. An rTMS device was approved for the
first time by the FDA in 2008 for treatment of MDD in adult
patients who have failed one antidepressant trial in the current
episode of illness. In MDD, TMS is usually delivered 5 days a
week with each course lasting between 20-30 sessions. One of
the parameters used in TMS is motor threshold (MT). MT is
defined as the minimal amount of energy required to activate a
motor area. In each session, current is delivered at a strength of
80- 120% of the MT over the left dorso-lateral prefrontal cortex
(DLPFC). Though the mechanism of action is not clear, TMS is
hypothesized to modulate serotonin transmission and alter
regional blood flow. In an interesting study, Li et al. reported
that 1 Hz TMS over the left prefrontal cortex was associated
with increased activity at the site of stimulation as well as in
connected limbic regions, including bilateral middle prefrontal
cortex, right orbital frontal cortex, left hippocampus, medio-
dorsal nucleus of the thalamus, bilateral putamen, pulvinar,
and insula. Interestingly, significant deactivation was found in
the right ventromedial frontal cortex.
TMS is a completely noninvasive procedure administered in
an outpatient setting. Unlike ECT which has a recovery period
with caution against driving on the same day, patients receiv-
ing TMS can immediately resume their day-to-day activities.
Some of the more common side effects include headache and
pain at the application site. In some studies, TMS has been
associated with transient increase in the auditory threshold
which can rarely become permanent. Use of ear plugs has been
recommended as a protective measure and should be accompa-
nied by regular monitoring for emergence of side effects such as
hearing difficulties and tinnitus. One of the most serious side
effects reported till date with TMS is the occurrence of seizures
and its incidence is estimated at 1 for every 10 000 treatment
sessions administered. The risk of seizures increases with
increase in stimulation intensity, increased frequency, increased
train duration, and decreased intertrain interval. Another rare
side effect reported has been treatment-emergent mania. There
are ongoing clinical trials involving TMS for schizophrenia,
MDD, bipolar, anxiety disorders, Alzheimer's disease, addic-
tions, obsessive compulsive disorder, smoking cessation,
chronic pain, Parkinson's disease, migraine, aphasia, tinnitus,
stroke, and Tourette's syndrome.
See also: Depression.
Further Reading
Carpenter L (2010) Sub-convulsive therapies. In: Nahas 1 (ed.) Brain Stimulation
Therapies in Psychiatry. 163rd Annual Meeting. New Orleans, LA: American
Psychiatric Association, 22 May 2010, Course 8.
Chouinard PA and Paus T (2010) What have we learned from ‘perturbing’ the human
cortical motor system with transcranial magnetic stimulation? Frontiers in
Human Neuroscience 4: 173.
Greenberg RM and Kellner CH (2005) Electroconvulsive therapy: A selected review.
The American Journal of Geriatric Psychiatry 1 3(4): 268-281 .
Kim DR, Pesiridou A, and O’Reardon JP (2009) Transcranial magnetic stimulation in the
treatment of psychiatric disorders. Current Psychiatry Reports 11(6): 447-452.
Li X, Nahas Z, Kozel FA, Anderson B, Bohning DE, and George MS (2004) Acute left
prefrontal transcranial magnetic stimulation in depressed patients is associated with
immediately increased activity in prefrontal cortical as well as subcortical regions.
Biological Psychiatry 55(9): 882-890.
Payne NA and Prudic J (2009a) Electroconvulsive therapy: Part I: A perspective on the
evolution and current practice of ECT. Journal of Psychiatric Practice 15(5):
346-368.
Payne NA and Prudic J (2009b) Electroconvulsive therapy: Part II: A biopsychosocial
perspective. Journal of Psychiatric Practice 15(5): 369-390.
Rossi S, Hallett M, and Rossini PM (2009) Safety of TMS Consensus Group. Safety,
ethical considerations, and application guidelines for the use of transcranial
magnetic stimulation in clinical practice and research. Clinical Neurophysiology
120(12): 2008-2039.
Sadock BJ and Sadock VA (2007) Brain stimulation methods. In: Sadock BJ and
Sadock VA (eds.) Kaplan and Sadock 1 s Synopsis of Psychiatry: Behavioral
Sciences/Clinical Psychiatry, 10th edn., p. 1007. Philadelphia, PA: LWW.
Electroencephalography
R Srinivasan, University of California, Irvine, CA, USA
P L Nunez, Tulane University, New Orleans, LA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cerebral cortex The outer layer of mammalian brains;
in humans about 2-5 mm thick with surface area
of roughly 1500-3000 cm 2 .
Coherence A squared correlation coefficient
measuring the phase consistency between two
signals, expressed as a function of signal
frequency.
Conductivity (ft -1 mm The property of a
material (living tissue, for example) that determines
the ease with which charges move through the
medium to produce current.
Current density (|jim mm 2 ) The flux of positive
plus negative charge passing through a unit cross-
sectional area.
Electric potential (mV) The negative gradient of the local
electric field vector.
Electrocorticogram (ECoG) Electric potential recorded
directly from the brain surface.
Electroencephalography (EEG) The electric potential
recorded from the human scalp; sometimes used more
generally to include intracranial recordings in humans and
lower mammals.
Event-related potential (ERP) Average brain electric
potential produced by a combination of sensory stimulus
and cognitive task.
Evoked potential (EP) Brain electric potential produced
on average by a sensory stimulus.
Ohm's law for a volume conductor Vector current density
equals the product of conductivity and vector electric
field. Ohm's law simplifies to the usual scalar expression.
Voltage change = Current x Resistance For current flow
confined to one spatial direction, for example, in electric
circuit wires.
Window on the Mind
Human electroencephalography (EEG) provides a convenient,
but often opaque, 'window on the mind,' allowing observa-
tions of electrical processes near the brain surface. The outer
brain layer is the cerebral cortex, believed to be largely respon-
sible for cognition: perception, memory, thinking, emotions,
actions, and behaviors. Cortical processes involve electrical
signaling between neurons that change over many times in
the 10 ms (0.01 s) range. EEG is the only widely available
technology with sufficient temporal resolution to follow
these quick dynamic changes.
EEG can be recorded using electrodes placed inside the
skull to study nonhuman mammals or human epilepsy
patients. Such intracranial recordings provide measures of cor-
tical dynamics at small spatial scales, dependent on electrode
size. However, there are significant limitations to intracranial
EEG recording for studies of cognition and behavior. Intracra-
nial recordings in humans are mostly limited to patients with
intractable epilepsy, often in preparation for brain surgery.
These recordings are called electrocorticograms (ECoGs).
ECoG recordings are usually obtained only over a very limited
portion of the cortex, areas which vary widely across indivi-
duals, partly guided by EEG recordings of epileptic activity
using electrodes placed on the scalp prior to surgery.
In both clinical and research studies, EEG is nearly always
recorded from electrodes placed on the scalp. Each scalp elec-
trode records electrical activity at large scales, measuring electric
currents (or potentials) generated in cortical tissue containing
about 30 million to 500 million neurons. Luckily, these large-
scale estimates provide important measures of brain dysfunction
for clinical work and cognition or behavior for basic scientific
studies. Human 'mind-measures' are easily obtained at the large
scale of scalp recordings. EEG monitors the state of consciousness
of patients in clinical work or experimental subjects in basic
research. Oscillations of scalp voltage tell a very limited but
important part of the story of brain functioning. For example,
states of deep sleep, coma, or anesthesia are mostly associated
with very slow EEG oscillations and larger amplitudes. Modern
signal analyses allow for identification of distinct sleep stages or
quantitative measures of the depth of anesthesia. More sophisti-
cated experimental designs and methods of signal analysis have
revealed robust connections to detailed cognitive events.
On the other hand, EEG spatial resolution is poor, com-
pared to modern brain functional imaging methods such as
PET and MRI. But these latter methods have very poor tempo-
ral resolutions on the timescale of seconds and thus do not
offer detailed information about the rapid neural dynamics
available to EEG. The related technology, magnetoencephalog-
raphy (MEG), consists of recordings of the magnetic field
generated by brain current sources. MEG also provides high
temporal resolution and low spatial resolution, similar to EEG.
MEG is preferentially sensitive to brain current sources ori-
ented tangential to the scalp surfaces, which are typically
located in the sulcal walls (folded cortex), while EEG is more
sensitive to radial sources that are mainly located in the gyral
surfaces (see Figure 1).
Sources of Scalp Potentials
The generators of scalp potentials are best described as micro-
current sources at cell membranes. Relationships between such
very small-scale sources and macroscopic potentials at the
15
1 6 Electroencephalography
Figure 1 Sources of EEG and MEG. Neocortical sources can be
generally pictured as dipole layers (or ‘dipole sheets’, in and out of
cortical fissures and sulci) with mesosource strength varying as a
function of cortical location. EEG is most sensitive to correlated dipole
layer in gyri (regions ab, de, gh), less sensitive to correlated dipole layer
in sulcus (region hi), and insensitive to opposing dipole layer in sulci
(regions bed, efg) and random layer (region ijklm). MEG is most sensitive
to correlated and minimally apposed dipole layer (hi) and much less
sensitive to all other sources shown, which are opposing, random or
radial dipoles. For this reason, MEG signals are more likely to ‘fit’
localized equivalent sources while ignoring the dominant EEG sources.
scalp are made easier by employing an intermediate ( mesoscopic )
descriptive scale. This approach makes use of the columnar
structure of the neocortex, believed to contain the dominant
sources of spontaneous scalp potentials. For macroscopic mea-
surements, the 'mesoscale source strength' of a volume of tissue
is defined by its electric (current) dipole moment per unit
volume
ws(r', w, t)dW(w) [1]
w
Here, the three integral signs indicate integration over a small,
local volume W of tissue where dW(w) is the tissue volume
element. s(r', w, f) is the local volume source current
(|rAmm~ 3 ) near membrane surfaces inside a tissue volume
with vector location r'. w is the vector location of sources
within dW(w). The current dipole moment per unit volume
P(r', t) in a conductive medium is fully analogous to charge
polarization in a dielectric (insulator). Macroscopic tissue
volumes satisfy the condition of electroneutrality at EEG fre-
quencies. That is, current consists of the movement of positive
and negative ions in opposite directions, but the total charge in
any mesoscopic tissue volume is essentially zero. Cortical mor-
phology is characterized by its columnar structure with pyra-
midal cell axons aligned normal to the local cortical surface.
Because of this layered structure, the volume elements dW(w)
may be viewed as cortical columns with height ~ 2-5 mm. For
purposes of describing scalp potentials, the choice of basic
cortical column diameter is somewhat arbitrary. Anything
between the cortical minicolumn (f« 0.03 mm) and macrocolumn
scales (~ 1 mm) may be used to describe scalp potentials. The
main consideration is that tissue volumes are assumed to be
sufficiently large such that current conservation within the
volumes is satisfied. Otherwise, monopole contributions to
distant potentials are expected.
P(r', t) =
W
The microsources sfr 7 , w, f) are generally mixed positive and
negative due to local inhibitory and excitatory synapses, respec-
tively. In addition to these active sources, the sfr 7 , w, t) include
passive membrane (return) current required for current conser-
vation. Dipole moment per unit volume Vfr 1 , t) has units of
current density (|rAmm~ 2 ). For the idealized case of sources
of one sign confined to a superficial cortical layer and sources of
the opposite sign confined to a deep layer, P(r', t) is roughly the
(negative) diffuse current density across the column, - J(r', t).
This might occur due to superficial inhibitory synapses and
deep excitatory synapses, for example. Generally, column
source strength Ptr 7 , £) is reduced as excitatory and inhibitory
synapses overlap along column axes. In the case of fully mixed
synapses, P(r', t) ^ 0, and a so-called closed field results.
Human neocortical sources may be viewed as forming a
large dipole sheet (or layer) of perhaps 1500-3000 cm 2 over
which the function Pfr 7 , t) varies continuously with cortical
location r 7 , measured in and out of cortical folds. Figure 1
shows an idealized cortical surface with sources generally ori-
ented normal to the cortical surface. In limiting cases, this
dipole layer might consist of only a few discrete regions
where P(r 7 , f) is large, consisting of localized or focal sources.
But, more generally, Pfr 7 , f) is distributed over the entire folded
surface. The question of whether P(r', t) is distributed or loca-
lized in particular brain states is often controversial. The aver-
aging of evoked potentials over trials substantially alters the
nature of this issue. Such time averaging strongly biases evoked
potentials toward (trial to trial) time stationary sources, for
example, sources confined to the primary sensory cortex.
Volume Conduction of Head Currents
Scalp potential may be expressed as a volume integral of dipole
moment per unit volume over the entire brain, provided
P(r 7 , f) is defined generally rather than in columnar terms.
For the important case of dominant cortical sources, scalp
potential may be approximated by the following integral of
dipole moment over the cortical volume 0
V(r,f)
| G(r, r')-P(r 7 , t)d<9(r')
0
[ 2 ]
If the volume element dOfr 7 ) is defined in terms of cortical
columns, the volume integral may be reduced to an integral
over the folded cortical surface. Equation [2] indicates that
the time dependence of scalp potential is the weighted sum
(or integral) of all dipole time variations in the brain. Deep
dipole volumes typically make negligible contributions
because of the greater separation from the scalp electrodes.
The weighting function is called the vector Green's function
G(r, r 7 ), essentially an inverse electrical distance. G(r, r 7 ) con-
tains all geometric and conductive information about the
head volume conductor. The most common head models
consist of three or four concentric spherical shells, representing
the brain, cerebrospinal fluid, skull, and scalp tissue with
different electrical conductivities a as shown in Figure 2.
In genuine heads, however, G(r, r 7 ) is much more complicated.
More complicated numerical methods may also be used to
estimate G(r, r 7 ), sometimes employing MRI to determine
tissue boundaries. However, the accuracy of both analytic
Electroencephalography 17
Figure 2 Volume conduction models for EEG. (a) A dipole is shown in
the inner sphere of a 4-concentric spheres head model consisting of
the inner sphere (brain) and three spherical shells representing CSF
(cerebral spinal fluid), skull, and scalp. The parameters of the model
are the radii (rl, i2, r3, r4 ) of each shell and the conductivity ratios
(<ri/<72, <7-1 /<r 3 , <7i/<7 4 ). Typical values are radii (8, 8.1, 8.6, 9.2 cm) and
conductivity ratios (0.2, 40, 1). (b) A realistically shaped boundary
element model (BEM) of the head. The brain and scalp boundaries were
found by segmenting the images with a threshold, and opening and
closing operation respectively, while the outer skull boundary was
obtained by dilation of the brain boundary (ASA, Netherlands). Although
geometrically more accurate than the spherical model, the
(geometrically) realistic BEM suffers a potentially fatal flaw because
tissue resistivity is poorly known.
and numerical methods may be severely limited by incomplete
knowledge of tissue conductivities. MRI has also been sug-
gested as a future means of estimating tissue conductivities,
based on the close relationship of diffusion to current flow
processes.
Despite these limitations preventing highly accurate esti-
mates of the function G(r, r'), a variety of studies using concen-
tric spheres or numerical methods have provided reasonable
quantitative agreement with the experiment. The cells generat-
ing scalp EEG are believed to have the following properties:
1. In the case of potentials recorded without averaging, cells
generating EEG are mostly close to the scalp surface. Poten-
tials fall off with distance from source regions. In real heads,
tissue inhomogeneity (location-dependent properties) and
anisotropy (direction-dependent properties) complicate
this issue. For example, the low conductivity, three-layered
skull tends to spread currents (and potentials) in directions
tangent to its surface. Brain ventricles, the subskull cerebro-
spinal fluid layer, and skull holes (or local reductions in
resistance per unit area) may provide current shunting.
But generally, sources closest to electrodes are expected to
make the largest contributions to scalp potentials.
2. The large pyramidal cells in the cerebral cortex are aligned in
parallel, perpendicular to local surface. This geometric
arrangement encourages large extracranial electric fields
due to linear superposition of contributions by individual
current sources. Columnar sources P(r', £) aligned in parallel
and synchronously active make the largest contribution to
the scalp potential. For example, a 1-cm 2 crown of cortical
gyrus contains about 110 000 minicolumns, approximately
aligned. Over this small region, the angle between P(f, t) and
G(r, r 7 ) exhibits relatively small changes. By 'synchronous'
sources, it is meant that the time dependence of Pfr 7 , f) is
roughly consistent (phase locked) over the area in question.
In this case, eqn [2] implies that individual synchronous
column sources add by linear superposition. By contrast,
scalp potentials due to asynchronous sources are due only
to statistical fluctuations, that is, imperfect cancellation of
positive and negative contributions to the integral in
eqn [2], Scalp potential maybe estimated as roughly propor-
tional to the number of synchronous columns plus the
square root of the number of asynchronous columns. For
example, suppose 1% (s^ 10 3 ) of the gyrial minicolumns
produce synchronous sources P(r 7 , £) and the other 99% of
minicolumns (s 2 ~ 10 5 ) produce sources with random time
variations. The 1% synchronous minicolumn sources are
expected to contribute roughly Si/sls 2 or about three times as
much to scalp potential measurements as the 99% popula-
tion of random minicolumn sources.
3. The observed ratio of brain surface (dura) potential magni-
tude to scalp potential magnitude for widespread cortical
activity like alpha rhythm is roughly in the 2-6 range.
By contrast, this attenuation factor for very localized corti-
cal epileptic spikes can be 100 or more. A general clinical
observation is that a spike area of at least 6 cm 2 of cortical
surface must be synchronously active to be identified on the
scalp. This area contains about 700 000 minicolumns or
70 000 000 neurons forming a dipole layer. These experi-
mental observations are correctly predicted by eqn [2],
4. For dipole layers partly in fissures and sulci, somewhat
larger areas are required to produce measurable scalp
potentials. First, the maximum scalp potential due to a
cortical dipole oriented tangent to the scalp surface is esti-
mated to be about 1/3 to 1/5 of the maximum scalp poten-
tial due to a dipole of the same strength and depth, but
orientated normal to the surface. Second, tangential dipoles
tend to be located more in fissures and (deeper) sulci, and
1 8 Electroencephalography
may also tend to cancel due to opposing directions on
opposite sides of the fissures and sulci. Third, and most
importantly, synchronous dipole layers of sources with
normal orientation covering multiple adjacent gyri can
form, leading to large scalp potentials due to the product
P(r', £)G(r, f) having a constant sign over the integral
in eqn [2],
Recording Methods
EEG Machines
Human EEG is recorded using electrodes with diameters typi-
cally in the 0.4-1-cm range, held in place on the scalp with
special pastes, caps, or nets, as illustrated in Figure 1. EEG
recording procedures are noninvasive, safe, and painless. Nev-
ertheless, experimental subjects used in research laboratories
must provide informed consent. Special gels are applied
between electrodes and scalp to improve electrical contact.
Wires from scalp electrodes connect to special EEG machines
containing amplifiers to boost raw scalp signals, which are
typically in the 5-200-mV range or roughly 100-1000 times
smaller than EKG (heart) signals.
Electrode Placement
The electrode positions (which are referred to collectively as
the electrode 'montage') vary considerably across laboratories.
Standard electrode-placement strategies use the 10-20, 10-10,
and 10-5 placement systems. These montages are based on
systematic extensions of the clinical standard 10-20 electrode
montage, and they are widely but not universally used. The
basis of these standard electrode placements is to define con-
tours between skull landmarks (e.g., nasion and inion) and to
subdivide the resulting contours in proportional distances. The
10-5 system uses proportional distances of 5% of the total
length along contours between skull landmarks, compared to
the 20% and 10% distances used in the 10-20 and 10-10
systems, respectively. EEG caps are commonly commercially
available with these electrode positions. However, most mod-
ern EEG systems in research laboratories that use a larger
number of channels (perhaps 64-130 or more) make use of
other placement systems developed with the aim of obtaining
more regularly spaced sampling of the EEG, with nearly equal
nearest-neighbor electrode separations.
Potential differences between electrode pairs are recorded
with EEG machines containing amplifiers, filters, and other
hardware. A ground electrode is also required to provide a
reference signal to the differential amplifiers. The distinction
between 'recording' and 'reference' electrodes is mostly artificial
since it is the potential differences between the two sites
that allow closed current loops through the tissue and the
EEG machine. In referential recordings, potentials between each
recording electrode and a fixed reference electrode are measured
over time. The potential difference between any pair of electro-
des can then be easily obtained by subtracting one signal from
the other. A widely used strategy in research EEG is the common
average reference, where the average potential of all channels is
subtracted from each channel. Bipolar recordings measure poten-
tial differences between adjacent scalp electrodes and are a
commonly used strategy in clinical EEG. When such bipolar
electrodes are placed close together (say 1 or 2 cm apart), poten-
tial differences are estimates of the tangential electric fields
(or current densities) in the scalp between the electrodes.
Standard clinical practice makes use of a 10-20 electrode
montage consisting of only 21 electrodes. Often 48-136 re-
cording electrodes are used in research; some laboratories now
use as many as 256 channels to obtain more detailed informa-
tion about brain sources by means of sophisticated computer
algorithms. The resulting multichannel data are submitted to
two distinct types of computer algorithms: (1) High-resolution
EEG methods estimate potentials on the brain surface by
accounting for distortions caused by intervening tissue and
the physical separation of electrodes from the brain. (2) Dipole
localization methods can estimate the location of source regions
in the brain in a few specialized applications where EEG is
expected to be generated mainly in only one or two isolated
source regions.
Artifact
Potentials recorded from the scalp are generated by brain
sources, environmental and hardware system noise, and
biological artifacts. Biological artifacts often contaminate EEG
records and generally pose a more serious problem than envi-
ronmental or system noise, the latter substantially reduced by
modern amplifier systems. Common artifact sources include
whole body movement, heart, muscle, eye movements, and
eye blink. EEG records containing large artifacts are usually
discarded. Artifact removal by computer is typically successful
only for a highly stereotypical low-frequency artifact such as
eye blinks. The effectiveness of automated artifact editing is
severely limited because the frequency bands of biological
artifacts substantially overlap the important EEG bands,
making distinctions between artifact and brain signal very
difficult. This problem is severe at frequencies above 20 Hz,
where the electrical artifact from scalp muscles is recorded
along with the EEG. Such high-frequency muscle artifact can
easily dominate EEG in the same high-frequency range.
The Reference Electrode
Ideally, an EEG would represent the reference-free potential or
nominal potential with respect to infinity. The label 'nominal'
means that even if it were possible to measure scalp potentials
with respect to 'infinity' (or a distant wall of the laboratory),
such large potentials (measured in volts) would have no phys-
iological relevance, being dominated by power line fields and
other environmental sources. The 'nominal potential with
respect to infinity' is defined as the potential with respect to
infinity due only to sources generated by the brain. For many
years, EEG scientists seemed to insist (based on both word and
deed) that if only some body location could be found with no
active local sources, such a reference location would allow for
genuine reference-free recordings. This old idea has long been
discredited using both simulations and genuine data; EEG
always represents the potential difference between two body
locations. Any set of data recorded with some reference loca-
tion may be referenced to another location. For example, the
average reference is obtained by simply adding up all recorded
Electroencephalography 1 9
potential differences and subtracting this sum from each indi-
vidual potential difference. Like any other choice of reference,
the average reference provides imperfect estimates of reference-
independent potentials, but when used with large number of
electrodes (>64) it often performs reasonably well.
Time Dependence
Oscillatory Waveforms
Voltage traces of EEG signals recorded from each electrode pair
oscillate with mixtures of component waveforms with different
frequencies. Standard EEG analyses often begin with the spec-
tral analysis of the EEG signal using Fourier methods. Each
frequency component may be defined in terms of three para-
meters, its amplitude (A nm ), frequency (/„„,), and phase (0„ m ).
The subscript n denotes the frequency component and the
subscript m indicates an electrode pair used to record a potential
difference. The electrical power associated with each frequency
component is proportional to the square of the corresponding
amplitude. One may express any physical waveform as a sum of
components with different frequencies, amplitudes, and phases
called a Fourier series. Fourier series are analogous to expres-
sions of music or other sounds as compositions of tones or of
white light composed of many colors. The EEG voltage V m (t)
recorded from any electrode pair m may be expressed generally
as a sum over frequency components
N
( t) = ^ Smdilfnmt f) nm ) [3]
n=l
Waveform frequencies f„ m are expressed in terms of number of
cycles per second (or Hertz, abbreviated Hz). EEG frequency
ranges are categorized as delta (1-4 Hz), theta (4-8 Hz), alpha
(8-13 Hz), and beta (>13 Hz). Very high frequencies (typically
30-40 Hz) are referred to as gamma activity. These distinctive
labels correspond roughly to frequency ranges (or bands)
that often dominate particular human brain states. Disparate
rhythms may be associated with behavioral or cognitive state,
brain location, or other criteria. For example, delta activity
with frequencies lower than about 1 or 2 Hz provides the
largest EEG amplitudes (or power) during deep sleep and in
many coma and anesthesia states. Alpha, often mixed with
low-amplitude delta, theta, and beta, is typically predominant
in awake-resting states. Alpha is the dominant rhythm in alpha
coma and is superimposed on delta activity during some sleep
stages. Distinct rhythms can also occur in the same frequency
range. Thus, the plural terminology, alpha rhythms, beta
rhythms, and so on, appropriately describes the wide variety
of EEG phenomena.
Modern methods of time series analysis are often used to
simplify complicated waveforms like EEG. Many industrial
applications involve such methods, including electric circuits,
signal processing (television, radar, astronomy, etc.), voice
recognition, and so on. Most time series analyses are based
on spectral (or Fourier) methods. Computers extract the ampli-
tudes A nm and phases 4>„ m associated with each data channel
( m ) and frequency (n) from the often complicated EEG, repre-
sented by eqn [3] . The computer 'unwraps' the waveform V m (t)
to reveal its individual components. Such spectral analysis is
analogous to the physical process performed naturally by
atmospheric water vapor to separate white light into its com-
ponent colors. Each color is composed of electromagnetic
waves within a narrow frequency band, providing the basis
for rainbows.
The Rhythmic Zoo
Human EEG exhibits many waveforms, especially in the expe-
rience of clinical electroencephalographers (neurologists with
this specialized training). Some EEG have known clinical
significance and some do not. Any complicated waveform
can be described as a mixture of oscillations with different
frequencies and amplitudes, as indicated by eqn [3]. But more
picturesque descriptions are often preferred by clinical elec-
troencephalographers to characterize the 'zoo' of EEG wave-
forms. Such labels include paradoxical alpha, spike and wave,
delta focus, sharp transient, sleep spindle, nonspecific dysrhythmia,
and others.
Cortical EEG (ECoG) typically consists of even more com-
plex waveforms, composed of rhythms with different frequen-
cies, locations, and spatial extent. Such highly localized cortical
rhythms are not recorded on the scalp. Cortical beta and
gamma rhythms are often strongly attenuated between cortex
and scalp because they are more localized or less synchronized
over cortical areas than some of the alpha, theta, and delta
band activity. EEG during sleep, coma, and anesthesia typically
exhibits large scalp amplitudes of low-frequency oscillations,
implying widely distributed and highly synchronous cortical
source activity.
Alpha Rhythms
Alpha rhythms provide an appropriate starting point for clini-
cal EEG examinations. Some initial clinical questions are: Does
the patient show alpha rhythms, especially over the posterior
scalp? Are their spatial-temporal characteristics appropriate for
the patient's age? How do they react to eyes opening, hyper-
ventilation, drowsiness, etc.? For example, pathology is often
associated with pronounced differences in EEG recorded over
opposite hemispheres or with low alpha frequencies. A resting
alpha frequency lower than about 8 Hz in adults is considered
abnormal in all but the very old.
Alpha rhythms may be recorded in roughly 95% of healthy
adults with closed eyes. The normal waking alpha rhythm
usually has larger amplitudes over posterior regions, but is
typically recorded over widespread scalp regions. Posterior
alpha amplitude in most normal adults lies in the range of
15-50 mV; alpha amplitudes recorded from frontal electrodes
are usually lower. A posterior rhythm of ~4 Hz develops in
babies in the first few months of age. The amplitude increases
with eye closure and is believed to be a precursor of mature
alpha rhythms. Maturation of the alpha rhythms is character-
ized by increased frequency and reduced amplitude between
the ages of about 3 and 10.
Normal resting alpha rhythms may be substantially
reduced in amplitude by eye opening, drowsiness, and in
some subjects, by moderate to difficult mental tasks. Alpha
rhythms, like most EEG phenomena, typically exhibit an
inverse relationship between amplitude and frequency.
20 Electroencephalography
For example, hyperventilation and some drugs (alcohol,
for example) may cause reductions of alpha frequencies
together with increased amplitudes. Other drugs (barbiturates,
for example) are associated with increased amplitude of
low-amplitude beta activity superimposed on scalp alpha
rhythms. The physiological bases for the inverse relation
between amplitude and frequency and most other properties
of EEG are largely unknown, although physiologically based
dynamic theories have provided several tentative explanations.
Figure 3 describes a 60-s recording of alpha rhythm from
two scalp locations. The subject is a healthy waking adult,
relaxed with eyes closed. By examining the power spectra for
the occipital (Figure 3(a)) and frontal sites (Figure 3(b)) for
individual epochs with frequency resolution of 0.5 Hz, support
for two different kinds of oscillation within the alpha band is
obtained. At the occipital channel, individual epochs display
distinct peaks at 9.5 and 10.5 Hz. The first 15 epochs show a
strong response at 10.5 Hz but the later epochs show a stronger
response at 9.5 Hz. The dominant frequency in each epoch is
summarized in the peak power histograms in Figure 3(c)
showing that individual epochs displayed peak frequencies at
both 9.5 and 10.5 Hz. By contrast, very few epochs have a peak
frequency of 10.5 Hz at the frontal site (Figure 3(d)); most
epochs have peak frequencies either at 9.5 Hz or in the delta
band (<2 Hz). Note that during most epochs with strong delta
activity as in Figure 3(b), the alpha peaks are attenuated. The
amplitude spectra show mixtures of frequencies that depend
partly on scalp location; however, alpha rhythm is dominant at
all locations in this typical example. This subject has two
frequency peaks near 10 Hz, a relatively common finding.
These two alpha oscillations are partly distinct phenomena as
revealed by their separate distributions over the scalp and their
distinct behaviors during mental tasks. The multiple underly-
ing source regions are believed to overlap substantially.
Alpha, Theta, Cognitive Tasks, and Working Memory
Two prominent features of human scalp EEG show especially
robust correlation with mental effort. First, alpha band ampli-
tude normally tends to decrease with increases in mental effort.
Second, frontal theta band amplitude tends to increase as tasks
require more focused attention. A common but oversimplified
view of alpha rhythms is one of brain idling. But upper and
lower alpha band amplitudes may change independently,
depending on scalp location and task. Some working memory
tasks cause lower alpha band amplitude to decrease while
upper alpha band amplitude increases. Alpha amplitude reduc-
tions may reflect local cortical desynchronization occurring in
task-relevant brain areas, whereas task-irrelevant regions may
be unchanged or even produce larger alpha amplitudes.
Another (possibly complementary) hypothesis is that increases
in higher frequency alpha amplitudes reflect a specific memory
processing function and not simple idling.
Generally, intracerebral electrodes record a variety of alpha
rhythms. Some intracerebral alpha rhythms are blocked by
opening the eyes and some are not. Some respond in some
way to mental activity and some do not. The alpha band
rhythms recorded on the scalp represent spatial averages of
multiple alpha phenomena and a large number of alpha com-
ponents. For an alpha rhythm of a particular type to be
observed on the scalp, it must be synchronized (roughly in
phase) over a large cortical area.
Figure 3 An example of power spectra from a single subject (female, 22 years). The subject is at rest with eyes closed, (a) Plots of 30 (individual
2-s epoch) power spectra for a channel over the occipital lobe, (b) Plots of the same 30 individual epoch spectra for a frontal channel, (c) Peak
power histograms show the distribution of peak frequencies for the 30 epochs shown in (a), (d) Peak power histograms for the 30 epochs shown in (b).
Electroencephalography 21
Topography
Dynamic Measures that Depend on Spatial Location
The EEG recorded from a single electrode pair is fully charac-
terized by its time dependence, as in eqn [3]. In spatially
extended systems, however, dynamic behavior generally de-
pends on time and spatial location, the usual independent
variables of spatially extended dynamical systems. Thus, multi-
channel recordings potentially introduce many new measures
of brain dynamic behavior. Amplitude, phase, and frequency,
for example, may vary with scalp or cortical spatial location.
Spatial Distribution of Alpha Rhythms
Alpha rhythms have been recorded from nearly the entire upper
cortical surface (ECoG), including frontal and prefrontal areas.
High-resolution EEG scalp recordings also show widespread
distribution of alpha rhythms over nearly the entire scalp in
healthy, relaxed subjects. EEG clinical populations differ, typi-
cally involving patients who are older, have neurological pro-
blems, and may be anxious during recording. These factors all
tend to work against the production of robust, widespread alpha
rhythms. Second, the clinical definition of alpha is based on raw
waveforms rather than the spectra. Often alpha is identified
simply by counting the number of zero crossings of recorded
waveforms. This can sometimes provide a misleading picture
because raw EEG composed of broad frequency bands can
appear very 'nonalpha' to visual inspection, even though its
amplitude spectrum shows substantial contribution from the
alpha band. Such alpha rhythms may consist of mixtures of
both localized and widely distributed activity.
Larger amplitude frontal alpha often occurs as subjects
become more relaxed, for example, by employing relaxation
or meditation techniques. Alpha rhythms of unusually large
amplitude or exhibiting frontal dominance may be associated
with mental retardation and some types of epilepsy. Large
amplitude and dominant frontal alpha rhythm may also be
recorded in some coma and anesthesia states. In summary,
frontal alpha rhythms of moderate amplitude are common in
healthy relaxed subjects with closed eyes, but very large frontal
alpha is associated with disease or anesthesia. The physiologi-
cal relationships between these disparate alpha phenomena are
unknown, but they appear to share some underlying physio-
logical mechanisms since their frequencies and widespread
distributions are similar.
Coherence
Other dynamic measures involve a combination of location
and time measures. For example, the (normalized) covariance
of two signals is a correlation coefficient expressed as a func-
tion of time delay for characteristic waveforms recorded at the
two locations. Covariance is used in ERP studies of cognition.
A measure similar to covariance is the coherence of two signals,
which is also a correlation coefficient (squared). It measures
the phase consistency between pairs of signals in each fre-
quency band. Scalp potential (with respect to a reference)
recorded at many scalp locations over (say) a 1-min record
may be represented by eqn [ 1 ] . Consider any two locations
with time-dependent voltages V)(f) and Vy(£). The methods of
Fourier analysis may be used to determine the phases (j> p m and
4> p n j associated with (say) each 1-s period or epoch (indicated
by superscript p) of the full 60-s record. The frequency com-
ponent is indicated by subscript n and the two electrode loca-
tions are indicated by subscripts i and j. If the voltage phase
difference {</> p ,• - <£^) is fixed over successive epochs p ( phase
locked), the estimated EEG coherence between scalp locations
i and j is equal to 1 at frequency n. On the other hand, if
the phase difference varies randomly over epochs, estimated
coherence will be small at this frequency.
The low spatial resolution of conventional EEG limits the
information available using coherence (or any correlation
measure) between electrodes. High-resolution EEG methods
improve the spatial resolution of EEG, minimizing the smear-
ing of the potentials by volume conduction through the
head. High-resolution EEG coherence between one electrode
(labeled x) and a set of electrodes (1-8) at progressively larger
distances from posterior to anterior locations of the left hemi-
sphere is shown in Figure 4. (High-resolution EEG was esti-
mated by using the entire 111 channel data set.) The subject
was at rest with eyes closed to facilitate a robust alpha rhythm.
Power spectra for two midline channels of this subject are
shown in Figure 3, demonstrating two peaks in the alpha
band (9.5 and 10.5 Hz). Even the closest neighbor pair (x:l)
indicates very low coherence except at the coherence peaks at
9.5 and 10.5 Hz. At the next closest pair (x:2) a different pattern
is seen with stronger coherence at 10.5 Hz. The electrode pairs
involving prefrontal electrodes (x:7 and x:8) show higher
coherence at 9.5 Hz than 10.5 Hz. For two pairs (x:3 and x:6)
coherence is very low over the entire frequency range. The
highest coherences at 9.5 and 10.5 Hz involve temporal elec-
trodes (x:4 and x:5). For these pairs additional peaks can be
seen at the harmonics (19 and 21 Hz). In summary, coherence
estimates show significant spatial specificity and reveal much
more detailed patterns of synchronization of neuronal activity
across the brain.
Dynamic Behavior of Sources
EEG waveforms recorded on the scalp are due to a linear
superposition of contributions from billions of microcurrent
sources or, expressed another way, by thousands to millions of
columnar sources P(r', f) located in the cerebral cortex, as
indicated by eqn [2], However, the underlying physiological
bases for the dynamic behavior of the sources themselves are
mostly unknown. The 10-Hz range oscillations of alpha
rhythm, the 1-Hz range oscillations of deep sleep, and other
waveforms in the EEG 'zoo' must be based on some sort of
characteristic time delays produced at smaller scales. Such
delays can evidently be developed in cell assemblies ( neural net-
works) that cover a wide range of spatial scales. Locally gener-
ated activity in small networks and more globally generated
activity involving spatially extensive networks up to the global
scale of the entire cerebral cortex may be reasonably assumed.
The local network category includes the so-called thalamic pace-
makers that could possibly impose oscillations in specific fre-
quency ranges on the cortex ( local resonances). Other possible
mechanisms occur at intermediate scales between the local and
22 Electroencephalography
x:8
i
0.8
> 0.6
5
j
1 0.4
)
0.2
0
1
1
10 20 30 40
Frequency (Hz)
x:7
10 20 30 40 50
Frequency (Hz)
Frequency (Hz)
Frequency (Hz)
Figure 4 Coherence spectra from a 22-year-old female subject at rest with eyes closed (to maximize alpha coherence). Coherence was estimated
with T = 2 s (nf = 0.5 Hz) in a 60-s record. The head plot shows the location of 9 electrodes, labeled x and 1 through 8. Coherence spectra between
electrode x and each of the other electrodes 1-8 are shown, with increasing separations along the scalp. High-resolution EEG methods were used to
increase the spatial resolution of 128 EEG signals by removing the effect of smearing by the skull. This process can also remove genuine EEG
signals that are broadly distributed. The coherence spectra show complex patterns of coherence according to frequency and scalp locations, changing
even with small changes in frequency or location.
the global. These involve feedback between the cortex and the
thalamus or between specific cortical locations. Preferred fre-
quencies generated at intermediate scales may be termed
regional resonances. At the global scale, the generation of reso-
nant frequencies ( global resonances ) due to standing waves of
synaptic action has been proposed.
Delays in local networks are believed to be mainly due to
rise and decay times of postsynaptic potentials. By contrast, global
delays occur as a result of propagation of action potentials along
axons connecting distant cortical regions (corticocortical fibers).
Delays in regional networks may involve both local and global
mechanisms. A working conjecture is that local, regional, and
global resonant phenomena all potentially contribute to source
dynamics. However, the relative contributions of networks with
different sizes may be quite different in different brain states.
The transition from awake to anesthesia states is an example of
a local to global change. The ECoG changes from rhythms
depending strongly on location to rhythms that appear similar
over widespread cortical locations. Another example is the
desynchronization (amplitude reduction) of alpha rhythms
that occurs with eye opening and certain mental tasks.
A number of mathematical theories have been developed
since the early 1970s to explain the physiological bases for
source dynamics, that is, the underlying reasons for specific
time-dependent behaviors of the source function P(r', £). Dis-
tinct theories may compete with, or complement each other,
or both. Some common EEG properties for which plausible
quantitative explanations have emerged naturally from math-
ematical theories include the following observed relations:
frequency ranges, amplitude versus frequency, spatial versus
temporal frequency, maturation of alpha rhythm, alpha
frequency-brain size correlation, frequency versus corticocor-
tical propagation speed, frequency versus scalp propagation
speed, frequency dependence on neurotransmitter action, and
mechanisms for cross-scale interactions between hierarchical
networks. Because the brain is so complex, such theories must
involve many approximations to genuine physiology and
anatomy. As a result, verification or falsification of specific
theories for the physiological bases for EEG is difficult. How-
ever, such mathematical theories can profoundly influence
our general conceptual framework of brain processes and
suggest new studies to test these ideas. Furthermore, this
Electroencephalography 23
framework can strongly influence the design of cognitive
studies, for example, when estimating various measures of
functional connections between cortical regions such as EEG
coherence or ERP covariance.
Further Reading
Braitenberg V and Schuz A (1 991 ) Anatomy of the Cortex: Statistics and Geometry.
New York: Springer.
Gevins AS, Le J, Martin N, Brickett P, Desmond J, and Reutter B (1994) High
resolution EEG: 124-channel recording, spatial enhancement, and MRI
integration methods. Electroencephalography and Clinical Neurophysiology
90: 337-358.
Gevins AS, Smith ME, McEvoy L, and Yu D (1997) High-resolution mapping of cortical
activation related to working memory: Effects of task difficulty, type of processing,
and practice. Cerebral Cortex 7: 374-385.
Hamaleinen M, Hari R, llmoniemi RJ, Knutilla J, and Lounasmaa 0V (1993)
Magnetoencephalography: Theory, instrumentation, and applications to
noninvasive studies of the working human brain. Reviews of Modem Physics
65: 413-497.
Klimesch W (1996) Memory processes, brain oscillations and EEG synchronization.
International Journal of Psychophysiology 24: 61-100.
Malmuvino J and Plonsey R (1995) Bioelectromagetism. New York:
Oxford University Press.
Neuper C and Pfurtscheller G (2001) Event-related dynamics of cortical rhythms:
Frequency-specific features and functional correlates. International Journal of
Psychophysiology 43: 41-58.
Niedermeyer E and Lopes da Silva FH (eds.) (1999) Electroencephalography. Basic
Principles, Clinical Applications, and Related Fields, 4th edn. London: Williams and
Wilkins.
Nunez PL (1 995) Neocortical Dynamics and Human EEG Rhythms. New York:
Oxford University Press.
Nunez PL (2010) REST: A good idea but not the gold standard. Clinical
Neurophysiology 1 21 (1 2): 21 77-2180.
Nunez PL and Srinivasan R (2006) Electric Fields of the Brain: The Neurophysics of
EEG, 2nd edn. New York: Oxford University Press.
Nunez PL, Srinivasan R, Westdorp AF, et al. (1997) EEG coherency I: Statistics,
reference electrode, volume conduction, Laplacians, cortical imaging, and
interpretation at multiple scales. Electroencephalography and Clinical
Neurophysiology 103: 516-527.
Nuwer M (1997) Assessment of digital EEG, quantitative EEG, and EEG brain mapping:
Report of the American Academy of Neurology and the American Clinical
Neurophysiology Society. Neurology 49: 277-292.
Penfield W and Jasper HD (1954) Epilepsy and the Functional Anatomy of the
Human Brain. London: Little, Brown.
Picton TW, Bentin S, Berg P, et al. (2000) Guidelines for using human event-related
potentials to study cognition: Recording standards and publication criteria.
Psychophysiology 37: 127-152.
Scott AC (1995) Stairway to the Mind. New York: Springer.
Srinivasan R, Nunez PL, and Silberstein RB (1998) Spatial filtering and neocortical
dynamics: Estimates of EEG coherence. IEEE Transactions on Biomedical
Engineering 45: 814-825.
Srinivasan R, Winter W, Ding J, and Nunez PL (2007) EEG and MEG coherence:
Measures of functional connectivity at distinct spatial scales of neocortical
dynamics. Journal of Neuroscience Methods 166: 41-52.
Uhl C (ed.) (1999) Analysis of Neurophysiological Brain Functioning. Berlin:
Springer.
Embarrassment and Blushing
R J Edelmann, University of Roehampton, London, UK,
© 2012 Elsevier Inc. All rights reserved.
Glossary
Appeasement display A display by which individuals
seek to placate or pacify others or as a 'remedial display'
serving to mitigate others' negative reactions to behaviors
that might otherwise result in public humiliation or loss
of self-esteem.
Blushing Spontaneous reddening or darkening the
face, ears, neck, and/or upper chest caused by blood
vessels dilating and increasing blood volume in those
areas in response to social evaluation or public
scrutiny.
Embarrassment A state of emotional discomfort
resulting from a projection of an undesired image to
others present.
Face saving All coping actions, both verbal and nonverbal,
taken by both the embarrassed person and observers to
the event to overcome feelings of embarrassment in
an attempt to regain social composure.
Self-consciousness The focusing of attention upon oneself
either as a result of the situation (e.g., in the presence of
an audience) or as a result of individual differences in the
tendency to be self-attentive.
Self-presentation Attempts to control the image of
ourselves that we present to others.
Social phobia A marked or persistent fear in social
situations that the person concerned will behave in a
manner which is perceived by others as being embarrassing
or humiliating.
Introduction
Embarrassing blunders, faux pas, and improprieties are every-
day social occurrences. However, because embarrassment is
essentially a state in which a person has failed to maintain
social poise or a desired image in the presence of others, it is
an uncomfortable social state which people seek to avoid.
Because of the aversive nature of embarrassing situations peo-
ple try to avoid them when possible or, if embarrassment has
occurred, make efforts to restore identity or self-view, as well as
attempt to save face or preserve a desired identity in the eyes of
others present. In addition to external cues, which can relate
either to the nature of the social accident or to the reactions of
others to our faux pas, a feeling of embarrassment can also be
elicited by internal cues (notably blushing) in the absence of
an intrinsically embarrassing situation. Indeed, Arnold Buss
described blushing as the hallmark of embarrassment and
numerous authors including Charles Darwin writing in 1872
have agued that the blush is the primary nonverbal signal of
embarrassment. While there is not a one-to-one link between
embarrassment and blushing, it is possible to be embarrassed
without blushing and to blush without being embarrassed;
nevertheless, for some people the fear of blushing can be a
socially debilitating experience. Many people who suffer from
social phobia are particularly concerned that they will blush and
that others will then interpret this as a sign of social ineptitude.
Theoretical Explanations for Embarrassment and
Blushing
Although embarrassment is a common and dramatic experi-
ence with which nearly everyone is familiar, there is a lack of
agreement about both the precise psychological circumstances
underlying embarrassment and the function of blushing.
Indeed, proposals about the factors underlying embarrassment
do not link precisely to those delineating the function of
blushing. Three broad theoretical explanations for embarrass-
ment have been advanced; the first focuses upon self-presenta-
tional concerns and fear of negative evaluation; in this context,
blushing is also explained as being evoked following undesired
attention; the second emphasizes the individual's flustered
uncertainty about how to proceed following a social accident
and the third focuses upon anxiety and worry about the need to
create a desired impression. Additional explanations for blush-
ing are that it serves as an appeasement display by which indi-
viduals seek to placate or pacify others or as a 'remedial display'
serving to mitigate others' negative reactions to behaviors that
might otherwise result in public humiliation or loss of self-
esteem. These latter accounts are discussed in more detail in
section 'Blushing and the Expression of Embarrassment'.
Theories focusing upon self-presentation concerns suggest
that embarrassment has to do with a failure to present a desired
image to others whom we regard as evaluating our perfor-
mance. This assumes that the individual concerned is moti-
vated to make a particular impression on others. Falling over in
front of friends is likely to cause us less concern about the
impression we create than falling over just as we are about to
introduce a guest of honor to a large audience at an important
function. Factors that increase the negative consequences of
failure, such as larger, more competent audiences, novel or
ambiguous contexts, or facets of one's own personality, are
likely to increase our desire to protect our image before others.
A further assumption, however, is that the prevailing cir-
cumstances during an interaction do not allow us to present a
desired image to others present. The latter circumstances
involve the violation, witnessed by others, of some taken-
for-granted social rule. This, in turn, assumes that the person
concerned has violated the rule unintentionally while at the
same time being aware that a violation has occurred. Someone
unaware of the rules of conduct will remain unembarrassed
by their 'inappropriate' behavior no matter how concerned
24
Embarrassment and Blushing 25
they are with the evaluative reactions of others, unless others
present make it clear that their behavior is inappropriate.
If we are unaware of the rules of etiquette at a social function,
transgressing such rules will not cause us concern unless
others draw our attention to our failings. Similarly, someone
aware of the rules but unconcerned with the evaluative reaction
is likely to remain unembarrassed by social mishaps. Con-
versely, someone acutely concerned with negative audience
evaluations is likely to be embarrassed by seemingly trivial
events. It is only when our knowledge of, and motivation to
concur with, these rules exists that embarrassment is a conse-
quence of social failure.
Others have argued that it is not self-presentational con-
cerns per se that cause embarrassment but flustered uncertainty
about how to proceed once an unexpected or disruptive social
event has occurred. Thus, forgetting someone's name or falling
over in public disrupt the working consensus of identities that
exist in social interaction, leaving a new consensus to be cre-
ated. Until such a new consensus exists, awkwardness or flus-
tered uncertainty predominates. Although such a reaction may
well be important, flustered uncertainty is itself likely to be
directly related to a concern with the evaluation of others.
Appearing to be flustered or uncertain is not an image we
would wish to present to others and hence may be embarras-
sing in its own right.
Finally, some have pointed out that anxiety and worry about
whether one has actually created, or indeed will create, an unde-
sirable impression is a central feature of self-presentational con-
cerns. Indeed, some people seem to be so concerned about
the impression they will create that they suffer extreme embar-
rassment in the absence of any clearly defined evoking event.
Indeed, they may be so concerned that they might behave in a
way which is humiliating or embarrassing that they actively seek
to avoid social situations. Indeed, one of the defining features
of social phobia as detailed in the Diagnostic and Statistical
Manual of the American Psychiatric Association is a marked
and persistent fear of social or performance situations in which
embarrassment may occur.
Clearly, there are common features to the three theories
described. Although flustered uncertainty and anxiety about
one's social performance seem to be important aspects of
embarrassment, self-presentational concerns are central. Thus,
the emphasis in each theory is that embarrassment has to do
with the actual, or anticipated, occurrence of an unwanted event
that communicates undesired information about oneself to
real, or imagined, others, whom we regard as evaluating our
performance.
Causes of Embarrassment and Blushing
There have been a number of attempts to collect descriptions
of, and categorize, embarrassing events on the basis of the
situations involved; fewer studies have sought to classify
blush-inducing rather than embarrassing circumstances. One
factor illustrated by the former studies is the sheer variety of
embarrassing circumstances people document. It is perhaps
not surprising therefore that the various attempts to systemati-
cally categorize embarrassing acts produces a variety of differ-
ent solutions. The broad categories do though correspond with
categories of situations which elicit blushing. The first two
categories describe inappropriate identity (e.g., uncontrollable
laughter or tears) and loss of poise. The latter includes loss
of physical poise (e.g., tripping over furniture), impropriety
(e.g., failure to control bodily behaviors, for example, when
fainting), forgetting someone's name or introducing them by
the wrong name, and awkward interactions (e.g., asking inap-
propriate or unfortunate questions or simply not knowing
how to proceed). The third category concerns false identity
(i.e., telling someone untruths about oneself and being
found out), the fourth consists of breaches of privacy (e.g.,
unintentional exposure of bodily parts, touching or coming
too close to someone, or revealing feelings that the person
wished to hide), while the fifth involves being made to feel
conspicuous as when over-praised (e.g., discomfort occasioned
by excessive compliments). Still others have included a cate-
gory of empathic or vicarious embarrassment to describe
instances when we feel embarrassed for another person who
is in a seemingly embarrassing situation (e.g., when an actor
forgets their lines). Although there are clearly a number of ways
of classifying embarrassing events, these tend to reflect the
methodology used; there do not tend to be radical differences
in the general descriptions of embarrassing events obtained.
Thus, some categorization schemes refer to specific actions
(e.g., tripping and falling) while others pinpoint dimen-
sions (e.g., who is embarrassed, as in empathic embarrassment).
Therefore, rather than considering categories of acts, a more
useful strategy for classifying embarrassing episodes is to con-
sider psychologically meaningful dimensions. Two such
dimensions appear to be of particular importance: who is
embarrassed by the events and who is responsible for causing
the embarrassment. The first dimension, who is embarrassed,
considers whether it is an individual and/or social group,
both (or all) the interactants, or the bystander/ viewer. The
second dimension, who is responsible for causing the embar-
rassment, considers whether it is the embarrassed person them-
selves (i.e., the actor), someone present and involved when
the embarrassment occurs (i.e., the observer) and someone
present but not directly interacting (i.e., person viewed).
By combining these two dimensions, four broad categories of
situation can be recognized: (i) those in which the individual
and/or social group is responsible for their own embarrass-
ment (individual/group behavior, actor responsible); (ii) those
in which the interaction itself is embarrassing (interactive
behaviors, shared responsibility); (iii) situations in which
embarrassment is caused by an observer (individual/group
behavior, observer responsible); and (iv) an empathic reaction
of embarrassment to someone else's behavior (bystander/
viewer, person viewed responsible).
The actual types of embarrassing and blush-inducing
episodes described by earlier classification schemas can readily
be incorporated by the dimensional categorization approach.
Examples from the first category (individual/group behavior,
actor responsible) include loss of poise (e.g., forgetting some-
one's name), impropriety (e.g., wearing the wrong clothes to a
social function), and false identity (i.e., telling someone untruths
about oneself and being found out). Examples from the second
category (interactive behavior, shared responsibility) include
any awkward interaction where neither person is sure what to
say, when there are long uncomfortable silences, unsuccessful
26 Embarrassment and Blushing
attempts by both parties to end the conversation, and so on.
Examples form the third category (individual/group behavior,
observer responsible) include conspicuousness (e.g., being the
'victim' of people singing "happy birthday to you" or when
being complimented or congratulated), being the victim of
teasing, or when people make public our past embarrassing
predicaments. Examples of the final category (bystander/viewer,
person viewed responsible) include any occasion of empathic
embarrassment when we feel embarrassed because of someone's
or something's behavior (e.g., going to the zoo and seeing
animals engage in sexual behavior).
Central to each is a desire to present oneself in a way that
meets one's own standard for self-presentation and/or a desire
to avoid focusing attention toward the public aspect of oneself.
Thus, any event that involves the transgression of an unwritten
social rule (e.g., falling over in public, dressing inappropriately
for social occasions, awkward interactions), results in both the
presentation of an undesired image and self-attention. Being
made the centre of attention (e.g., through teasing) inevitably
draws attention to the public aspect of ourselves and a concern
with the image we are likely to present.
Interestingly, blushing, which is often regarded as the hall-
mark of embarrassment, may actually make us more self-aware
and lead us to assume we are presenting an image (i.e., embar-
rassment) that is discrepant from the image we wish to present.
We tend to assume that people feel embarrassed when they
blush, although visible facial reddening also occurs with other
emotional states such as anger. Thus, it is possible to make
someone embarrassed simply by telling them they are blush-
ing, even in nonembarrassing situations. Blushing may then be
sufficient to generate embarrassment in the absence of specific
situations that are defined as embarrassing. People who are
over-attentive to internal and expressive cues may be particu-
larly prone to generate a feeling of embarrassment from a
perception of their own blushing. Because of the negative
experience of embarrassment with which blushing is asso-
ciated, this may, in some cases, lead to a chronic fear of blush-
ing. This issue is developed further in the following section.
Blushing and the Expression of Embarrassment
Embarrassment is marked by a well-defined set of behaviors.
Eye contact is reduced, body movements, speech dysfluencies,
and smiling all increase, and blushing is likely to occur. This
reaction seems nearly universal, although the intensity and
relative importance of different components of the reaction
seem to differ across cultures. Indeed, the display of embarrass-
ment is readily recognizable by others. But why should an
emotion we wish to hide be so evident to others? There are a
number of suggestions for the functions of behaviors asso-
ciated with embarrassment.
The desire to look away or hide behind one's hands when
embarrassed has been studied in laboratory experiments as
well as being widely referred to in people's reports of their
embarrassment. A common feeling when embarrassed is a
wish to disappear or escape from the situation. Unfortunately,
departing the scene of an embarrassing mishap is rarely possi-
ble to achieve without attracting even more attention to one-
self. Hence, as we are unable to increase our physical distance
from others, we may use avoidance of eye contact, turning
away, or covering our eyes as a way of increasing our psycho-
logical distance from them. A further possible explanation is
that people look away when embarrassed to avoid seeing the
signs of rejection on the faces of other people. Alternatively,
reduced eye contact may be a reflection of tension or anxiety
associated with embarrassment. Such nervous responses may
also include manipulative gestures (such as fingering one
clothes, hair, or surrounding objects), turning one's body
away and stammering.
There are a number of ways in which anxiety associated
with embarrassment might interfere with speaking. First, pre-
occupation with the reason for our embarrassment may make
it difficult to pay attention to what we wish to say. Alterna-
tively, increased anxiety may make us pay too much attention
to our speech with the result that we stumble over our words.
Finally, anxiety may interfere with normal breathing, with
shorter, faster respiration interfering with speech output.
A 'silly' self-conscious grin and a 'strangled' laugh are also
common reactions to embarrassment. The fact that, during
embarrassment, smiling usually occurs while the person
looks away (aversion of gaze occurs when the smile reaches
its peak intensity), distinguishes it from a smile of pleasure
(when looking away is more likely to occur when the smile
intensity is diminishing). In addition muscular actions asso-
ciated with smiling such as pressing the lips together, stretch-
ing, biting or sucking them, puffing the cheeks or bulging the
tongue occur more frequently, for longer and in greater num-
ber in relation to smiling when embarrassed than when such
actions occur in association with amusement. There are four
possible explanations for such a reaction. First, people may use
smiling in an attempt to hide or conceal their embarrassment;
such attempts may occasionally succeed, as embarrassment can
be mistaken for amusement. Second, smiling and laughter
when embarrassed may serve as an admission of the person's
inappropriate behavior, while at the same time attempting to
reduce the importance of the event giving rise to the embar-
rassment. Third, laughter may serve to reduce the tension or
anxiety engendered by the embarrassing event. Fourth, it may
be a purposeful nonverbal strategy used by the embarrassed
person to change the meaning and focus of the situation in an
attempt to save face. By diminishing the importance of the
event, the identity 'victim' can be transformed to 'co-actor'
and the label 'embarrassing' transformed to 'humorous.'
Finally, the 'silly grin' of embarrassment may be related to the
submissive, appeasement grin which occurs in other primates
when threatened. Embarrassment presents a threat to the per-
son's sense of self and, in such circumstances, smiling may
serve to conciliate a potential 'aggressor.'
Although it is possible to feel embarrassed without visibly
blushing and to visibly blush without experiencing embarrass-
ment, blushing and embarrassment have been inextricably
linked by a number of authors. Although the faces of nonhu-
man primates redden or 'flush' with rage, blushing is a pecu-
liarly human expression that is restricted to certain specific
regions of the body. The 'blush region' is located in the face,
ears, neck, and upper chest. The redness itself is caused by the
bed of capillaries close to the skin filling with blood in order to
allow the body to lose heat. It seems that the blood vessels in
the face and upper body differ structurally from blood vessels
Embarrassment and Blushing 27
in other parts of the body. In addition, people seem to differ
with regard to their degree of response, hence, individual dif-
ferences in actual intensity of blushing. But, how might blush-
ing and embarrassment be interlinked?
One possibility is that people use their own expressive
behavior to interpret their inner feelings. Thus, skin blood-
flow changes associated with blushing may play a significant
part in influencing the intensity of reported embarrassment.
Our attention shifts between ourselves and the environment
on a continual nonrandom basis. If a social accident has
occurred, attention may initially be directed toward an evalua-
tion of the event itself. This may be followed quickly by an
inward focus of attention directed specifically toward those
aspects of the self that are presumed to be associated with
embarrassment, most notably blushing. The subsequent envi-
ronmental focus of attention may be directed toward a search
for the evaluation of others, again followed by an inward focus
toward our blushing. The fact that the process is rapidly
repeated serves to explain why blushing and embarrassment
escalate. Further, for those who are particularly sensitive to, or
predisposed to attend to their bodily reactions, the sequence of
reactions might be initiated internally, perhaps by a mere
suspicion that one is blushing, rather than by a specific external
event. Reactions to the suspicion that one might be blushing
could explain why people can be made to blush by telling them
that they appear to be blushing even when they are not. Such
information leads them to infer that others might infer they are
embarrassed, thereby leading them to blush.
Blushing is additionally curious in that it is a clearly visible
sign of an emotion that, in most cases, we would prefer to
conceal. Its very visibility has led some to argue that blushing is
indeed communicative and acts as a 'nonverbal apology.' It is a
visible statement to others that we are aware that our behavior
was untoward, inappropriate, or inadequate and, because we
care about other people's evaluations, our blushing is a way of
seeking forgiveness from other group members. Indeed, stud-
ies indicate that those who blush after a social accident are
evaluated more positively than those who do not do so.
It is also possible that blushing, like the silly grin, serves a
function similar to nonhuman appeasement gestures. For non-
human primates, appeasement frequently results in loss of
interest in the victim by the threatening animal. One purpose
of blushing may, paradoxically, be to reduce attention from
observers. Although some people may deliberately embarrass
or humiliate others, or deliberately draw attention to another's
blushes, in general we tend to avoid focusing on people when
they are blushing.
Individual Differences in Embarrassment
People differ markedly both in their tendency to blush and in
their reported frequency of embarrassment experienced. Four
factors related to such differences are examined here: age,
gender, personality, and culture.
Age
Embarrassment is systematically built into our social system.
Societies everywhere have unwritten social rules which regulate
our everyday behavior. It seems that the aversive consequences
of embarrassment are such that one reason for our tendency
to follow the rules, conventions, and behavior expected by
others is our fear of the possibly embarrassing consequences
of not doing so. Indeed, laughing at, ridiculing, or in short,
embarrassing another person for rule-breaking is likely to
result in attempts to avoid such rule-breaking in future. Some
have argued that, via such mechanisms, embarrassment exerts
a powerful influence on the acquisition of social rules in
children and adolescents. Thus, children are introduced to
modesty, self-control, and manners by means of teasing, laugh-
ter, and ridicule for mistakes. During adolescence there is an
increasing desire to manage the impressions we create for
others and hence we are increasingly motivated to avoid
being singled out for laughter and ridicule. Embarrassment
thus becomes a particularly powerful method by which social
mles can be acquired.
Although there is limited research pertaining to develop-
mental aspects of embarrassment, there seem to be discernible
age related changes. Embarrassment seems to occur infre-
quently in very young children, research suggesting that the
first signs seem to be around the age of 2 years. Over the next
few years the tendency to experience embarrassment increases,
until by the ages of 5-7 embarrassment is experienced as
frequently as in adulthood. A peak period of embarrassment
is during adolescence. There is then a decline in reported
embarrassment in later adult life.
The experience of embarrassment is related to the emer-
gence of the social self and associated self-consciousness.
Some have argued that this occurs at about 7 years of age, an
increase in self-conscious awareness then occurring during
early adolescence with a peak at around 13 or 14 years of age,
followed by a gradual decline. Thus, while children as young as
two may show embarrassed behaviors as judged by others
whether they personally experience the emotion of embarrass-
ment is a different matter.
Central to age-related changes in felt embarrassment is the
extent to which children are aware of how the impression they
create is viewed by others. At a young age a child's ability to
view things from someone else's perspective is limited. Their
behavior is more likely to be directed by what appears to be
right to him or her than by what appears to be right to others.
Although they may well be aware that certain behaviors pro-
voke ridicule or laughter they may not be aware of why such
reactions occur. At this age they are thus likely to experience a
fairly 'primitive' form of embarrassment. By the age of 5-7
children are more aware of the way in which the impression
they create is viewed by others. They become more concerned
with creating an impression appropriate to the circumstances
and which will be viewed positively. Thus, for an older child, a
social transgression or social accident is likely to result in
'mature' embarrassment, similar to that experienced by adults.
The peak period of embarrassment is during adolescence
which parallels the peak of self-presentational concerns and
self-consciousness which occurs at this time. Conformity to
peer expectations, marked by a need to wear the 'right' clothes
and listen to the 'right' music is especially important during
adolescence. The desire to win approval from others is an
important goal. In addition, adolescence is associated with
marked changes in physical and psychological makeup.
28 Embarrassment and Blushing
No longer being a child, but not yet being an adult, the adoles-
cent has yet to acquire the necessary repertoire of behaviors to
play out an adult role. Lack of knowledge can lead to indeci-
siveness and lack of confidence and hence more likelihood that
behaviors will be viewed negatively or have embarrassing
consequences.
There is then a gradual decline in self-consciousness after
adolescence which is paralleled by both a decrease in experi-
enced embarrassment and a decrease in the tendency to blush.
With increased age, novel experiences and hence the potential
for embarrassment are likely to decline; this is paralleled by an
apparent decrease with age in the propensity for blood vessels
in the face to dilate.
Gender
The lay view is that women are more likely to blush and are
more readily embarrassed than men. Certainly, within a his-
torical context, coy, shy reactions such as exhibited by the
'blushing bride' were considered appropriate signs of modesty
in women. In general, however, such stereotypical differences
are not supported by research findings. Although there are
some gender differences in relation to circumstances associated
with embarrassment and while women in comparison with
men do report experiencing more intense embarrassment and
a greater propensity to blush there are few differences with
regard to actual skin temperature changes in potentially
blush-inducing situations. Situational differences in reported
embarrassment and blushing no doubt reflect differing sociali-
zation experiences while differences in reported embarrass-
ment and propensity to blush may merely reflect the greater
propensity for women to express their feelings.
Personality
A number of research studies indicate that some individuals
are more prone to experience embarrassment and more likely
to report a propensity to blush than others. A central factor
is the degree of concern the person has for others' impressions
of them. Those concerned with projecting an image which
engenders social approval, or at least avoids engendering
social disapproval, are more likely to be more readily embar-
rassed. Furthermore, those who are more readily embarrassed
tend to be intensely affected by the slightest social transgres-
sion, more likely to report enduring embarrassment even
years later at the recall of the event, are more likely to feel
empathic embarrassment and are more likely to report a high
propensity to blush. Indeed, a proneness to embarrassment is
associated with feelings of being observed and a high regard for
how one is viewed by others. Indeed, this is a pattern similar to
those who experience social anxiety and social phobia. It is
thus perhaps not surprising that the degree to which people
fear social situations in general is also related to the degree
of reported blushing and embarrassment. At the extreme,
people troubled by chronic fear of blushing obtain scores on
measures of social anxiety similar to those clinically diagnosed
as socially phobic; indeed, it is now widely recognized that
for a subgroup of social phobics at least fear of blushing is
their central concern.
Cultural differences
There seems to be clear evidence that blushing is universal and
that embarrassment is expressed in a similar way across cul-
tures. Blacks and other dark-skinned people experience vasodi-
lation in the kinds of social situations that provoke blushing in
whites. However, blushing in the former appears to be marked
by a further darkening of the skin rather than the reddening
which occurs in lighter skinned people. The occurrence of
blushing is reported with greater frequency in some cultures
(e.g., the United Kingdom) than in others, while averting eye
contact and laughter which is reported as frequently occurring
in some cultures is reported as occurring less frequently in
others (e.g., Italy). These apparent differences tend to be
derived from self-reports, rather than from actual observations
or direct measures in the cultures concerned, and hence any
variations may simply reflect reporting differences. However,
if the differences are real, they may involve variations across
cultures in the meaning attributed to embarrassing events.
Thus, embarrassment might be a more serious matter in certain
cultures and hence laughter or smiling would be less likely to
occur. There may also be variations in skin thickness across
cultures and hence variability in the visibility of blushing. It is
also important to bear in mind that different situations are
likely to elicit embarrassment and blushing across cultures.
Face Saving and Embarrassment
Because of its aversive and disruptive nature, people generally
attempt to avoid embarrassing situations when possible. How-
ever, as complete avoidance is rarely possible, we often make
attempts to redress the situation in order to maintain face,
regain social approval, and restore the interaction following
an embarrassing event. In dealing with embarrassing events
four main face-saving strategies are recognized: apologies,
accounts, avoidance/escape (evasive facework), and the use of
humor. Two additional strategies, referred to as remediation
and hostility, have also been discussed. Remediation involves
offers or requests of help or comfort, expressions of concern, or
offers to correct the problem, for example, clearing up after an
accident. In contrast to situations inviting remediation, there
are some circumstances in which anger or hostility is used as a
method of coping with embarrassment. In general, however,
the latter reaction is restricted to circumstances in which the
person has been criticized or teased by others with the deliber-
ate intent of causing embarrassment. In this context, the goal of
intentional embarrassment is to negatively sanction the other's
behavior, to establish power, or simply self-satisfaction. The
embarrassed person may respond with hostility in an effort to
keep the same behavior from occurring again.
The simplest way of dealing with embarrassment is to offer
an apology, such as 'I'm sorry' or 'pardon me.' This serves to
acknowledge blameworthiness for the undesired behavior in
the hope that a minor discretion can be passed over quickly
and forgotten by those concerned. Indeed, the very act of being
embarrassed can itself serve as an apology, the visual sign of
blushing indicating that the person accepts responsibility for
the untoward act. Several theorists have sUessed the functional
properties of expressing embarrassment and have argued that
Embarrassment and Blushing 29
these expressions may help restore the person's public image
after a mishap or transgression. That is, blushing or publicly
conveying other signs of embarrassment can show that the
person recognizes that they have committed a social or moral
transgression, and that they regret having done so. Accord-
ingly, it has been argued that blushing has appeasement-
related properties which can actually positively influence
observers' judgments. An interpretation of a sign of embarrass-
ment as an apology is most likely in a situation in which the
person's undesired behavior may have caused discomfort to
someone else who was present. Brazening out the situation
may serve to increase both parties' embarrassment, while a
display of embarrassment accompanied by laughter may dif-
fuse the situation.
While an apology offers an admission of responsibility, an
account is an attempt to explain the undesired behavior.
Accounts can take two forms: excuses and justifications. The
former consists of attempts to play down one's own contribu-
tion to the event while at the same time admitting that the
behavior was wrong. A justification similarly admits some
responsibility for the event while at the same time containing
an element of denial that there are any negative consequences
of the behavior. Categories of accounts involve appeals to
environmental circumstances, lack of intent, highlighting per-
sonal characteristics, or scapegoating. Thus, if we embarrass
ourselves by tripping in front of others, we might comment
'that's the trouble with slippery floors' (environmental circum-
stances), 'I didn't mean to do that' (lack of intent), 'that's
typical, I seem to have two left feet' (personal characteristics),
'the cleaners have polished the floor again' (scapegoating).
Categories of justification include 'self-fulfilment,' which
justifies the action on the grounds that it leads to some form
of enlightenment, and 'effect misrepresented,' which justifies
the action because the consequences were trivial or because no
harm was done. Because they add to excuses and attempt to
minimize or deny the undesirability of the event, justifications
may be too strong for dealing with minor transgression asso-
ciated with embarrassment. An attempt to justify ones actions
may serve to prolong, heighten, and add to the implications of
the embarrassing event.
Although a verbal apology or account may be used fre-
quently to deal with embarrassing events, there are occasions
when they may be inappropriate or may fail to deal adequately
with the situation. In addition, the person concerned may feel
unable to offer a verbal apology or account for their action.
In such circumstances, avoidance may be a common reaction.
This may take the form of either withdrawal subsequent to the
embarrassing event (escape) or attempts to avoid encounters
in which embarrassment might occur (avoidance). However,
physically retreating from the situation, rather than achieving
the goal of averting attention, is likely to draw further attention
to the embarrassed person. There are occasions, however,
when averting an awkward event can be achieved by changing
the topic or remaining silent rather than by escape per se. It is
also clear that some people will go to extreme lengths to avoid
social situation due to a fear that they will behave in a way
which is embarrassing. This is particularly true for people with
a chronic fear of blushing. Avoidance in this case is more an
attempt to avoid losing face through blushing inappropriately
rather than an attempt to save face.
A further strategy involved in coping with embarrassment
involves the use of humor or laughter. Laughter serves to
reduce the tension, while joking can turn a potential loss of
social approval into a gain in social approval.
In order for a face-saving strategy to be successful, it must
be acceptable, reasonable, and adequate. In the case of minor,
nonserious events, humor and laughter may be appropriate.
As events increase in their significance, the likelihood that
apologies and excuses will be used with greater frequency
also increases. In general, however, apologizing is the most
appropriate response regardless of the type of embarrassing
situation. Excuses and justifications tend to be used infre-
quently, except in social situations involving forgetfulness or
awkwardness. In addition, research findings tend to indicate
that the face-saving strategy used will depend not only on the
type of embarrassing event, but also the gender and status of
those involved. In terms of gender differences, women are
more likely than men to offer an apology or excuses outlining
mitigating circumstances. In addition, women are more likely
to offer help and more likely to act on such offers, if their
embarrassing actions have caused potential distress to others.
When the embarrassing event is not serious, people of higher
status, in comparison to those of lower status, are less likely
to acknowledge responsibility for the event or to attempt to
account for their actions.
Interestingly, there seems to be little variation across cul-
tures in the type of face-saving strategy used. Both avoidance
and a simple expression of embarrassment are common reac-
tions to all types of embarrassing events across cultures. The
main variation seems to be in the use of humor, which may be
almost absent in those cultures which have stronger sanctions
for social rule-breaking.
Observer Reactions
When someone is embarrassed, others present are likely to
react to and become involved in that person's embarrassment.
Each may affect the other so that embarrassment can either
spiral or be mutually controlled. In seeking to offer assistance,
observers may excuse or explain the embarrassed person's
behavior, 'playing along' with them, for example, by the use
of humor or an offer of support or empathy. Such reactions
may be used alone or in combination, and seem to be relatively
constant across cultures.
The most frequently reported reactions from observers are
support and empathy. Indeed, these may often be used in
combination, most usually with a supportive comment fol-
lowed by a sign of empathy. Support includes any com-
ment or nonverbal expression intended to reassure the
person that they are still viewed positively despite the embar-
rassing event. A typical comment might be 'don't worry,
it's not important,' while typical gestures include a warm
smile or friendly touch. Empathic reactions involve com-
ments intended to assure the embarrassed person that the
event that occurred could have happened to anyone. A typical
comment might be 'I know how you feel, I'm always doing
the same thing.' Such a reaction can also be a useful way
for the observer to show that they accept an apology offered
by the embarrassed person.
30 Embarrassment and Blushing
Humor is also used by observers with moderate frequency.
However, humor and laughter will clearly have different mean-
ing dependent upon whether the observer laughs alone or
whether both the observer and embarrassment person laugh
together. There is a fine dividing line between 'laughing with,'
and 'laughing at' someone. If the embarrassed person fails to
join in with the observer in their attempt to help, they may well
end up being the butt of the joke rather than helping to make
light of the event. In such circumstances embarrassment is
likely to be intensified.
A similar intensification of embarrassment is likely to occur
if the observer fails to respond or responds in a manner which
is interpreted negatively by the embarrassed person. As with
humor, a lack of response can be interpreted as a sign of lack of
support. The difference between a positive or negative reaction
is often signaled by subtle behaviors such as a fleeting smile or
facial expression acknowledging the event, or even an appear-
ance of embarrassment from the observer. Assuming that the
observer wishes to provide assistance in remedying the situa-
tion, their attempts can fail either because they do not know
how to respond or because their attempts backfire.
There are also occasions when an observer may seek to
intentionally increase another's embarrassment. For example,
if they feel that the embarrassed person has not shown suffi-
cient 'remorse,' if the embarrassing behavior has disturbed
others present, or if the embarrassed person was apparently
responsible for the event. Such deliberate intensification of
another's embarrassment may be achieved either by drawing
attention to the event or by expressing reproach. The latter may
involve criticism or correction. Clearly, such reactions may well
serve to further disrupt social interaction.
As well as instances when an observer may seek to intensify
another's embarrassment, there are occasions when an
observer may seek to intentionally initiate embarrassment in
others. Intentional embarrassment is frequently used as a sanc-
tion to teach children correct or tactful behavior. Men, in
comparison to women, are more likely to intentionally embar-
rass children and subordinates. This may reflect the fact that
males are stereotypically more competitive and more likely to
engage in controlling behaviors. Intentional embarrassment
also occurs frequently among same sex friends, often as part
of a 'game' to test each other's social poise.
Coping with Blushing
While embarrassment is a common but uncomfortable state
which can disrupt social interaction, chronic blushing or fear
of blushing can severely disrupt the lives of those concerned.
Blushing seems to cause most distress when it occurs in the
absence of a clearly defined embarrassing event, for example,
when someone fears that their red face will be taken as a sign of
shyness or social anxiety. Chronic blushers are characterized by
two facets of self-consciousness, they are over-concerned with
both their internal sensations and thoughts, as well as obser-
vers' subsequent reactions to their blushing. Blushing causes
a particular problem in social situations because we imagine
others can see the blush which we assume we are displaying but
which we are unable to see for ourselves. In some instances, the
fear of blushing can be psychologically debilitating. For many
people such fear of blushing can have devastating effects
leading to social withdrawal, friendship, and work related
difficulties and a state of panic when faced with what are to
other people quite ordinary social encounters. Many chronic
blushers are unable to cope with their difficulty, resorting
to medication or surgery. A number of authors have noted
that fear of blushing is one of the core concerns for at least
a subgroup of those who are socially phobic.
One of the key findings from a number of different research
studies is that there does not seem to be a perfect relationship
between the extent to which people report blushing or fear of
blushing with actual measured changes in skin temperature
or facial blood flow. In other words, fear of blushing is not
necessarily to do with external signs that are visible to others
but seems rather more related to our thoughts and evaluations.
Thus, a fear of the prospect of blushing provokes anticipatory
anxiety thereby increasing the possibility that other signs of
anxiety (trembling, shaking, stammering, sweating) will be
displayed during the ensuing social encounter. This further
increases feelings of discomfort and anxiety which in turn is
further fuelled by the assumption that others present can see
the signs of our anxiety and in particular our red face. Such
fears can lead to increasing avoidance of social situations.
Such findings have clear implications for treatment. If fear
of blushing, or a feeling that you are blushing intensely or
frequently is not necessarily related to actual blushing, then
the focus of treatment should not be actual blushing but the
underlying psychological parameter, that is, fear or anxiety
about blushing. Appropriate cognitive behavior therapy
aimed at helping the person cope more effectively with their
anxiety is thus the key and has been shown to be effective.
Because the negative pattern of thinking associated with blush-
ing is sufficient to cause embarrassment in the absence of a
clearly defined embarrassing event, the most effective strategy
for dealing with blushing is for the person to modify their
thinking pattern. This can be achieved by reevaluating and
redirecting attention away from oneself and blushing.
Thus, while everyday social embarrassment may be a com-
mon but uncomfortable psychological state which can disrupt
social interaction, a chronic fear of blushing can severely
disrupt the lives of those concerned. While a number of
strategies can be used to cope with social embarrassment, the
more significant the situation and the more concern we have
about our reaction the more difficult the embarrassment is to
deal with. At the extreme end of the continuum are those
who live in a constant fear of blushing and/or being seen as
behaving in an embarrassing manner. For such people seeking
appropriate therapeutic help may be the most appropriate
course of action.
See also: Bullying; Facial Expression of Emotion; Self-Esteem; Social
Anxiety Disorder.
Further Reading
Bogels SM and Lamers CT (2002) The causal role of self-awareness in
blushing-anxious, socially-anxious and social phobic patients. Behaviour
Research and Therapy 40: 1367-1384.
Embarrassment and Blushing 31
Bogels SM and Voncken M (2008) Social skills training versus cognitive therapy for
social anxiety disorder characterized by fear of blushing, trembling, or sweating.
International Journal of Cognitive Therapy 1:138-150.
Crozier R (2006) Blushing and the Social Emotions. Basingstoke: Palgrave Macmillan.
Dijk C, de Jong PJ, and Peters ML (2009) The remedial value of blushing in the context
of transgressions and mishap. Emotion 9: 287-291.
Drummond PD and Lim HK (2000) The significance of blushing for fair-and
dark-skinned people. Personality and Individual Differences 29: 1123-1132.
Edelmann RJ (1987) The Psychology of Embarrassment. Chichester: Wiley.
Edelmann RJ (1994) Embarrassment and blushing: Factors influencing face saving
strategies. In: Ting-Toomey S (ed.) The Challenge ofFacework: Cross Cultural and
Interpersonal Issues. Albany, NY: State University of New York Press.
Edelmann RJ (2001) Blushing. In: Crozier WR and Alden LE (eds.) International
Handbook of Social Anxiety. Concepts, Research and Interventions Relating to the
Self and Shyness. Chichester: Wiley.
Edelmann RJ (2004) Coping with Blushing, 2nd edn. London: Sheldon Press.
Gerlach AL, Wilhelm FH, Gruber K, and Roth WT (2001) Blushing and
physiological arousability in social phobia. Journal of Abnormal Psychology
110: 247-258.
Keltner D and Busswell BN (1997) Embarrassment: Its distinct forms and appeasement
function. Psychological Bulletin 122: 250-270.
Leary MR, Britt TW, Cutlip WD II, and Templeton JL (1992) Social blushing.
Psychological Bulletin 112: 446-460.
Miller RS (1996) Embarrassment. Poise and Perils in Everyday Life. New York:
Guildford Press.
Miller RS (2001) Embarrassment and social phobia: Distant cousins or close kin?
In: Hofmann SG and DiBartolo PM (eds.) From Social Anxiety to Social Phobia:
Multiple Perspectives. Needham Heights MA: Allyn & Bacon.
Miller RS (2007) Is embarrassment a blessing or a curse? In: T racy JL, Robins RW, and
Tangney JP (eds.) The Self-Conscious Emotions: Theory and Research. New York:
Guildford.
Sharkey WF, Kim MS, and Diggs R (2001) Intentional embarrassment: A look at
embarrassors’ and targets' perspectives. Personality & Individual Differences
31: 1261-1272.
Relevant Website
http://www2.hawaii.edu/sharkey/embarrassment/embarrassment_references.html -
A bibliography of embarrassment research complied by Professor W. Sharkey,
University of Hawaii.
Empathy
E L Stocks, University of Texas at Tyler, Tyler, TX, USA
D A Lishner, University of Wisconsin - Oshkosh, Oshkosh, Wl, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Affective empathy A class of feeling states evoked by the
real, implied, or imagined experiences of another person.
Altruism A motivational state with the ultimate goal of
protecting or promoting the welfare of another individual.
Attitude An individual's relatively stable positive or negative
evaluation of an object (e.g., person, place, or worldview).
Cognitive empathy A class of cognitive-perceptual
processes oriented toward understanding the perspectives,
feelings, or situations of others.
Egoism A motivational state with the ultimate goal of
increasing one's own welfare.
Empathy burnout Distress and psychological distancing
that result from prolonged exposure to the suffering of
others.
External attribution An inference that a behavior was
caused by situational factors rather than personality factors.
Internal attribution An inference that a behavior was
caused by personality factors rather than situational
factors.
The term empathy has been used as a label for a number of
distinct psychological processes and related consequences over
the past century. Most psychologists agree that understanding
empathy (however it may be defined) is cmcial to understand-
ing human social behavior, yet there is much disagreement
and, consequently, much confusion about what it is. Adding
to the confusion is the related concept of sympathy. Although
some psychologists use the terms empathy and sympathy inter-
changeably, others use each to refer to a distinct phenomenon.
Even when each term is used to refer to a distinct phenome-
non, there is little consistency among researchers in their
application. At the most basic level, empathy (or sympathy)
may be said to encompass processes that are responsible for,
or that result from, ( 1 ) understanding what another person is
thinking and feeling; (2) a broad range of perceptual, cog-
nitive, and affective responses to negative and positive events
in the lives of others; or (3) both. The term empathy may also
be used to describe individual differences or a stable personal-
ity characteristic. To make sense of the many uses of the term
empathy and, more importantly, to understand the role that it
plays in human behavior, it is useful to differentiate the various
cognitive and affective processes involved and to differentiate
these processes from their effects.
Empathy as Knowing or Understanding Another’s
Internal States
The term empathy was coined by Titchener in the early 1900s
as a translation of Lipps' concept, einfuhlung. Both terms
referred to the process by which an individual actively projects
or intuits his or her way into another person or object, thereby
experiencing a situation from the other's point of view. This
early conceptualization of empathy emphasized the process
whereby a person deliberately steps outside of himself or her-
self to acquire an accurate understanding of another's psycho-
logical world.
In this respect, empathy is quite similar to the process
that is sometimes labeled role taking or perspective taking.
However, there are a number of variations on this view of
empathy. For example, one may actively imagine what another
is thinking and feeling as a consequence of his or her situation,
based solely on the perception of that person's current values,
goals, resources, and the like. Alternatively, one may instead
imagine how one would feel in the other's situation, given
one's own current values, goals, resources, and the like. The
former cognitive activity has been labeled an imagine-other
perspective, and the latter, an imagine-self perspective. Although
these two approaches to understanding another's internal
states appear to be quite similar at first blush, research suggests
that these two perspective-taking strategies have different emo-
tional, motivational, and behavioral consequences and, as
such, should remain conceptually distinct.
Further complicating the distinction between imagine-self
and imagine-other perspectives are distinctions in the different
aspects of another's perspective that one may seek to under-
stand. Specifically, when attempting to understand another's
perspective one may seek to understand his or her visual per-
spective, cognitive perspective, or affective perspective.
Visual Perspective Taking
Visual perspective taking has been defined as the ability to
understand what another can visually perceive. A prototypical
investigation of this form of perspective taking was provided
by Piaget's Three Mountains Task. In this task, the child is
seated at one of four chairs situated around a table contain-
ing three mountain-shaped objects. The child is asked to ver-
bally describe or to select a pictorial representation of what
another person, seated at a different chair, would see of these
mountains.
Piaget assumed that correctly indicating what the other
person could and could not see suggests an appreciation of
different points of view of a visual stimulus and, by exten-
sion, a rudimentary form of decentering, an awareness that
others may have access to different information about, or
interpretations of, stimuli. Flavell and his colleagues extended
this research by examining the active suppression of one's own
32
Empathy 33
visual or conceptual perspective as a means of facilitating
communication with others. Visual perspective taking is likely
the initial developmental manifestation of the capacity to
empathize with another person, but at least two other mani-
festations of this skill emerge within the first few years of life.
Cognitive Perspective Taking
Cognitive perspective taking has been defined as the ability
to identify or understand what another is thinking in a specific
situation. There is clearly some degree of conceptual overlap
between the visual and cognitive manifestations of perspective
taking in the sense that both involve the recognition of differ-
ing points of view. However, researchers tend to separate these
in terms of both the focus of each skill (viz., visual vs. concep-
tual content of the other's psychological experience of a situa-
tion) and, as noted below, the nature of the research protocols
used to study them.
Building on Piaget's work, Flavell and colleagues' apple-
dog story is the prototype for much of the early research on
cognitive perspective taking. In this task, children are pre-
sented with a cartoon depicting a boy climbing an apple tree
after being chased by a dog. The section of the story in
which the dog chases the boy is removed and the child
is asked to indicate what a naive participant would think is
happening in the story (i.e., what caused the boy to climb
the apple tree). The child's ability to suppress the privileged
knowledge of the dog chase scene and to construct a story
based only on the information that is physically or psycho-
logically available to the other is taken as evidence of cogni-
tive perspective taking.
Researchers working in the Piagetian tradition have also
investigated more complex instances of cognitive perspective
taking. For example, in Flavell and colleagues' nickel-dime
task, one child is asked to trick a second child into choosing
one of the two cups that contains the lesser amount of money
(nickel) so that the first child can keep the cup that contains
the greater amount of money (dime). It is assumed that suc-
cessful completion of the task requires the child to engage in
recursive cognitive perspective taking, whereby the child makes
an inference about the other's cognitions and then adjusts
his or her own in order to manipulate the other's cognitions
(e.g., she will believe B if I do C, so I should do Y in order
to trick her into believing Z). This research has demonstrated
that cognitive perspective taking is a skill that emerges rela-
tively early in life, but one that develops and becomes more
complex and flexible with age and experience.
Affective Perspective Taking
Affective perspective taking has been defined as the ability to
identify and understand what another feels in a particular
situation. Borke's Interpersonal Perception Test is prototypical
of research on one aspect of affective perspective taking. In this
task, the child is asked to indicate a hypothetical person's likely
emotional responses to a series of emotion-relevant situations
(e.g., losing a pet). The child is typically asked to either describe
or to select a pictorial depiction of the target's emotional state.
Research on this aspect of affective perspective taking tends
to follow the same logic as research on cognitive perspective
taking and shares its emphasis on generalized inferential stra-
tegies and cognitive processing.
A second aspect of affective perspective taking was investi-
gated by researchers interested in participants' ability to decode
a target's emotional response either in terms of feeling or
displaying the same emotion that the target is depicting, or
in terms of supplying the appropriate emotion label for the
target's emotion expression. Thus, affective perspective taking
has been conceptualized as both (1) the process of making
inferences about how a generic person would feel in a certain
hypothetical situation, and (2) the cognitive capacity to infer
what a specific person is feeling by decoding (or feeling) the
other's displayed emotion as a consequence of the other's
situation. It has also been treated as a consequence of the
discrepancy between the other's displayed emotion and the
emotion that would be expected, given the other's situation
(e.g., if someone is sad at his or her birthday party).
Empathy as Feeling What Another is Feeling
Empathy has been defined as feeling the same thing that
another is feeling by a number of neuropsychologists, psychol-
ogists, and philosophers. However, it was more common for
research conducted prior to the 1950s to employ the term
sympathy rather than empathy to describe this shared emo-
tional response. Since that time, researchers have coined a
number of phrases to emphasize the unique characteristics of
this process, such as emotional contagion, or parallel emo-
tional responding.
This view of empathy is potentially problematic in that it is
unclear whether an observer must experience an emotional
response that, when compared to the emotion of the other, is
of either exactly the same tone, the same valence, the same
intensity, or some combination of the three to qualify as
empathy. For example, one researcher may define empathy as
experiencing an emotional response that is similar to the emo-
tional response of the other on one of these dimensions
whereas another researcher may require similarity on all three
dimensions. It is also unclear whether the observer must expe-
rience only the emotional response of the other, or if experien-
cing it in addition to other, different emotional responses still
qualifies as empathy.
Empathy as Being Upset by Another’s Situation
Witnessing another individual's suffering can, under certain
circumstances, produce feelings of distress or anxiety. Such
feelings have been labeled empathy by a number of research-
ers, especially in light of the strong statistical association
between measures of empathy and measures of distress in
certain circumstances. However, this label becomes problem-
atic when the other's suffering is either graphic or extreme. In
such cases, traditional measures of empathy and distress do not
appear to assess different aspects of a unified emotional
response. To be sure, distress or anxiety can be caused by
witnessing another suffer, but this emotional response is not
experienced for the other. Rather, it is experienced for oneself.
Given the difference in focus of empathy and distress, recent
34 Empathy
research on the topic maintains a conceptual distinction
between empathy and feelings of distress that may also be
produced by the same situation. This research also suggests
that each of these emotional responses has distinct motiva-
tional and behavioral consequences, further highlighting the
need to conceptually separate empathy from the distress that
may accompany it.
Empathy as Feeling for Another
Perhaps the most common definition of empathy in modern
psychology is feeling a vicarious emotion that is congruent
with, but not necessarily identical to, the welfare of another.
This view of empathy highlights the distinction noted above
regarding distress - empathy is felt for another whereas, in
some circumstances, distress is felt for oneself. Another impor-
tant feature of this view of empathy is that it broadens the scope
of potential empathic emotions. Specifically, empathy is typi-
cally construed as an emotion experienced when witnessing
another suffering or experiencing a negative situation. How-
ever, it is also possible to respond with empathic joy when the
other is experiencing positive states, with empathic anger when
the other has been dealt an injustice, with empathic embarrass-
ment when a cherished other commits a public faux pas, and
so on. Thus, it may be more appropriate to view empathy as
a category of emotional responses or as a way of responding
emotionally rather than as a unitary, prototypical emotion.
Empathy as Matching Another’s Overt Responses
Another variation on the concept of empathy views it as the
process by which one matches another's overt responses to his
or her current situation. Researchers have focused on various
aspects of the other's responses that are to be matched, includ-
ing his or her posture, emotional expressions, physiological
state, and physical activities. This view of empathy has also
been labeled motor mimicry or imitation to reflect its primary
characteristic, some form of behavioral matching.
More recently, neuropsychological research suggests that
matching in empathy goes well beyond motor mimicry or
imitation. According to the Perception-Action Model of empa-
thy, matching at the level of neural representations is central to
the experience of empathy. Specifically, due to the automatic
link between perception and action (viz., involving the dorso-
lateral prefrontal cortex and left inferior parietal region), sim-
ply perceiving another's physical activities results in a pattern
of neural activation similar to what would occur if the observer
were engaging in the same physical activities himself or herself.
It is important to note, however, that the neural matching is far
from complete, and that empathy involves a decoupling mech-
anism (viz., involving the insula and right inferior parietal
cortex) that differentiates one's own experiences and physical
activities from those of the observed other.
It is important to note that response matching does not
necessarily lead to an accurate understanding of the other. In
many cases, matching is a goal-directed act that serves to com-
municate support for the other rather than an automatic and
unavoidable imitative act. And it is, of course, possible that
another's overt physical activities or emotional expressions
are intentionally inaccurate, so mimicking those actions or
expressions would lead to an inaccurate understanding of
the other. Above all, one need not match or imitate another's
responses in order to understand them. Humans are quite
adept at communicating what they are thinking and feeling,
and one often needs to simply listen carefully to the other in a
nonjudgmental manner to gain accurate knowledge of his or
her internal states.
Summary
The various processes outlined above bear some conceptual
overlap, but it is important to note the differences between
them. The first (empathy as perspective taking) is primarily a
cognitive account of empathy. In contrast, the second, third,
and fourth accounts (empathy as feeling what another is feel-
ing, empathy as feeling upset by another's situation, and empa-
thy as feeling for another) are affective accounts of empathy,
and the fifth account (empathy as matching another's overt
responses) is a behavioral account of empathy.
Although the various processes labeled as empathy share
similarities, it is important to note that the antecedents and
consequences of these different views of empathy are not iden-
tical. Interestingly, important distinctions among these pro-
cesses often result from their differential links to one another.
For example, research suggests that adopting an imagine-self
perspective toward another in need results in feeling a similar
emotion to what one expects the other to feel, whereas adopt-
ing an imagine-other perspective toward a person in need has
been shown to result in feeling for the other.
These two forms of affective empathy in turn can result in
different motivational outcomes. Feeling a similar negative
emotion as the other has been shown to evoke a selfish motive
to reduce the emotion, resulting in attempts to escape the emo-
tion by either escaping the situation or avoiding the other in
need, both of which are involved in eliciting the unpleasant
negative emotion. In contrast, feeling for the other evokes
an altruistic motive to help the other, resulting in offering
direct help to the other or attempting to get someone else to
help him or her. The link between these two perspectives and
their distinct emotional, motivational, and behavior conse-
quences has been demonstrated numerous times in scientific
literature, particularly by social psychologists. The link between
being upset by another's situation and specific motivational
and behavioral consequences is, however, less clear and is in
need of further study.
As a further layer of complexity, research suggests that
empathy as matching another's overt responses sometimes
leads to cognitive empathy, affective empathy, or both. But,
cognitive and affective empathy can also lead to a matching of
overt responses. Thus, it is not surprising that the distinctions
in various forms of empathy are often lost in the literature,
given their interrelated and reciprocal nature.
Empathy as a Personality Characteristic
The research described above has sought to understand the
cognitive and affective processes that have been referred to
as empathy. However, other scholars have sought to describe
Empathy 35
individual differences in either the capacity for experiencing
empathy or in the tendency to use this capacity. That is, instead
of thinking of empathy as a process, some scholars think of
empathy as a distinct and stable aspect of one's personality that
transcends situations and potential eliciting stimuli.
To this end, researchers have worked to develop valid and
reliable measures of individual differences in the capacity or
tendency to experience empathy for others, be it understanding
what another is thinking and feeling, feeling what the other is
feeling, or feeling for the other. Most of these measures are self-
report in nature and, consequently, require that respondents
have accurate insight into their empathic capacities or tenden-
cies, have the means for communicating these capacities or
tendencies, and are not motivated to present a distorted view
of these capacities or tendencies. It is important to note that
these preconditions are unlikely to be met at all times in all
studies. As such, when considering various individual differ-
ence measures of empathy, one should keep in mind these
potential limitations. Examples of individual difference mea-
sures of empathy include the Interpersonal Reactivity Index
(IRI), developed by Davis, the Questionnaire Measure of
Emotional Empathy (QMEE), developed by Mehrabian and
Epstein, and the Emotional Contagion Scale (ECS) developed
by Doherty.
The Interpersonal Reactivity Index
The structure of the IRI is based on the assumption that
empathy consists of four related, but conceptually distinct,
constructs. Specifically, the IRI consists of four subscales:
perspective taking, empathic concern, personal distress, and
fantasy. Taking these in turn, the perspective taking subscale
measures an individual's tendency to spontaneously adopt
other people's points of view. The empathic concern subscale
measures an individual's tendency to feel sympathy and com-
passion for those in need. The personal distress subscale mea-
sures an individual's tendency to experience discomfort when
in the presence of those in need. The fantasy subscale measures
an individual's tendency to imagine himself or herself in fic-
tional situations (e.g., in the scene of a movie or as a character
in a book). Taken together, the IRI subscales measure empathy
as various types of both cognitive processes (perspective taking
and fantasy subscales) and affective processes (empathic con-
cern and personal distress subscales).
Questionnaire Measure of Emotional Empathy
The QMEE is designed to measure a person's tendency to
become emotionally aroused in response to the emotional
behavior or situations of others. It includes items that measure
feeling the same emotion as another, feeling upset about the
situation of another, and feeling for another. Consequently,
the QMEE is probably best conceptualized as a general mea-
sure of empathy as an affective process.
Emotional Contagion Scale
The ECS is a measure of one's tendency to feel the same emo-
tions as others. It consists of five subscales that measure ten-
dencies to 'catch' the experiences of happiness, sadness, anger,
fear, and love. Alternatively one can combine scores on all five
subscales to create a general measure of the tendency to feel the
emotions of others.
Sex Differences in Empathy
A commonly held belief is that women are more empathic
than men, particularly in response to the needs of others.
Indeed, an analysis of existing empathy research by Eisenberg
and her colleagues suggests that women report more emotional
reactivity to the plight of others in need than do men. How-
ever, a closer look across studies reveals there is little evidence
of a gender difference when physiological measures of emo-
tional intensity are used to measure empathy. This suggests
that the differences found in research that employs self-report
methodologies may be due to self-presentation concerns and
gender role-based expectancies rather than to biological gender
differences. Thus, (1) women may overreport their genuine
feelings of empathy because of what they think is expected of
them, (2) men may underreport their genuine feelings of
empathy because of what they think is expected of them, or
(3) both.
Research does suggest that women are slightly better than
men at decoding visual and auditory cues from others (i.e.,
adopting the perspective of another individual to understand
what he or she is thinking and feeling) when self-report mea-
sures of empathic accuracy are used; however, no gender dif-
ferences are found when this aspect of empathy is measured
using objective measures. Ironically, this difference on self-
report measures of perspective taking may be due to the same
self-presentation concerns and gender-role expectations as the
differences found using self-report measures of affective empa-
thy (which do not appear to reflect actual differences in the
intensity of emotional response to others).
Summary
Much like the theoretical literature, studies on individual dif-
ferences in the capacity or tendency to be empathic have trea-
ted the empathy concept in a variety of different ways. There
are also gender-based differences in the tendency to be
empathic when empathy is conceptualized as the ability to
correctly infer what another person is thinking or feeling
(even though objective measures do not detect this difference).
Yet, there is little evidence of gender differences when physio-
logical measures are used to assess affective empathy (even
though self-report methodologies often mistakenly suggest
that gender differences in affective empathy do exist).
Intrapersonal and Interpersonal Consequences
of Empathy
Empathy, Prosocial Behavior, and Antisocial Behavior
One of the most widely studied consequences of empathy is its
effect on prosocial behavior. The relation between empathy for
a victim and an increased tendency to help him or her has been
well known for several decades. However, disagreement over
36 Empathy
the reasons for this relation has garnered much research atten-
tion of late.
According to the Empathy-Altruism Hypothesis, empathy
evokes an altruistic motive with the ultimate goal of protecting
or promoting the welfare of the person for whom empathy is
felt. Note that, as they are used here, the term empathy refers to
another-oriented emotional response to the victim's welfare
(feeling for the other), and the term altruism refers to a moti-
vation rather than to any specific form of helping behavior.
The claim that humans are capable of genuine altruism is
somewhat controversial and, as a consequence, a number of
egoistic explanations for the empathy-helping association
have been proposed. The term egoism refers to a motivational
state with the ultimate goal of protecting or promoting one's
own welfare. Among the more popular egoistic explanations for
empathy-based helping are (1) the Aversive-Arousal Reduction
Hypothesis, which claims that empathy is aversive and that
empathically aroused individuals help in order to rid themselves
of the unpleasant feeling; (2) punishment-avoidance explana-
tions, which claim that empathically aroused individuals help
because they anticipate negative social evaluations or shame
and guilt for failing to help; and (3) reward-seeking explana-
tions, which claim that empathically aroused individuals help
because they anticipate praise or positive mood for helping
those for whom empathy is felt. Although each of these expla-
nations has some degree of empirical support, the Empathy-
Altruism Hypothesis is considered by many researchers to have
the weight of evidence on its side.
In contrast to the large body of work on the causal relation
between empathy and prosocial behavior, there is relatively
little research on the causal relation between empathy and
antisocial behavior. Of the research that has been done,
most has focused on the behaviors of deviant populations or
young children, has assessed empathy as a general personality
disposition rather than as a cognitive or emotional process,
or used methodological designs that make it difficult to
determine the causal direction of the relation (i.e., does empa-
thy decrease antisocial behavior or does antisocial behavior
decrease empathy?). As such, evidence for a negative relation
between empathy and aggression has been reported, but it is
largely inconsistent across populations, research paradigms,
techniques for conceptualizing and measuring empathy, and
for different forms of antisocial behavior. Consequently, it is
difficult to generalize beyond the samples and methods
employed in existing research.
Empathy, Accuracy, and Social Relationships
Empathy has long been acknowledged as an important com-
ponent of social relationships. When defined as a cognitive
process, empathy has been linked to conflict avoidance or
management, improved communication among relationship
partners, increased relationship satisfaction, and a tendency to
be considerate of the partner's wants and needs. However,
research suggests a number of potential barriers to accurate
perception of what another is thinking and feeling, which can
mitigate the positive effects of empathy. For example, it is
common for individuals to use themselves as a basis for asses-
sing how and why another individual will react to a given
situation. Making judgments from an egocentric perspective
(viz., that self and other are similar) can be useful in many
cases, but accuracy is impaired when the empathizer and the
other are dissimilar on situation-relevant domains. Accuracy is
also impaired when the empathizer is currently experiencing
strong emotion or is in a state of need, both of which tend to be
unconsciously incorporated into his or her judgment of the
other's thoughts and feelings.
The role that empathy plays within the context of close
relationships is even more complex. Research suggests an inter-
esting paradox: empathic accuracy increases relationship satis-
faction in certain situations, but decreases relationship
satisfaction in others. Specifically, in nonthreatening contexts
(e.g., discussing mundane events or planning a night out), a
moderate-to-high level of motivation to accurately understand
what a partner is thinking and feeling increases relationship
satisfaction and stability. However, in relationship-threatening
contexts (e.g., potential infidelity or disappointment with the
partner's performance), individuals tend to be motivated
toward empathic inaccuracy - that is, to intentionally distort
his or her perception of what the partner is thinking and
feeling. In such cases, empathic inaccuracy actually increases
relationship satisfaction and stability. In connast, when the
motive to be empathically accurate is strong in relationship-
threatening contexts, the relationship is likely to suffer. Thus,
empathy is important in social relationships in general, but
inaccuracy is a common (and occasionally useful) exception,
especially in close relationships when dealing with potentially
threatening contexts.
Empathy and Attributions
Attributions are beliefs about the cause of an individual's
behavior. According to past research, there is a pronounced
actor-observer difference when making attributions. Specifi-
cally, individuals tend to explain their own behavior as a
consequence of the situation (an external attribution) whereas
they tend to explain the behavior of others as a consequence of
personality traits or dispositions (an internal attribution),
especially when the behavior is negative or otherwise results
in failure. Although the actor-observer difference is found
primarily when the actor is highly idiosyncratic or when the
actor and observer are in a close relationship, the effect can be
reversed by simply instructing the observer to empathize with
the actor. As a consequence of empathizing with the actor, the
observer makes external attributions for the actor's behavior,
just as he or she would make external attributions for himself
or herself. Thus, empathy promotes the evaluation of others in
a relatively self-like (i.e., generous) manner.
Empathy and Attitudes
Empathy has also been shown to increase valuing of, and
attitudes toward, individuals for whom empathy is felt. This
is, at least in part, due to the reversal of the attributional biases
noted above, especially in situations that involve some form of
victimization, in that viewing another's suffering as a result
of situational factors rather than as a result of dispositional
Empathy 37
factors reduces the likelihood that derogation and blaming the
victim will occur.
The effects of empathy on valuing have important conse-
quences for attitudes toward members of stigmatized groups
(e.g., minorities or people with HIV/AIDS) and for attitudes
toward social causes (e.g., animal rights or the environment).
Research suggests the following three-step model of how
empathy can be used strategically to improve attitudes toward
such groups: (1) actively trying to understand what a stigma-
tized person is thinking and feeling will increase empathic
feelings for him or her, (2) feeling empathy for him or her
will increase valuing of his or her welfare, and (3) valuing this
person will generalize to valuing the stigmatized group as a
whole. Initial tests of this model demonstrated the success of
this strategy by improving attitudes toward a broad range of
stigmatized individuals and groups (e.g., people with AIDS,
homeless individuals, and convicted murderers). Subsequent
research has demonstrated the power of empathy to reduce
prejudice and discrimination against members of racial or
ethnic minorities, and homosexuals, and to improve attitudes
toward protecting the natural environment and its nonhuman
inhabitants.
Implications and Future Directions
The term empathy has been used to describe a broad range of
cognitive-perceptual and emotional processes, all of which
play an important role in human behavior. As noted above,
empathy is potentially a powerful force not only in personal
relationships, but also in an individual's interaction with social
groups and with the world more generally. However, much is
left to be discovered about the processes and consequences of
this class of phenomena.
Perhaps the most pressing need at present is the theoretical
integration of the cognitive and emotional processes involved
in empathy. There is much disagreement and little consistency
among researchers on exactly what empathy is, which has
almost certainly slowed progress toward understanding its
causes and consequences. Second, a number of important issues
pertaining to empathy have received little or no empirical atten-
tion. For example, research on the neurological bases of empa-
thy has emerged only within the last few years and much is left
to be learnt about which brain processes are responsible for
the various empathy phenomena and how organic brain
damage can affect an individual's capacity to understand and
respond to the thoughts and feelings of others. Similarly, little
is known about the genetic basis or evolutionary history of
empathy. Given its importance in human social functioning,
it is remarkable that this topic has received so little empirical
attention.
Also, much research has been conducted to understand
the relation between empathy and antisocial behavior, yet the
results are largely inconsistent from study to study. Future
research may benefit from adopting a broader purview of the
topic and by employing ( 1 ) measures of empathy as a process
rather than as a general personality disposition; (2) a
representative range of antisocial behaviors rather than focus-
ing exclusively on physical or verbal retaliation; (3) diverse
populations, including normal, physically healthy adults; and
(4) experimental methodology so that the causal direction of
an association can be determined. A broader purview would
also be useful when investigating other, socially relevant con-
sequences of empathy, such as volunteerism and a related
phenomenon, empathy burnout. Empathy burnout is an
involuntary process by which empathy is transformed into
feelings of distress after prolonged exposure to the suffering
of others that causes caregivers (e.g., spouses, nurses, volun-
teers, social workers, etc.) to become callous and, potentially,
disengaged from service to the victims. The majority of research
on empathy and prosocial behavior has focused on single-
instance forms of helping and, as a consequence, little is
known about the effects of empathy in long-term care situa-
tions. Future research on these and related issues could
advance both theoretical understanding and real-world appli-
cation of this important and complex class of phenomena.
See also: Altruism and Helping Behavior; Attitude Change; Moral
Development.
Further Reading
Batson CD (1991) The Altruism Question: Toward a Social-Psychological Answer.
Hillsdale, NJ: Erlbaum Associates.
Batson CD, Ahmad N, and Stocks EL (2004) Social benefits and liabilities of empathy-
induced altruism. In: Miller AG (ed.) The Social Psychology of Good and Evil:
Understanding Our Capacity for Kindness and Cruelty, pp. 359-385. New York:
Guilford.
Buck R and Ginsburg B (1997) Communicative genes and the evolution of empathy.
In: Ickes W (ed.) Empathic Accuracy, pp. 17-45. New York: Guilford.
Cialdini RB, Schaller M, Houlihan D, Arps K, Fultz J, and Beaman AL (1987)
Empathy-based helping: Is it selflessly or selfishly motivated? Journal of Personality
and Social Psychology 52: 749-758.
Davis MH (1994 ) Empathy: A Social Psychological Approach. Madison, Wl: Brown &
Benchmark.
Decety J and Jackson PL (2004) The functional architecture of human empathy.
Behavioral and Cognitive Neuroscience Reviews 3: 71-100.
Eisenberg N and Miller PA (1987) Empathy and prosocial behavior. Psychological
Bulletin 101: 91-119.
Flavell JH, Botkin PT, Fry CL, Wright JW, and Jarvis PE (1968) The Development of
Role-Taking and Communication Skills in Children. New York: Wiley.
Hoffman ML (2000) Empathy and Moral Development: Implications for Caring and
Justice. Cambridge: Cambridge University Press.
Ickes W, Simpson JA, and Orina M (2005) Empathic accuracy in close relationships.
In: Malle M and Hodges SD (eds.) Other Minds: How Humans Bridge the Divide
Between Self and Others, pp. 310-322. New York, NY: Guilford.
Krebs DL (1975) Empathy and altruism. Journal of Personality and Social Psychology
32: 1134-1146.
Omoto AM and Snyder M (1995) Sustained helping without obligation: Motivation,
longevity of service, and perceived attitude change among AIDS volunteers. Journal
of Personality and Social Psychology 68: 671-686.
Schroeder DA, Penner LA, Dovidio JF, and Piliavin JA (1995) The Psychology of
Helping and Altruism: Problems and Puzzles. New York: McGraw-Hill.
Sober E and Wilson DS (1998) Unto Others: The Evolution and Psychology of Unselfish
Behavior. Cambridge, MA: Harvard University Press.
Stotland E (1969) Exploratory investigations of empathy. In: Berkowitz L (ed.)
Advances in Experimental Social Psychology, vol. 4, pp. 274-314. New York:
Academic Press.
Empirical Challenges to Conventional Mind-Brain Theory
E F Kelly and E W Kelly, University of Virginia, Charlottesville, VA, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Automatism Motor activity not initiated consciously.
Eidetic imagery Mental images that are almost
photographically exact.
Mediumship The ability of some persons, usually in a
trance or other altered state, to produce information relating
to a deceased person, which they have not learned normally.
Psi A theory-neutral term designating phenomena that seem
not to derive from ordinary sensorimotor processses,
including phenomena such as extrasensory perception
(telepathy or clairvoyance), psychokinesis ('mind over
matter'), or precognition.
Introduction
Nearly all contemporary psychologists, neuroscientists, and
philosophers subscribe - explicitly or implicitly - to some
version of physicalism. Physicalist conceptions of the human
mind and personality, contrary to traditional and everyday
notions, run along roughly the following lines: We human
beings are nothing but extremely complicated biological
machines. Everything we are and do is in principle causally
explainable from the bottom up in terms of our biology, chem-
istry, and physics - ultimately, that is, in terms of local contact
interactions among bits of matter moving in accordance with
mechanical laws under the influence of fields of force. Some of
what we know and the substrate of our general capacities to
learn more are built-in genetically as complex resultants of
biological evolution. Everything else comes to us directly or
indirectly by way of our sensory systems, through energetic
exchanges with the environment of types already largely under-
stood. Mind and consciousness are generated by - or in some
mysterious way identical with - neurophysiological events and
processes in the brain. Mental causation, free will, and the 'self
do not really exist; they are mere illusions, ineffectual by-pro-
ducts of the grinding of our neural machinery. And since mind
and personality are entirely products of our bodily machinery,
they are necessarily extinguished, totally and finally, by the
demise and dissolution of that body.
Views of this sort unquestionably hold sway over the vast
majority of contemporary scientists and they have also perco-
lated widely through the public at large. We believe, however,
that they are at best seriously incomplete, and at certain critical
points, demonstrably false, empirically. In this article, we will
briefly catalogue a variety of interrelated empirical phenomena
that appear difficult or impossible to explain in conventional
physicalist terms. We emphasize from the outset that these
phenomena must be considered collectively, not piecemeal;
they not only challenge the conventional physicalist picture
individually, but converge in pointing to the need for a radi-
cally novel way of understanding the intimate relationship of
mind and brain. We also emphasize that we are presenting here
only a skeletal outline of the kinds of phenomena to which we
wish to direct readers' attention. More detailed treatments of
relevant empirical evidence and the issues raised can be found
in works given in the Further Reading.
Extreme Psychophysiological Influence
We begin with a variety of phenomena especially suggestive of
the involvement of direct mental agency in the production of
physiological effects not fully explainable in terms of physio-
logical mechanisms alone. The following examples will serve
to capture their flavor.
Placebo effects and related kinds of psychosomatic phe-
nomena have long been informally recognized and are
now widely accepted, but they were accepted by modem
biomedical science only grudgingly, as new mechanisms
of brain-body interaction that seemed potentially capable
of explaining them came to light. In particular, psychoneu-
roimmunology has demonstrated the existence, previously
unknown, of interactions between the central nervous sys-
tem and the immune system. Nevertheless, the adequacy
of such explanations to account even for placebo effects
remains in question, and there are many types of kindred
phenomena that pose progressively greater challenges to
explanation in such terms.
For example, Sigmund Freud and F. W. H. Myers were
impressed by hysterical 'glove anesthesias,' in which a patient
loses sensation from the skin of a hand in the absence of
organic lesion, in a way that typically corresponds only to the
patient's idea, in complete disregard of the underlying anato-
mical organization. Related phenomena have often been
reported in the context of hypnosis. For example, highly sug-
gestible persons who can vividly imagine undergoing an in-
jurious circumstance such as receiving a burn to the skin
sometimes suffer effects closely analogous to those that the
physical injury itself would produce, such as a blister. More
rarely, the correspondence between the hypnotic blister and its
imagined source extends even to minute details of geometric
shape, details too specific to account for in terms of known
mechanisms of brain-body interaction. A closely related and
well-documented phenomenon is that of 'stigmata,' in which
fervently devout or pious believers in Christ develop wounds
analogous to those inflicted during the crucifixion. The injuries
are again localized and specific in form, vary in locus and
character in accordance with their subjects' differing concep-
tions of Christ's own injuries, and appear and disappear, often
suddenly and regularly, also in accordance with the subject's
expectations. Similarly, dramatic phenomena have occasionally
38
Empirical Challenges to Conventional Mind-Brain Theory 39
been documented in psychiatric patients in connection with
their recall of prior physical trauma.
The conventional hope, of course, is that even the most
extreme of the phenomena just mentioned might ultimately
be explained in terms of brain processes. Continuing allegiance
to this hope, despite the indicated explanatory difficulties, is
undoubtedly encouraged by the fact that the phenomena
described so far all involve effects of a person's mental states
on that person's own body. Still more drastic explanatory
challenges are posed, however, by cases in which one person's
mental state seems to have directly influenced another person's
body. Such phenomena include 'maternal impressions' (birth-
marks or birth defects on a newborn that correspond to an
unusual and intense experience of the mother during the preg-
nancy), distant healing (including experimental studies of
effects of prayer on healing), experimental studies of distant
mental influence on living systems, and cases in which a child
who claims to have memories of the life of a deceased person
also displays extremely unusual birthmarks or birth defects
corresponding closely with marks (usually fatal wounds) on
the body of that person. In addition, there has been a consid-
erable accumulation of spontaneous cases and experimental
evidence demonstrating the reality of psychokinesis (PK),
which by definition involves direct mental influence on the
physical environment (see section 'Psi Phenomena').
Extremes of Informational Capacity and Precision
A number of well-documented psychological phenomena
involve levels of detail and precision difficult to account for
in terms of a brain operating in statistical fashion with neural
components of low intrinsic precision and reliability. Here are
some examples.
The first involves a case of 'automatic writing' observed by
William James. The subject wrote with his extended right arm
on large sheets of paper, his face meanwhile buried in the
crook of his left elbow. For him to see what he was doing was
'a physical impossibility.' Nevertheless, James continues: "Two
or three times in my presence on one evening, after covering a
sheet with writing (the pencil never being raised, so that the
words ran into each other), he returned to the top of the sheet
and proceeded downwards, dotting each i and crossing each
£ with absolute precision and great rapidity."
This episode illusuates two features that have often ap-
peared together in the large but neglected scientific literature
dealing with automatic writing: The subject is in an altered
state of consciousness, and the motor performance, itself
remarkable, is apparently guided by an extremely detailed
memory record, an essentially photographic representation of
the uncompleted page.
The latter property relates to the phenomenon of eidetic
imagery, the second example, the most dramatic demonstra-
tion of which has been provided by Charles Stromeyer using
Julesz stereograms. These are essentially pairs of computer-
generated pictures, each of which by itself looks like a matrix
of randomly placed dots, but is constructed in such a way that
when viewed simultaneously (by presentation to the two eyes
separately) a visual form emerges in depth. Stromeyer pre-
sented pictures of this type to the eyes of his single subject, a
gifted female eidetiker, at different times, ultimately as much as
3 days apart. Under these conditions, the subject could exttact
the hidden form only if she could fuse current input to one eye
with an extremely detailed memory image of previous input to
the other eye. Remarkably, she was able to succeed under a
wide variety of increasingly demanding conditions. The origi-
nal stereograms, for example, were 100 x 100 arrays, but she
ultimately succeeded under double-blind conditions with
arrays as large as 1000 x 1000, or a million 'bits,' viewed up
to 4 h apart.
These results were understandably shocking to many psy-
chologists, who sought to escape their force by pointing to the
dependence on a single subject and the absence of replications.
At least one successful replication has subsequently occurred,
however. Moreover, the literature already contains many addi-
tional examples of prodigious memory. Stromeyer mentions
Luria's famous mnemonist and the case of the 'Shass Poliaks,'
who memorized all 12 volumes of the Babylonian Talmud,
and Oliver Sacks has reported a similar case of a person who
among other things knew by heart all nine volumes and 6000
pages of Grove's Dictionary of Music and Musicians. Other exam-
ples could easily be cited. Prodigious memory of this sort is a
real psychological phenomenon.
Third in this group is the family of 'calculating prodigies.'
Of special interest is the 'savant syndrome,' often associated
with autistic disorders, in which islands of spectacular ability
appear in the midst of generalized mental disability. The abil-
ities are of many types, but almost invariably involve prodi-
gious memory. The depth of the problems they pose for brain
theory is exemplified by the case of 'The Twins,' also described
by Sacks. These profoundly impaired individuals, unable to
perform even simple additions and subtractions with any accu-
racy, nonetheless proved able to generate and test prime num-
bers in their heads. Sacks was able to verify the primacy up to
10 digits, but only by means of published tables, while the
twins themselves went on exchanging numbers of steadily
greater length, eventually reaching 20 digits. Sacks makes the
inttiguing suggestion that they may not literally be calculating
these enormous numbers, but discovering them by navigating
through some vast inner iconic landscape in which the rele-
vant numerical relations are somehow represented pictorially.
The twins themselves of course cannot say how they do it.
Phenomena of these sorts look hard to explain in terms of
brain processes. The most serious attempt to do so that is
known to us is in fact devoid of specific neural mechanisms.
Its cenUal argument is rather that early-stage brain processes
like those subserving visual perception, for example, must also
be savant-like in terms of their speed, precision, and infor-
mational capacity; what is unusual about savants, therefore,
may consist merely in their access to these mechanisms. This
explanation of course presupposes a positive answer to the
fundamental question at issue, whether the brain alone can
accomplish any of these things including perceptual synthesis
itself (see section 'The Unity of Conscious Experience').
As proved long ago by mathematician John von Neumann,
the only practical way to get increased arithmetical precision
out of individually unreliable neurons is to use more of them.
This biocomputational perspective clearly implies that calcu-
lating prodigies must use large portions of their brains in very
abnormal ways to achieve the observed effects. The cognitive
40 Empirical Challenges to Conventional Mind-Brain Theory
deficits that often accompany savant-type skills could conceiv-
ably reflect such substitutions, but we must remember that
comparable skills sometimes also occur in geniuses such as
the mathematicians Gauss and Ampere.
Memory
The previous section focused on phenomena such as high-
precision calculations and prodigious memory that appear
incompatible with the physical properties of the brain consid-
ered as a kind of computing device. Problems also arise, how-
ever, in regard to memory in its more familiar and everyday
forms. Here we briefly sketch some relevant issues.
Memory is central to all human cognitive and perceptual
functions, yet we remain largely ignorant of where and in what
forms our past experience is stored and by what means it is
brought to bear upon the present. Generations of psychologists
and neurobiologists have taken it as axiomatic that all stored
memories must exist in the form of 'traces,' physical changes
produced in the brain by experience, but there has been little
real progress toward scientific consensus on the details of these
mechanisms despite many decades of intensive research.
Significant progress has recently been made, to be sure, in
regard to 'learning' and 'memory' in simple creatures such as
the sea slug (Aplysia), and more generally in regard to what
might be called 'habit memory,' the automatic adjustments of
organisms to their physical environments. But these discov-
eries fall far short of providing satisfactory explanations of
the most central and important characteristics of the human
memory system, including in particular our supplies of general
knowledge (semantic memory) and our ability to recall volun-
tarily and explicitly our own past experience (autobiographical
or episodic memory). Furthermore, recent functional neuro-
imaging studies, although generating vast amounts of data,
have yielded little, if any, progress toward a comprehensive
and coherent account of memory based on trace theory.
Meanwhile, deep conceptual problems have been identi-
fied in trace theory itself. For example, autobiographical
memory clearly involves something more than mere revival
of traces of past experiences, something that allows us to
interpret what is experienced now as a representation of our
own past rather than a contemporary perception, dream, or
hallucination. That is, traces as such only provide memory
aids rather than memories per se, and it has proved extremely
difficult to specify in conventional physicalist terms what that
extra something is, without falling into regressive forms of
explanation that presuppose and hence cannot explain the
phenomenon of memory itself. Similarly, the content of a
concept or semantic memory typically transcends any finite
set of experienced circumstances that can plausibly be imag-
ined as having deposited corresponding 'traces' in a form
capable of explaining its effective deployment in an unlimited
variety of novel contexts.
These conceptual problems regarding trace theories of
memory have deep connections with issues discussed in sec-
tion 'The Heart of the Mind', and similar issues arise in relation
to allied components of current cognitive theory such as 'infor-
mation' and 'representation.' See also section 'Psi Phenomena'
for an additional empirical problem.
Psychological Automatisms and Secondary
Centers of Consciousness
Phenomena catalogued under this heading involve what looks
like multiple concurrent engagement, in potentially incompat-
ible ways, of major cognitive skills (linguistic skills, for exam-
ple) and the corresponding brain systems.
Current cognitive neuroscience pictures the mind or 'cog-
nitive system' as a hierarchically ordered network of subpro-
cessors or 'modules,' each specialized for some particular task
and corresponding (it is hoped) to some particular brain
region or regions. Leaving aside major issues regarding the
details of its specification, this picture seems broadly consis-
tent with the overall manner in which our minds normally
seem to operate. That is, our basic way of consciously doing
things is essentially one at a time in serial fashion. Although
psychologists recognize that with suitable training people can
do more things in parallel than they customarily suppose,
this generalization applies mainly to relatively divergent
things, and conspicuously fails as the simultaneous tasks
become more complex and more similar.
Nevertheless, a large body of credible evidence, some dating
back to the late nineteenth century, demonstrates that addi-
tional 'cognitive systems,' dissociated psychological entities
indistinguishable from full-fledged conscious minds or per-
sonalities as we normally understand these terms, can some-
times occupy the same organism simultaneously, carrying on
their varied existences as it were in parallel, and largely outside
the awareness of the primary, everyday consciousness. In
essence, the structure that cognitive psychology conventionally
pictures as unitary, as instantiated within and identified with a
particular organization of brain systems, can be functionally
divided - divided, moreover, not 'side to side,' leading to iso-
lation of the normal cognitive capacities from each other, but
'top to bottom,' leading to the appearance and concurrent -
not alternating - operation of what seem to be two or more
complete cognitive systems, each of which includes all of the
relevant capacities. Emergent 'multiple' or 'alter' personalities
can also differ widely, not only in demeanor, interests, and
knowledge but even in regard to nonvoluntary physiological
characteristics such as visual defects and susceptibilities to
allergies. Even worse, it sometimes happens that one of these
personalities appears to have direct access to the conscious
mental activity of one or more others, but not vice versa.
Two brief examples, drawn from an enormous literature,
may help convey a more concrete sense of the character of
these phenomena.
The first comes from a report by Oxford philosopher
F. C. S. Schiller on automatic writing produced by his brother.
As is characteristic of this genre of automatisms, the writer was
typically unaware of the content of his writing, which went on
continuously while he was fully and consciously engaged
in some other activity such as reading a book or telling a
story. Of particular relevance here, however, were occasions
on which he wrote simultaneously with both hands and on
completely different subjects, one or the other of these streams
of writing also sometimes taking mirror-image form.
The second example is the case of Anna Winsor, described
by William James in his report on automatic writing. The case
Empirical Challenges to Conventional Mind-Brain Theory 41
was protracted and bizarre, but only superficially resembles the
neurological 'alien hand' (Dr. Strangelove) syndrome. Its cen-
tral feature is that the patient, Anna, at a certain point lost
voluntary control of her right arm, which was taken over by
a distinctive secondary personality. This personality, whom
Anna herself named 'Old Stump,' was benign, often protecting
Anna from her pronounced tendencies toward self-injury. As in
the case of Schiller's brother, Stump typically wrote or drew
while Anna was occupied with other matters. But Stump also
continued writing and drawing even when Anna was asleep,
and sometimes in total darkness. This secondary personality
also remained calm and rational during periods when Anna
was feverish and delusional, and it manifested knowledge and
skills which Anna herself did not possess.
Psi Phenomena
Here we refer to experimental and field observations systemat-
ically adduced in the course of over a century of effort by
workers in 'psychical research' and its modern descendent,
'parapyschology.' The phenomena in question involve, by def-
inition, correlations occurring across physical barriers that
should be sufficient, on presently accepted physicalist princi-
ples, to prevent their formation. This occurs, for example,
when person A spontaneously experiences an apparition of
his friend B, as B unknown to A lies dying from a fatal accident.
Over a thousand detailed cases of this sort - carefully docu-
mented experiences that are not dismissible en masse as mere
'anecdotes' - have been published in the peer-reviewed litera-
ture. It also occurs when an experimental subject consistently
succeeds in identifying randomly selected forced-choice targets
displayed in a remote location. It is not difficult to set up
controlled experiments of this sort and to evaluate their out-
comes using rigorous statistical procedures. A large amount of
careful experimental work has been carried out along these
lines, with results more than sufficient, in our opinion, to
demonstrate beyond reasonable doubt to open-minded per-
sons that the sheer existence of the basic input-output
phenomena - 'extrasensory perception' (ESP) and 'psychoki-
nesis' (PK) in the popular vocabulary, or in more theory-
neutral terminology, 'psi' - is a fact of nature with which we
must somehow come to scientific terms.
Psi phenomena in general are important because they
provide examples of human behavioral capacities that
appear impossible to account for in terms of presently recog-
nized computational, biological, or classical physics princi-
ples. Even more important for our purposes, however, is a
further body of evidence suggestive of postmortem survival,
the persistence of elements of mind and personality follow-
ing bodily death. It is simply not true, as most scientists
presume, that we possess no such evidence. We in fact pos-
sess a lot of such evidence, much of it of very high quality,
deriving, for example, from studies of veridical apparitions,
trance mediumship, and 'cases of the reincarnation type,' in
which young children spontaneously report verifiable events
from the lives of distant and ordinary persons now deceased.
Ironically, the primary threat to a survivalist interpretation
of this accumulated evidence arises not from considerations
of evidential quality, but from the difficulty of excluding
alternative explanations based upon psi interactions involv-
ing only living persons.
Quite apart from any personal or theological interests read-
ers may bring to this subject, it should be evident that post-
mortem survival, if it occurs, demonstrates dramatically the
limitations of present-day reductive physicalism. If it is the
case, for example, as much evidence indicates, that autobio-
graphical, semantic, and procedural (skill) memories can sur-
vive bodily death, then memory in living persons must
presumably exist at least in part outside the brain and body
as conventionally understood.
Either horn of this interpretive dilemma - postmortem
survival or psi among the living - is lethal to current physi-
calist orthodoxy, which undoubtedly explains the widespread
scientific resistance to both. But as we are arguing here, and
have argued in much more detail elsewhere, these phenom-
ena cannot be isolated and quarantined, because similarly
difficult explanatory challenges are posed by many other
well-evidenced psychological phenomena. Evidence for the
occurrence of psi phenomena in general and postmortem
survival in particular must, we believe, play an important role
in the formulation of an empirically adequate mind-brain
theory, and our efforts here will be amply rewarded if they
lead scientifically minded readers to examine these subjects
more seriously than they otherwise might.
Genius-Level Creativity
Any scientific theory of personality and cognition truly worthy
of the name surely must help us to understand this humanly
vital topic, but by this standard we have so far made distress-
ingly little progress. The reason, in our opinion, is that for the
most part we have tried to understand the exceptional - real
genius, in its fullest expressions - as an amplification of the
commonplace - 'creativity,' as found in convenience samples
of undergraduates and the like.
All of the challenging phenomena catalogued in this
article - including extreme psychophysiological influence,
psychological automatisms and secondary centers of con-
sciousness, flashes of inspiration involving unusual forms of
thinking and symbolism, prodigious memory, spontaneous
psi phenomena, and altered states of consciousness verging
on the mystical realm - are inescapably bound up with genius
in its fullest expressions, but these connections go virtually
unmentioned in contemporary mainstream discussions. A
particularly dramatic case which exemplifies our central
point is that of the Indian mathematical genius Ramanujan,
rated by his distinguished discoverer Hardy as standing alone
at 100 atop a scale of mathematical ability on which most of
us lie at or near zero, while the magnificent David Hilbert
rated 80 and Hardy himself a mere 25. Replete with examples
of prodigious memory, psychological automatisms, mathe-
matical discoveries presented in the form of dreams, and
profound and beautiful intuitions of hidden but ultimately
verifiable properties of the physical world, this astonishing
case fairly beggars the theoretical apparatus currently avail-
able to cognitive science and hence could well serve as a kind
of reality check and navigational aid for further investigations
of genius.
42 Empirical Challenges to Conventional Mind-Brain Theory
Mystical Experience
Experiences of this type lie at the core of the world's major
religious traditions and have continued to occur throughout
history and across cultures. Their existence as a distinctive and
important class of psychological phenomena can scarcely be
denied, yet they have largely been ignored by mainstream
psychology and neuroscience, and generations of clinical
psychologists, psychiatrists, and neuroscientists have tended,
with few exceptions, to devalue and pathologize them, treating
them as products of malfunctioning brains. Even when
acknowledging that such experiences are typically life-
transforming and self-validating for those who have them,
the historically standard epistemological approaches in psy-
chology and philosophy - beginning with William James in
his Varieties of Religious Experience - treat them as purely sub-
jective events having authority only for those who experience
them, and thus deny their objective significance and the test-
ability of the associated truth claims. However, a large though
scattered literature testifies to the common occurrence in con-
nection with such experiences, or in individuals who have
them, of genius-level creativity, spontaneous psi-type events,
and many other unusual but verifiable empirical phenomena
of the sorts described in this article. Mystical-type states of
consciousness are also at least partially reproducible by phar-
macological (psychedelic) means, and they can be induced by
protracted self-discipline involving transformative practices
such as the various forms of meditation. An objective and
informed appraisal of mystical experience thus finds within
it much additional support for an enlarged conception of
human personality and many new opportunities for empirical
research.
The Unity of Conscious Experience
Under this heading we briefly address two interrelated pro-
blems. The first and narrower is the so-called 'binding' prob-
lem, which emerged as a consequence of the success of
contemporary neuroscientists in analyzing sensory mechan-
isms, particularly in the visual system. It turns out that different
properties of a visual object such as its form, color, and motion
in depth are handled individually by largely separate regions or
mechanisms within the brain. But once the stimulus has been
thus dismembered, so to speak, how does it get back together
again as a unit of visual experience?
Only one thing is certain: The unification of experience is
not achieved anatomically. There are no privileged places or
structures in the brain where everything comes together, either
for the visual system itself or for the sensory systems altogether.
Some early theorists such as James and McDougall argued that
the evident disparity between the multiplicity of physiological
processes in the brain and the felt unity of conscious experi-
ence could only be resolved in materialist terms by anatomical
convergence, and since there is no such convergence, material-
ism must be false. This argument, although ingenious, relied
upon the faulty premise that the only possible physical means
of unification must be anatomical in nature. All current neuro-
physiological proposals for solving the binding problem are
instead functional in nature; the essential concept common to
all of them is that oscillatory electrical activity in widely
distributed neural populations can be rapidly and reversibly
synchronized, particularly in the 'gamma' band of EEG fre-
quencies (roughly 30-70 Hz), thereby providing a possible
mechanistic solution to the binding problem.
A great deal of sophisticated experimental and theoretical
work over the past 20 years has demonstrated that such
mechanisms do in fact exist in the nervous system, and that
they are active in conjunction with normal perceptual synthe-
sis. Indeed, contemporary physicalism has crystallized neuro-
physiologically in the form of a family of 'global workspace'
theories, all of which make the central claim that conscious
experience occurs specifically - and only - in conjunction with
large-scale patterns of gamma-band oscillatory activity linking
widely separated regions of the brain.
The neurophysiological global workspace, however, cannot
be the whole story, because a large body of recent evidence
demonstrates that elaborate, vivid, and life-transforming con-
scious experience sometimes occurs under extreme physiologi-
cal conditions, such as deep general anesthesia and cardiac
arrest, that categorically preclude workspace operation. In
short, it appears to us that the early theorists were right after
all, albeit for the wrong reason. In effect, we believe, recent
progress in theoretical neuroscience, coupled with advances in
our capacity to retrieve patients from the borderland of death,
has provided new means for the falsification of physicalist
theories of mind-brain relations.
Availability of this emerging evidence emboldens us to
make some further and more speculative remarks regarding
the larger problem of perceptual synthesis, and the direction
in which things seem to us to be moving.
It is an historical fact that mainstream psychology has
always tended on the whole to try to solve its problems in
minimalist fashion and with as little reference as possible to
what all of us experience every day as central features of our
conscious mental life. The early workers in 'mechanical transla-
tion,' for example, imagined that they could do a decent job
simply by constructing a large dictionary that would enable
substitution of words in one language for words in the other.
This approach failed miserably, and we were slowly driven,
failed step by failed step, to the recognition that truly adequate
translation presupposes understanding, or in short a full appre-
ciation of the capacities underlying the human use of language.
A similar evolution is underway in regard to perceptual
theory. Most of the work to date has taken a strongly
'bottom-up' approach, which views perceptual synthesis as
a kind of exhaustive calculation from the totality of input
currently present at our sensory surfaces. Machine vision and
robotics, for example, necessarily took this approach, and even
in neuroscience it seemed to make sense to start with the most
accessible parts of the perceptual systems - the end organs and
their peripheral connections - and work our way inward. The
great sensory systems themselves - vision, audition, somato-
sensation, and so on - were also presumed to operate more
or less independently, and were in fact typically studied in
isolation.
A separate tradition dating back at least to Kant and the
early Gestalt theorists, and carried forward into the modern era
by psychologists such as Ulric Neisser and (erome Bruner, has
been sensitive to the presence of 'top-down' influences, both
Empirical Challenges to Conventional Mind-Brain Theory 43
within and between sensory modalities. Although a few per-
ceptual subsystems (such as those that produce incorrigible
visual illusions) may be truly autonomous or 'cognitively
impenetrable/ these seem to be isolated and special cases.
A very different overall picture of perceptual synthesis is cur-
rently emerging in which top-down influences predominate.
On this view perceptual synthesis is achieved not from the
input, but with its aid. This is necessarily the case for example
in regard to ambiguous figures such as the Necker cube, where
the stimulus information itself is insufficient to determine a
uniquely correct interpretation. More generally, we routinely
ignore information that is present in the input and supply
information that is not, speed reading providing a characteris-
tic example. Something within us, a sort of world-generating or
virtual-reality system, is continuously updating and projecting
an overall model of the perceptual environment and our posi-
tion within it, guided by limited samplings of the available
sensory information.
As in the case of understanding spoken or written language,
an enormous amount of general knowledge is constantly
mobilized in the service of this projective activity, which freely
utilizes whatever information it finds relevant. Top-down and
cross-modal sensory interactions have recently been recog-
nized as the rule rather than the exception in perception, and
neuroscientist Rodolfo Llinas and his coworkers have
advanced the view, which we believe is profoundly correct,
that dreaming, far from being an odd and incidental part
of our mental life, represents the fundamental form of this
world-creating activity. Ordinary perceptual synthesis, on this
inverted view of things, amounts to oneiric (dreamlike) activity
constrained by sensory input. Psychoanalyst Ernest Hartmann
has proposed similar ideas in regard to hallucinatory activity
more generally, with dreaming included. In his view such
activity is again a ubiquitous and fundamental feature of our
mental life, and the critical question is not 'why do we some-
times hallucinate?' but rather 'what keeps us from halluci-
nating most of the time?' The answer, he suggests, lies in
inhibitory influences exerted by the brain activity that accom-
panies ongoing perceptual and cognitive functions of the ordi-
nary waking sorts.
So far so good, but where exactly is the 'top,' the ultimate
source of this top-down world-creating activity? The main-
stream neuroscientists who have already recognized its exis-
tence invariably presume that it arises entirely within the brain
itself, but evidence such as that of near-death experiences
occurring under extreme physiological conditions and the
more direct evidence of postmortem survival suggest that it
may originate outside the brain as conventionally understood.
The Heart of the Mind
In this section we comment briefly on a hornet's nest of issues
lying at the core of mental life as all of us routinely experience
it, every day. These issues have been the focus of extensive
recent debates, especially in the philosophical literature, pre-
cisely because of their resistance to understanding in conven-
tional physicalist terms. The issues are deep, individually
complex, and densely interconnected, and what we can say
here will necessarily amount to little more than a summary of
our own opinions. Our central point is that the prevailing a
priori commitment to physicalism has rendered us systemati-
cally incapable of dealing adequately with the mind's most
central and characteristic properties. We should rethink that
commitment.
Consider first the issue of semantic content, the 'meaning'
of words and other forms of representation. Throughout
our history, we have tried unsuccessfully to deal with this by
'naturalizing' it, reducing it to something else that seems poten-
tially more tractable. An old favorite among psychologists was
that representations work by resembling what they represent,
by virtue of some sort of built-in similarity or structural iso-
morphism, but any hope along these lines was long ago
exploded by philosophical arguments. The central move sub-
sequently made by classical cognitive psychology is essentially
the semantic counterpart of the prevailing 'functionalist' doc-
trine in the philosophy of mind: Meanings are not to be con-
ceived as intrinsic to words or concepts, but rather as deriving
from and defined by the functional role those words or
concepts play in the overall linguistic system. Currently there
is great interest in 'externalist' causal accounts of this function-
alist type; in connectionism, dynamic systems theory, and
neuroscience, for example, the 'meaning' of a given response,
such as the settling of a network into one of its 'attractors' or
firing off a volley of spikes by a neuron in the visual cortex, is
typically identified with whatever it is in the organism's envi-
ronment that produces that response. But this simply cannot
be right: How can such an account deal with abstract things, for
example, or nonexistent things? Responses do not qualify ipso
facto as representations, nor signs as symbols. Something
essential is being left out. That something, as John Searle has
so effectively argued, is precisely what matters, the semantic or
mental content.
Closely related to this is the more general and abstract
philosophical problem of intentionality, the ability of any
and all representational forms to be 'about' things, events,
and states of affairs in the world. Mainstream psychologists
and philosophers have struggled to find ways of making inten-
tionality intrinsic to the representations themselves, but again
it just does not and cannot work, because something essential
is left out. That something is the user of the representations.
Intentionality is inherently a three-way relation involving
users, symbols, and things symbolized, and the user cannot
be eliminated. As Searle puts it in various places, the intention-
ality of language is secondary and derives from the intrinsic
intentionality of the mind. Searle thus agrees in part with the
nineteenth-century philosopher Franz Brentano, for whom
intentionality was the primary distinguishing mark of the men-
tal. At the same time, however, Searle ignores the other and
more fundamental part of Brentano's thesis, which is that
intentionality cannot be obtained from any kind of purely
physical system, including brains.
Talk of 'users' and the like raises for many contemporary
psychologists and philosophers the terrifying specter of the self
as a homunculus, a little being within, who embodies all the
capacities we sought to explain in the first place. Such a result
would clearly be disastrous, because that being would evi-
dently need a similar though smaller being within itself, and
so on without end. Cognitive modelers seeking to provide
strictly physicalist accounts of mental functions must therefore
44 Empirical Challenges to Conventional Mind-Brain Theory
do so without invoking a homunculus, but in attempting this
they routinely fail. Often the homuncular aspect is hidden,
slipped into a model by its designers or builders and covertly
enlisting the semantic and intentional capacities of its users or
observers. Much contemporary work on computational mod-
eling of memory, metaphor, and semantics harbors subtle
problems of this sort. Sometimes, however, the homunculus
is more brazenly evident. One example is David Marr's account
of vision, which applies computations to the two-dimensional
array of retinal input to generate a 'description' of the three-
dimensional world that provided that input, but then needs
someone to interpret the description. Another is Stephen
Kosslyn's model of visual imagery, which essentially puts up
an image on a sort of internal TV screen, but then needs
somebody else to view the image.
Cognitive models cannot function without a homunculus,
we believe, precisely because they lack what we have - minds,
with their capacities for semantics, intentionality, and all the
rest built in. No homunculus problem, however, is posed by
the structure of our conscious experience itself. The efforts of
Daniel Dennett and other physicalists to claim that there is
such a problem, and use that to ridicule any residue of dualism,
rely upon the deeply flawed metaphor of the 'Cartesian the-
ater,' a place where mental contents get displayed and we pop
in separately to view them. Descartes himself, James, and
Searle, among others, all have this right; conscious experience
comes to us whole and undivided, with the qualitative feels,
phenomenological content, unity, and subjective point of view
all built-in, intrinsic features. We and our experience cannot be
separated in this way.
Finally, we wish simply to record our own deepest intuition
as to where these issues lead. All of the great unsolved mysteries
of the mind - semantics, intentionality, volition, the self, and
consciousness - seem to us inextricably interconnected, with
consciousness somehow at the root of all.
The consciousness we have in mind, however, is emphati-
cally not that of people such as David Chalmers, irreducible
but ineffectual, consisting merely of phenomenological prop-
erties or 'qualia' arbitrarily tacked on to some sort of computa-
tional intelligence that supposedly does all the cognitive work.
Ordinary perception, memory, and action are saturated with
conceptual understanding, and conceptual understanding is
saturated with phenomenological content. Volition too has
an intentionality aspect, for as Nietzsche somewhere remarked,
one cannot just will, one must will something. And as William
James so forcibly argued at the dawn of our science, all of this
perceptual, cognitive, and volitional activity somehow ema-
nates from a mysterious and elusive 'spiritual self,' which can
often be sensed at the innermost subjective pole of our ongo-
ing conscious experience.
We find it astonishing, and predict that it will be found so by
our intellectual descendants as well, that so much of twentieth-
century psychology and philosophy sought - consciously! - to
slight or ignore these first-person realities of the mind, and some-
times even to deny their existence. There is perhaps no better
example of the power of preexisting theoretical commitments to
blind able persons to countervailing facts. The gloomy and coun-
terintuitive modern conclusions summarized in section 'Intro-
duction' about mind, consciousness, free will, and the self really
do follow - inexorably - from the physicalism that prevails
today. But as we will next briefly explain, that kind of physicalism
is itself incompatible with our deepest physical science.
Conclusion: Toward an Expanded Scientific Psychology
It cannot be emphasized too strongly that these unresolved
explanatory problems concerning consciousness, the heart of
the mind, and the other empirical phenomena surveyed in this
article all have a common source in the narrow physicalist
consensus which undergirds practically everything now going
on in mainstream psychology, neuroscience, and philosophy
of mind. But that consensus rests ultimately upon a classical
physics-based conception of nature, deriving from people such
as Descartes, Galileo, Newton, Laplace, and Kelvin, that began
its career by deliberately banishing conscious human minds
from its purview! Given that historical background, it should
occasion little surprise that William James - like Newton and
Leibniz before him, and like increasing numbers of philoso-
phers and scientists today - clearly recognized the inherent
impossibility of explaining consciousness and allied phenom-
ena within that Procrustean framework. James himself cau-
tioned that the physical science concepts underlying classical
physicalism were 'provisional and revisable things,' but he had
no good alternatives in sight. As he correctly anticipated, how-
ever, that conception of nature was soon radically undermined
by a tectonic shift in the foundations of physics itself, asso-
ciated especially with the rise of quantum mechanics.
The founders of quantum mechanics discovered to their
horror that the fundamental ideas of classical physics were
not just limited but wrong, leading repeatedly to predictions
falsified by experiment. The theory they were driven to in
response, quantum theory, is a more fundamental and better
physical theory that explains everything explainable in classical
terms and a host of additional things as well, often to extraor-
dinary levels of accuracy. No outcome predicted by it has ever
been experimentally falsified. Furthermore, in at least some of
its various interpretations, quantum mechanics appears able to
accommodate phenomena of the sorts surveyed here. Mathe-
matical physicist Henry Stapp in particular has shown that a
strictly orthodox interpretation derived from the mathematical
formalization achieved by von Neumann leads naturally to a
non-Cartesian form of dualism in which the human mind with
its powers of attention and decision making plays a necessary
and fundamental role in completing the quantum dynamics.
As a corollary, the classical doctrine of 'causal closure of
the physical,' which underlies most contemporary physicalist
denials of free will, is specifically rejected. And although details
remain to be supplied, many of the challenging behavioral
phenomena cited above, from stigmata and hypnotic blisters
to psi phenomena and even postmortem survival, seem poten-
tially understandable within this broader framework.
The empirical challenges briefly surveyed here should be
sufficient in themselves, we believe, to compel and to some
extent foreshadow a radical reworking of central parts of our
science of the mind. But it is also important to recognize that a
scientific psychology enlarged in these ways will likely prove
more compatible than present-day physicalist psychology both
with everyday human experience and with our most funda-
mental physical science.
Empirical Challenges to Conventional Mind-Brain Theory 45
See also: Creative and Imaginative Thinking; Creativity; Drugs, the
Brain, and Behavior; Electroencephalography; Free Will; Genius,
Eminence, and Giftedness; Hypnosis; Meditation: The Science and the
Art; Memory; The Mind-Body Problem; Near-Death Experiences;
Psychology and Religion; Savant Syndrome; Visual Perception.
Further Reading
Chalmers D (1996) The Conscious Mind: in Search ol a Fundamental Theory.
New York: Oxford University Press.
Gauld A (1982) Mediumship and Survival: A Century of Investigations. London:
Heinemann.
Hartmann E (1 991 ) Boundaries in the Mind: A New Psychology of Personality.
New York: Basic Books.
James W (1986) In: Burkhardt FH (ed.) Essays in Psychical Research. Cambridge, MA:
Harvard University Press.
James W (1890) Principles of Psychology, 2 vols. New York: Henry Holt.
Kan igel R (1 991 ) The Man Who Knew Infinity: A Life of the Genius, Ramanujan.
New York: Charles Scribner’s.
Kelly EF, Kelly EW, Crabtree A, etal. (2007) Irreducible Mind: Toward a Psychology for
the 21st Century. Lanham, MD: Rowman & Littlefield.
Llinas R and Ribary U (1994) Perception as an oneiric-like state modulated by the
senses. In: Koch C and Davis JL (eds.) Large-Scale Neuronal Theories of the Brain,
pp. 111-124. Cambridge, MA: MIT Press.
Marshall P (2005) Mystical Encounters with the Natural World: Experiences and
Explanations. Oxford: Oxford University Press.
Rad in D (2006) Entangled Minds: Extrasensory Experiences in a Quantum Reality.
New York: Simon and Schuster.
Sacks 0 (1 987) The Man Who Mistook His Wife for a Hat. New York: Simon and
Schuster.
Searle JR (1992) The Rediscovery of the Mind. Cambridge, MA: MIT Press.
Stapp HP (2007) Mindful Universe: Quantum Mechanics and the Participating Observer.
Berlin: Springer.
Stevenson I (1 997) Reincarnation and Biology: A Contribution to the Etiology of
Birthmarks and Birth Defects, 2 vols. Westport, CT: Praeger.
Strawson G (2006) In: Freeman A (ed.) Consciousness and Its Place in Nature: Does
Physicalism Entail Panpsychism?. Exeter: Imprint Academic.
Stromeyer CF III and Psotka J (1970) The detailed structure of eidetic images. Nature
225: 346-349.
T reffert D (1989) Extraordinary People: Understanding "Idiot Savants". New York:
Harper & Row.
Environmental Cognition
C Spencer and K Gee, University of Sheffield, Sheffield, UK
© 2012 Elsevier Inc. All rights reserved.
Glossary
Affective map Affective mapping as being the
opportunity to plot on maps the feelings that particular
places evoke. Feelings are shown by symbols,
possibly supplemented with annotation. Where patterns
of feelings can be identified on maps then areas
can be shaded accordingly on drawings of the
cognitive map.
Allocentric frame of reference Frameworks that are fixed to
the environment itself or to the individual object.
Being lost Lost is defined as being unable to identify or
orient present one's location with respect to known
locations; and having no effective method for reorientating
oneself. Note: one can have little accurate knowledge of
where one is, but not be lost.
Cognitive map A construct which enables a person to
predict the environment which is too large to be
perceived at once, and to establish a matrix of
environmental experience into which a new experience can
be integrated, simplifying and ordering the complexity of
human-environment transactions. Investigators often
capture the individual's cognitive map by asking for a sketch
map to be drawn.
Dead reckoning (or path integration) The process which
enables a traveler to constantly update their current position
with respect to their starting point without recording details
of the path already followed.
Egocentric spatial frameworks Frameworks that move with
the body as it moves through the environment.
Environmental grammar The terms of reference
acquired by an individual in relation to the spaces and
settings around them, from which they develop the
appropriate expectancies and hence, the corresponding
search strategies.
Legibility The legibility of a city as ". . . the ease with which
its parts may be recognised and can be organised into a
coherent pattern" (Kevin Lynch).
Space syntax The general idea is that spaces can be
broken down into components, analyzed as
networks of choices, then represented as maps and
graphs that describe the relative connectivity and
integration of those spaces: (Originally conceived by
Professor Bill Hillier, Julienne Hanson and their
colleagues at The Bartlett, University College
London in the late 1970s to early 1980s as a tool
to help architects simulate the likely effects of
their designs.)
Wayfinding The process of determining and following a
path or route between an origin and a destination, as a
purposive, directed, and motivated activity.
Introduction
Environmental cognition concerns the way we acquire, orga-
nize, store, and recall information about locations, distances,
and arrangements in the physical environment. Other aspects
of perception and cognition receive their own entries elsewhere
in the encyclopedia: Why then is there an entry specific to the
perception and cognition of the environment?
Most of the perception and cognition research in psychol-
ogy concerns isolated objects, often studied within a labora-
tory, and without a broader meaningful context. In contrast,
we are often surrounded by and are moving through the
(hugely complex) environments that are the focus of study
within environmental cognition. Theories of perception and
cognition have always been central to psychology as a whole,
and this is true of the relatively recent area of environmental
psychology (qv), as individuals perceive, interpret, and respond
appropriately to their surroundings.
Throughout our daily lives, literally hundreds of spatial
decisions are made each day. These rely not on formal refer-
ence systems such as maps, but on previously gained spatial
representations of the environment. Environmental psycholo-
gists define these as cognitive maps. As a result, the environ-
ment has a profound effect on human cognition, action, and
well-being.
The Nature of Stored Spatial Information: Analog,
Propositional, or Both?
We encode and manipulate environmental information in
numerous ways. Evidence exists for two kinds of coding: prop-
ositional and analog, as in a Conventional Map. The proposi-
tional storage of environmental information, in lists and
associated networks, is based on abstract representations of
meaning. Analog codes, however, parallel the physical struc-
ture of information in the world, emphasizing continuity over
space. Neither is sufficient alone: both are used according to
need and preference.
People are able to convert propositional into analog: for
example, the verbal descriptions of configurations or routes
into mental representations. Thus, cognitive maps must be in
effect both analog and propositional in nature.
Is configuration ( or survey) knowledge metric or topological?
A metric representation would imply that distances and
directions are directly encoded between environmental loca-
tions. Alternatively, if it is topological, then only the relative
spatial relationships would be encoded.
Challenging any theory is the seemingly instantaneous
development of configuration knowledge in certain situations
as opposed to other cases where it does not seem to develop at
all, even after extremely long periods of exposure.
46
Environmental Cognition 47
The debate continues as to what constitutes configural
knowledge as opposed to route knowledge. Most describe
configural knowledge as a well-connected network of locations
(or landmarks) with multiple potential paths between them.
However, reasonable estimations as to the direction and
distance of a remote location can often be made without
creating a clear route plan.
The Way the Brain Acquires, Stores, and Handles
Environmental Information. Can We Identify Key
Structures?
The representation of space may be distinct from other forms
of conceptual and perceptual knowledge. In many species of
mammals, there have been identified specific behavioral
mechanisms for locating objects, for dead reckoning, and for
constructing cognitive maps. Such mechanisms probably also
form the basis of human spatial cognition.
So what are the neural mechanisms supporting spatial cog-
nition, orientation, and wayfinding? The discovery of place
neurons in the hippocampus, and other evidence for the cog-
nitive map theory of hippocampal functioning, suggest that
this brain structure is the core of an extensive neural system
for representing and using information about the spatial
environment.
Evidence comes from lesion studies (damage invariably
impairs place learning and navigation) and from neurophysi-
ological recording studies, which show that neurons in this
system code for location, direction, and distance, which are the
elements needed for a mapping system.
A second brain structure involved in spatial cognition is the
posterior parietal cortex, where multimodal sensory systems
converge, and link to the motor cortex.
Evidence from lesion, human functional brain imaging,
and human functional brain interference studies indicate that
distinct areas within the posterior parietal cortex have different
functional contributions.
But the core function of the parietal cortex in spatial cogni-
tion is the constitution of 'egocentric' spatial frameworks: that
is, frameworks that move with the body as it moves through
the environment.
In contrast, the hippocampus not only plays an important
role in episodic and spatial memory but also provides a neural
substrate for the representation of spatial location within an
'allocentric' frame of reference (i.e., frameworks that are fixed
to the environment itself or to the individual object).
Thus, the hippocampus and the posterior parietal cortex
would act differently by handling topological and metric infor-
mation, respectively.
Only a close cooperation of both systems would enable the
individual to 'overlay' and integrate the relevant egocentric and
allocentric frameworks. This is the most important prerequisite
for performing many of the complex spatial tasks successfully.
Cognitive Maps in Animals and Humans as Drivers
of Action
Early psychology's account of a maze-running rat would
be of a sequence of several stimulus-response connections
that formed a route to the goal. Success at getting there
could not be taken as evidence that the rat had an image or
cognitive map of the whole area: it might only be recalling
a series of learned responses to the various choice-points in
the maze.
But observations of a rat escaping from a maze and run-
ning direct to the goal (i.e., being able to integrate the
various subroutes within the maze into a single direction)
lead Tolman in 1948 to use the term Cognitive Map in explain-
ing this ability.
Rather than state that animals, particularly the rats that
took short cuts across his mazes, stored spatial information
as a map, Tolman used the term Cognitive Map metaphori-
cally. In other words, he suggested that the animals used in
his studies appeared to be able to use spatial information as
though the places they remembered were recorded in a map-
like manner.
Much debate has since taken place on whether animals
do have cognitive maps, or whether for example they use
dead reckoning, or path integration, the process which enables
a traveler to constantly update their current position with
respect to their starting point without recording details of the
path already followed. Without a memory trace of the path
taken, exact route retrace might not be possible; but it would
still be possible to return home directly with foraged food, or
in the case of an emergency.
Much subsequent research has shown that birds and flying
insects variously use landmarks, star and sun compasses, and
magnetic compasses for their local, longer distance, and migra-
tory travel. Are these landmarks, etc. used as a memorized
sequence indicating a route to be followed or represented as a
layout or survey overview? Careful and ingenious studies are
needed to answer such questions.
The term cognitive maps had to wait until the early 1970s
until it received wide usage in the literature about human
spatial decision-making. Environmental and developmental
psychologists, human geographers, and others have used the
term with slightly differing implications, the broadest defi-
nition being the cognitive apparatus that underlies spatial
behavior. More specifically it has been defined as a construct
which enables a person to predict the environment which is
too large to be perceived at once, and to establish a matrix
of environmental experience into which a new experience
can be integrated, simplifying and ordering the complexity of
human-environment transactions.
To talk of a cognitive map is to imply that a human
or animal cognizes the familiar environment as a system of
interconnected places; is able to apply a set of transformation
rules to this system, optimizing goal-directed movements; and
is able to pilot and perform novel routes within it.
Decisions about action are not, however, purely based on
knowing where resources and routes are: that is, the factual
aspects of the cognitive map. They must also relate to the
attributes, meanings, and values that shape behavior: we
should thus talk of an affective map.
Therefore, we should see cognitive-affective maps as
integrating images, information, and attitudes about an
environment.
They are not isolated and contextless entities: they are
formed during purposive and social activity in the person's
48 Environmental Cognition
everyday world. Wayfinding (see below) relies on landmarks
imbued with meaning.
Cognitive maps are in effect a series of knowledge (and
feeling) structures which consist of different levels of detail
and integration. These knowledge structures develop with
age, education and experience, thus increasing the information
and associations held. We in effect extract and combine infor-
mation according to the task in hand: that is, there is no one
cognitive map in memory but rather we construct them for
specific events.
In this respect cognitive maps are dynamic, and are far
from a standard cartographer's map, in that they are sche-
matic, incomplete, distorted, simplified and idiosyncratic.
The metaphor of a map in the head is so persuasive that
we are tempted to believe that there is no metaphor, but rather
a real 'thing.'
The main argument against such a cognitive map construct
is the non-Euclidean properties discovered by some methods
of investigating our knowledge structures, and the fact that we
have incomplete knowledge not integrated into one single
'map' but rather disconnected components.
People's behavior in large-scale environments can be
explained more completely through recourse to internal,
subjective factors than by more traditional external, 'objec-
tive' factors. Ask people to plan a shopping trip round
town with a map in front of them, and you may get a
'rational route'; but plot an actual trip that they make, and
immediately these more subjective factors are seen to come
into play.
The capacity to act and move intelligently through space
may well occur before, and possibly in the absence of, the
capacity to represent that space.
Very young infants can navigate in ways that demonstrate
detailed spatial knowledge well before they could attempt any
public representation of that knowledge by, for instance, draw-
ing a sketch-map of a familiar area.
Sketch-mapping has nonetheless been a technique widely
used by researchers trying to elicit the elusive cognitive map,
despite the obvious limitations imposed by people's general
ability to draw. Systematic distortions can be demonstrated in
such external representations: distances become exaggerated
in areas which have many features of interest, and corre-
spondingly telescoped in less-featured areas; angles are misre-
membered, such that the map does not fit together; whole
areas are neglected.
Some researchers have preferred instead to ask subjects to
estimate a series of relative distances and angles, and thence
to use multidimensional scaling to reconstruct an underlying
cognitive map, thereby avoiding individual differences in
drawing abilities. Other techniques have included locating
points on a base map; recognizing features on aerial photo-
graphs; and obtaining think-aloud protocols.
Even within the sketch-map technique, very different
impressions of the presumed underlying cognitive map can
be obtained according to the particular demand that the
researcher's question imposes; and how much prior spatial
cueing is offered. No single test is best.
And some may indeed be misleading: consider that if
humans conceive of space as collections of familiar landmarks,
then the spatial representation of landmarks is thus
qualitative in the sense that it is imprecise: it does not depend
on precise metric information, and therefore should not be
elicited as if it were.
Direct and Indirect Sources of Information About the
Person’s Environment
All senses can be involved in one's direct learning about
the environment; but the sense most studied is vision, justi-
fiably so, as this sense affords one the most immediate,
complete, detailed relational information about the sur-
rounding environment. Whatever the scale, whether it be
at the moment one enters an unfamiliar room, or one has
a 360° view from a mountain top, vision provides an
instant ordering of the elements of the room-scape or land-
scape, in a way that tactile exploration of a room could
only slowly and partially allow. (In section 'Environmental
Cognition Without Sight; and Devices for the Blind,' cogni-
tion of space without vision will be further considered,
along with measures to afford the blind better awareness
of space.)
Visual exploration of the space around the newborn
starts from the earliest; and as soon as the infant is capable
of active movement, investigation of the immediate environ-
ment leads to the integration and interrelating of multiple
viewpoints when moving through. These integrative skills
in developing a cognitive map of the layout of a room and
then a house will be used in wider and wider contexts, from
immediate neighborhood up to whole township, district,
and region.
Perfect integration, however, is not easily achieved:
people find it difficult to interrelate places on different floors
of even a familiar house; and can often be surprised to see
how close two places in different sections of town are
when they are separately known from different habitual
journeys, especially if they cannot be both seen at the same
time. (The aerial view of a familiar place often affords
surprised discoveries.)
Such an overview can also be provided by a surrogate: a
map, whose raison d'etre is to show these interrelationships
and locations. We can learn relative distances and locations via
such a representation, and plan efficient journeys with their
aid. (As section 'Communicating and Sharing Environmental
Information' will claim, there is evidence that the act of
mapping is a cultural universal.)
In many cases, our knowledge of a particular area derives
from both direct and indirect inputs, from exploration and
map-use; and the interrelationship of these two in the cogni-
tive map is fascinating but as yet understudied.
Cognition of more distant environments may be predomi-
nantly or entirely from indirect sources: not only maps but also
television, film, photos, textbooks, travelers' tales, etc. Such
sources must inevitably offer only a partial and selective view
of a place and its people, and may indeed be a source of
prejudices or of over-optimistic expectations.
Even the very names of distant places may convey infor-
mation (sometimes misleading: think of Erik the Red's propa-
gandizing naming of Greenland as a good place to settle
and farm).
Environmental Cognition 49
Maps may be our primary graphic source of information at
the global scale, and information and impressions drawn can
be influenced by the map-projection selected.
Communicating and Sharing Environmental
Information
How do we communicate our environmental information to
others: for example, in sharing information about the location
of resources, or directing others to a destination?
Whether this is given via spoken instructions, or in graphic
form, the process involves selecting a subset of the stored
information, selected partly as a result of the sender's articulacy
or graphicacy, and partly with regard to an appreciation of any
shared understandings with the receiver of the message. An
interesting insight into the selection process can be given by
this simple comparison: first ask for verbal instructions; and
then on another occasion, ask for them to be given graphically:
subtly different subsets of information will be received.
How successful can we be in communicating route knowl-
edge verbally? Studies indicate that remembering and follow-
ing route directions can be facilitated by presenting the receiver
with directions in the correct temporal-spatial order; concen-
trating information in statements concerned with choice
points; suppressing irrelevant (decorative) detail; and by offer-
ing referents with which the listener will be familiar. Not
surprisingly, navigation from good descriptions results in
fewer mistakes than navigation from poor descriptions.
It has been claimed by the geographer Blaut that mapping is
a universal in human culture; that it is an important ecological
adaptation in all cultures (and the supportive evidence from
the anthropological literature is strong); that map-like models
of one sort or another have been made in all cultures since
prehistoric times (again, there is strong support from the
archaeological record); and that mapping ability emerges nat-
urally (without training) in very young children of all cultures,
a proposition hotly contested by conventional Piagetians.
The Piagetian position maintains that children younger
than seven are 'preoperational' and therefore cannot handle
or interpret the conventions involved in maps: of change of
size, aerial perspective, symbolism, etc. However, many studies
with much younger children show them capable of under-
standing such conventions and being able to use relatively
simple maps to guide their own behavior.
Thus for example, 3- and 4-year-olds can use a simple map
of a maze to reach their goal (i.e., a local area map); and cannot
only understand that an aerial photograph can represent a
familiar larger scale area, but can use it to navigate to 'hidden
treasure.' At a larger scale again, young preschoolers can use an
aerial photograph of a city to tell a story about imaginative
journeys across it, demonstrating clear understanding that the
size-reduced, aerial-perspective features on the photograph
stand for real, full-scale buildings, roads, trees, etc.
These untaught (hence 'natural') mapping abilities of
preschool-age children have been studied in many cultures.
And confirmatory evidence comes from toy play, in which
many young children have been able to use elements of a
map (schematic roads, houses etc.) to describe imaginary
places and journeys.
So, instead of the restricted, Piagetian, view of young chil-
dren's environmental cognitive abilities, the evidence from
many such studies using aerial photographs and simple
maps would seem to indicate that the cognitive modeling
of the environment may be an ecological necessity. Young
children in every culture seem capable of modeling, in two
or three dimensions, the landscape as through viewed from
overhead and reduced in scale; and this is prominently assisted
by toy-play and by constructing imaginary maps of what the
world may look like from overhead. Do not children every-
where make model landscapes with toys on the floor or mark-
ings on the ground?
The Development of Environmental Competence
In early childhood, we have to develop our general compe-
tence in understanding and navigating environments: this will
take a period of years (though probably not as many as early
developmental psychologists believed). As competent indivi-
duals, we still have to develop competence with respect to
places newly encountered: this will start immediately, with
our first impressions, and we will rapidly acquire a survey
structure, to which extra information will be added as our
travels through the place continue.
The first, the development in early childhood of the
basic competence, can be called the Ontogenesis of compe-
tence. The second, the application to a new place, can be called
Microgenesis. Some theorists have suggested that the stages
in microgenesis recapitulate those seen in the ontogenesis of
competence.
Perhaps the trend in developmental psychology to stress the
lack of early competence in many areas, including spatial, can
be mainly attributed to Piaget. Famously, in considering the
young child's failure on the 'Three Mountains' experiment,
(describing what a toy figure would see in a model landscape
as being what the child was seeing, regardless of the toy's
position), Piaget generalized this as evidence of an inevitable
egocentricity before 7 years or so.
Critics have suggested that young children find the instruc-
tions in these classic experiments confusing; and that with
more careful testing, we can see nonegocentric thought evident
in much younger children.
But it is clear that the youngest children are not fully com-
petent, and do go through a number of stages in acquiring it.
The account given, by Siegal and White, has become the pre-
dominant (but not unchallenged) one.
They suggest that initially, knowledge is structured around
landmarks. The second stage is where such locations are linked
as route knowledge; and only at the third stage can the individ-
ual be said to have survey knowledge. At first, the space between
the known points or landmarks is 'empty'; and the individual
cannot relate them to each other. Developing knowledge of set
routes gives the individual the beginnings of a structure, but
to start with it will consist of a series of local directions at
decision points, with little accurate idea of distances. Survey
knowledge encodes both distances and directions between
landmarks, and allows more innovative route discovery.
The attractiveness of the Siegal and White sequence is that it
can describe both the slow build up of the general skill during
50 Environmental Cognition
early childhood, and the rapid acquisition of knowledge about
a newly encountered place by a skilled older individual.
The original account of ontogenesis does not specify how
long it takes the young child to progress through the stages, but
implies early incompetence lasts several years. Does the empir-
ical evidence bear this out? As we saw in section 'Communicat-
ing and Sharing Environmental Information,' Blaut and
colleagues suggest not.
Similarly, the account of microgenesis, suggesting that the
survey is dependent on accurate estimates of distances along
routes and between landmarks is open to empirical challenge;
and Montello has recently shown that spatial knowledge,
including this metric knowledge is acquired gradually through
continuing experience of the new place. And the eventual 'map'
may well still include inconsistent data: a geometry of esti-
mates which do not fully match up.
When investigators ask respondents to draw a sketch map
to represent the individual's survey knowledge, the very
demands of the task may produce apparent consistencies
which are not borne out by that individual's actual behavior
in the real world. Only a very few studies have taken the time
and care to study the actual process of microgenesis over the
days and weeks that it may take to develop survey knowledge
of a new place; and although they give some support to the
landmark-route-survey sequence, it is clear that the resultant
'map' is highly individual, relates to the pattern of the indivi-
dual's exploration and interests, and will contain both highly
detailed and completely ignored subareas. Adults learn some
key landmarks very quickly, and start anchoring new knowl-
edge to major, features routes and boundaries. Direction and
distance estimates can develop rapidly, and then level off as
habitual local travel patterns are established.
But do adults integrate newly learned routes into metrically
accurate map-like survey knowledge? Could they demonstrate
this by, for example, accurately navigating a novel short-cut?
Experimental studies of microgenesis show that people make
these short-cuts in roughly the right direction and distance; but
only at a low level of accuracy, and that the implied cognitive
map is not fully Euclidian. People making short-cuts are highly
dependent upon visible landmarks.
Returning to the studies of ontogenesis, there are only a few
studies with the time to follow the whole process through; and
most are reliant on tests of the young child's competency at
particular ages. These latter would be carefully controlled tests
of spatial problem solving (often at table-top level) and spatial
cognition, rather than studying what children are actually
doing within their real world. Developmental psychology's
study of children's cognition in general has placed little
emphasis on the ecological or sociocultural factors supporting
and developing the competencies. In contrast, from the start,
environmental psychology has sought ecological validity for its
studies, and sees the individual and their environment in
transactions right from the start of life.
If one tests children with unfamiliar materials and in
strange contexts, then they are less likely to demonstrate the
skills they have developed in their everyday transactions with
the world. Culturally aware field studies of children's spatial
behavior by Hart, for example, have shown how independence
of travel by the young directly relates to their levels of local
knowledge and thence their basic spatial competences.
Wayfinding, Becoming Lost, and Signage
Environmental cognition, it could be argued, has its main role
in supporting wayfmding: in the planning and carrying out of
routes through the environment, in locating resources and
avoiding dangers, and having an awareness of possible novel
routes were the need to arise.
This involves long-term spatial memory of places and their
attributes, survey knowledge of spatial interrelationships, and
accessible place-tagged lists of resources: in other words, a way
of linking survey with listed knowledge.
Individual differences in success in wayfmding are often
anecdotally recounted, but can be borne out in part by
research: factors involved include people's spatial processing
abilities, as well as lifetime experience with spatial problem
solving in full scale environments (often related to an indivi-
dual's drive to explore and to seek novelty). Other, more
situation-specific factors can include: familiarity with a par-
ticular setting (and previous experience of similar types,
developing an 'environmental grammar'); the density and
salience of landmarks, and the complexity and ambiguity of
the layout.
All models of wayfmding distinguish between the processes
of the planning of a route, and its execution via actions that are
time- and place-appropriate. There will be a hierarchy of pro-
cesses which unfold as route-following progresses: for exam-
ple, there will be a sequence of expectancies, to find particular
landmarks (maybe personal markers) at expected places.
Where such expectancies are not met, the traveler will have to
formulate a new plan. Where this does not work, the person
may react as being lost (see below).
Routes to distant places may be planned only at a coarse
level, but closer to target the planning will become much finer-
detailed (in the same way that migrating birds' strategies shift
from the general directional to the specific when arriving close
to the familiar end of route).
Wayfmding strategies have been studied using a range of
techniques. Verbal report protocols, whether given en route
or in retrospect, may be vivid, but can only tell us of the
processes people are aware of and can articulate. They can,
however, demonstrate this two-stage process of an initial plan-
formulation stage and a detailed plan execution, in which new
and possibly unexpected features can be reacted to. Analysis
of such protocols can also show that initially rational plans
are often subverted by chance events and sightings en route.
Eye-tracking techniques will pick these attentional pat-
terns as people travel a real or indeed a virtual route; and
can be used in conjunction with protocols. Recent studies
have added fMRI to this range of investigative tools; and
together, they can track the moment-to-moment changes
in the way environmental cognitions direct responsive way-
finding, adding to the stock of place-located knowledge as
they go.
Such studies highlight individual differences in route
planning and following. A small proportion of people report
using cardinal directions; more use an egocentric frame of
reference. Occasional individuals report planning the route
back from the goal; more often, the plan is assembled from
well-rehearsed segments, even if a novel route might have
been more efficient. Studies of specialist wayfinders, such
Environmental Cognition 51
as taxi-drivers, have used the full range of investigative techni-
ques: their ability to produce highly detailed accounts of what
they had been thinking during wayfinding can be corroborated
by eye-tracking data.
Study of human spatial cognition should contribute to
the design of buildings and local layouts which are more
easily understood. Complex buildings such as hospitals
and airports, which are multilevel and perhaps counter-
intuitive, are likely to lead to wayfinding difficulties which
good signage can only go part way to resolve, particularly
if individuals are stressed or anxious at the time. Design
guidelines might include architecturally unambiguous entry-
points, after which there could be early open areas which
enable users to survey the possible routes forward. Transi-
tions between floors and areas should be clearly signaled,
and distinctiveness planned in with a view to developing
the newcomer's cognitive map of the area, perhaps with
the aid of a well-designed and orientated 'You are here'
map, to shape a new local cognitive map. Signage must then
support and develop this information, be consistent, fre-
quent, and comprehensible. There are implications for
the signage of emergency exit routes: research indicates
that under stress people do not so much panic as resort
to more basic route-following strategies; and could well be
better guided to safety by vocal instructions over a loud-
speaker system.
'Being lost' has been studied by environmental psychol-
ogists advising police and rescue services, using case studies to
show how people of different ages (and skills) are likely
to react; and thereby psychology can suggest optimal search
strategies. If the missing person is a child under 3-years-old
in open country, for example, they may not even think of
this as being lost: after a brief exploration of a small area,
they are likely to find a local place in which to shelter, thus
becoming hidden from searchers. In contrast, a more confi-
dent older child or adult is likely to travel much larger dis-
tances in attempting to return home, perhaps using faulty
search strategies in their increasing alarm. During the preced-
ing travel in a novel area, the landmarks may have only been
sketchily encoded, and so the would-be returner may use
falsely remembered cues, and travel considerable distances
from their start.
The Legibility of Places
This section considers the concept of legibility and interpret-
ability of buildings, areas, whole townscapes; and whether
legibility can be a key link to the aesthetics of buildings and
places.
Some environments are easier to cognize than others: the
concept of legibility was introduced by the planner Kevin
Lynch to describe this ease with which one can form a cognitive
map of a city. Asking what makes a place legible, and analyzing
the sketch maps drawn by residents, he suggested that legibility
is facilitated by the presence and organization of five main
elements:
Paths: the travel corridors of the city
Edges: features which limit or enclose districts (e.g., rivers)
Districts: subareas with common, distinctive features (e.g.,
architectural, functional)
Nodes: points of intersection of key paths
Landmarks: visible and distinct features (e.g., tall buildings)
The presence of such elements helps one to organize obser-
vations and knowledge of a city: by chunking it into districts,
one can cope with possible information overload.
Cities that have a clear spatial structure and physical dis-
tinctiveness are easy to 'read,' enabling clear cognitive maps.
But what of the link to aesthetic appreciation of places? Many
places which are famed for their beauty do indeed have high
legibility.
But there can be townscapes that are highly legible, yet
which fail to engage our aesthetic or emotional response. In
addition to Lynch's list, the architectural quality of the ele-
ments, the balance between simplicity and surprise, the level
of visual and actual 'clutter,' and the presence of natural fea-
tures, including water, have all been shown to contribute to the
judgment of environmental quality. Level of upkeep, mainte-
nance, and indications of care also contribute to the positive
impression.
As an example of simplicity and surprise, a town which
reveals itself straightway to the newcomer is adjudged less
interesting than one which unfolds as one travels inwards;
but beyond a certain level of complexity, this can become
aversive and confusing. Similarly, the most positive aesthetic
judgments of individual buildings can be shown to relate to an
optimal middle range of complexity.
Formal aesthetics includes the idea of order and variables
such as rhythm, complexity, and proportion. Symbolic aes-
thetics has been defined as "pleasurable connotative meanings
associated with the content of the formal organization".
Many studies have shown that the presence of natural ele-
ments in a scene can evoke positive emotions, induce positive
physiological conditions, facilitate cognitive functioning,
evoke positive and prosocial behaviors, and help restore the
individual. This is true whether the scene is experienced
directly (walking through an area) or indirectly (viewed
through a window, or even seen in a picture).
Restorative environments and scenic beauty may be linked.
Kaplan's Attention Restoration theory states that exposure to
restorative environments engages fascination or low-effort
attention, promoting recovery and the opportunity to rest
and reflect. In contrast, exposure to nonrestorative environ-
ments engages directed effortful attention and negatively
affects mood, performance, and psychophysical well-being.
Eye movements are a way to measure attention when view-
ing a scene: they are unconscious adjustments to the demands
of attention during visual experience.
Eye movements related to photographs low on fascination
were characterized by greater exploration and a greater number
of fixations compared to those rated high on fascination,
though the viewing time was the same. Scenes high on fascina-
tion were viewed without really focusing on particular features.
Differences in eye movements suggest that less effort is
required to view nature than urban scenes, which is consistent
with Kaplan's description of 'soft fascination'; which itself
derives from William James' distinction between involuntary
and voluntary attention.
52 Environmental Cognition
Environmental Cognition Without Sight; and Devices
for the Blind
Vision must be accepted as the most important spatial sense,
for developing the cognitive map and for updating it as
the individual travels. Yet blind and visually impaired people
can become accomplished independent travelers using the
other senses, sometimes supported by traditional and modern
technological aids.
Some 30% of adults with visual impairments make no
independent journeys out from their home; and many others
only use a few set routes. Hence studying the cognitive maps of
the blind, and their development, can have a social as well as
academic purpose.
Early theories stressed Deficiency: stating that the congeni-
tally blind would be unable to develop a general spatial under-
standing because they would never have experienced the
necessary perceptual process: this has been experimentally
discredited.
In contrast, Inefficiency theory argues that awareness from
auditory and haptic cues will be less efficient than that from
sight; and as a result, the blind will depend upon route-based
rather than survey or map-like ideas for travel.
Much empirical evidence now supports the Difference
theory, which proposes that visually impaired people pos-
sess the same abilities to process and understand spatial
concepts, and that although functionally equivalent to the
sighted, they are carried out in a different albeit often slower
manner. Their spatial reasoning, like an internalized geo-
graphical information system, may combine image-based
cognitions, which will hold relative positions of places within
a frame; and language-based cognitions, which will code
place a set of propositions.
For the sighted, much exploration is done visually, and
a whole new scene generates a new local cognitive map, with
all its implications for action. For the blind, usually much
more exploration will be needed to gain information about a
new place.
Mobility teachers working with blind children have
described them as 'movement recallers' who have move-
ment memories, rather than visual image memories. Their
cognitive maps include the relative positions of objects
(which act as local landmarks), and the routes leading to,
around, or away from landmarks, are composed of movement
memories.
Some individuals develop skills which enable them to
achieve very accurate independent travel, using, for example,
careful sensing of the position of walls. Others will rely on
supports such as canes and guide dogs for their mobility.
A whole range of higher technology devices are being devel-
oped: some of which are located out in the environment
(e.g. acoustic beacons and 'talking signs'), others of which are
handheld devices providing updatable local information.
Attempts have also been made to translate visual information
into haptic, to simulate the visual overview.
And tactile maps have been used for many years with
great success to develop the blind traveler's spatial expec-
tations of a new area; these can afford perhaps the most
immediate sense of the layout and features of a place yet to
be explored.
Individual, Specialist, and Cultural Differences in
Environmental Cognition
Popular beliefs abound about differences between the sexes in
spatial awareness and skills; similarly, claims have been made
for special awareness among specialist groups (taxi drivers,
orienteers, etc.); and for different cultural groups. What is the
empirical evidence?
The most controversial is the issue of sex differences:
there is evidence to support the contention that there is a
relationship between some measures of spatial ability and
the accuracy of cognitive maps. Males may frequently out-
score females on spatial tasks which involve the mental
rotation of abstract figures (and thus visualization), and on
some table-top scale manipulations. But these may not
translate into a real-world advantage, where wayfinding,
map use, and place learning are involved. There are many
spatial tasks where the sexes do not differ, or so substantially
overlap that personal styles in learning and recall become
more important.
How could we assess individual differences or gender dif-
ferences? We should differentiate between wayfinding tasks
and wayfmding means. Tasks include traveling to a previously
known destination, exploration with the purpose of returning
home, and traveling to a novel destination.
Means include oriented search, following a continuously
marked trail, piloting (between landmarks), habitual locomo-
tion, path integration, and reference to a cognitive map.
Truly individual differences do exist: there are a few people
who perform at astonishingly high levels, and who can acquire
surprisingly accurate metric knowledge relatively quickly,
including relational knowledge about places between which
they have not directly traveled. Could we train other people
to gain a better cognitive map of the environment? What are
good strategies adapted to people's existing strengths and
experience?
For strategies seem to be key to understanding the acqui-
sition of the exceptional spatial knowledge of a city by
its taxi drivers and other specialists; and the heightened
awareness of environmental cues developed by groups as
various as prospectors, hunters, police, rescuers, hang-gliders,
and divers.
The whole of the anthropological literature attests to
similar sensitivities being developed in support of living in
particular settings. Where these settings are especially challeng-
ing, then again a heightened awareness of subtle cues for
wayfinding and resource-location can be seen as vital to sur-
vival. There are, for example, well-attested accounts of the
sensitivity of Aboriginal peoples to individual rocks and plants
as cues when traveling through the Outback of Australia; and
of the Polynesian voyagers' heightened awareness of the
significance of cues in the sea and the stars.
Electronic and Other Devices for Storing and Using
Environmental Information
What impact upon the individual's environmental cognition
have the various electronic devices for storing and present-
ing information? In-car satellite navigation and pedestrian
Environmental Cognition 53
navigation assistance systems for wayfinding provide route-
based information, generally without any survey context, and
demonstrably reduce the user's spatial information processing
to a minimum. Driving simulators and virtual travel via a
computer screen are likely to offer something closer to real-
world exploration; but still here, knowledge acquisition and
the development of survey-understanding of places will be
inferior, and indeed the less-realistic conventional map of a
place will afford better survey and route knowledge.
Navigational assistance systems are nonetheless helpful and
supportive for the newcomer, often providing a sequence of
view-based (i.e., egocentric) visual images together with direc-
tions for travel. Maps, in contrast, offer an allocentric perspec-
tive, giving a survey model of place by requiring more active
spatial information processing.
Navigating real spaces using devices, or navigating large
virtual environments, will be optimized if we can link them
to 'normal' human processing of environmental information
and be aware of its limitations: virtual information environ-
ments that offer 3D representations, etc. can overwhelm many
users' abilities to filter and represent. As a result, users fre-
quently experience disorientation in navigating large digital
spaces to locate and use information.
Conclusion : As we have seen, the cognition of the environ-
ment differs from the perception and cognition of objects, the
usual focus of studies within psychology, in more than scale.
It guides decision making about movement through space;
it is linked with feelings and aesthetics, which in turn may
be associated with personal well-being. Research in this area
has potential benefits for professions as diverse as architec-
ture, the design of virtual worlds, and the mobility instruction
of the blind.
See also: Ecological Psychology; Environmental Psychology; Our
Cognitive Map; Perceptual Development; Spatial Orientation; Spatial
Perception.
Further Reading
Conroy-Dalton RA (2003) The secret is to follow your nose: Route path selection and
angularity. Environment and Behavior 35(1): 107-131.
Foo P, Warren WH, Duchon A, and Tarr MJ (2005) Do humans integrate routes
into a cognitive map? Map- versus landmark-based navigation of novel
shortcuts. Journal of Experimental Psychology: Learning, Memory, and Cognition
31(2): 195-215.
Garling T and Evans GW (1 991 ) Environment Cognition and Action. Oxford: Oxford
University Press.
Golledge RG (1992) Place recognition and wayfinding: Making sense of space.
Geoforum 23(2): 199-214.
Golledge RG (1999) Wayfinding Behavior: Cognitive Mapping and Other Spatial
Processes. Baltimore, MD: Johns Hopkins Press.
Jacobson RD (1998) Cognitive mapping without sight. Journal of Environmental
Psychology 18: 289-305.
Kitchin R and Blades M (2002) The Cognition of Geographic Space. London and New
York: Tauris.
Kuipers B (1978) Modeling spatial knowledge. Cognitive Science 2: 129-153.
Kuipers B (1982) The ‘Map in the Head 1 metaphor. Environment and Behavior 14(2):
202 - 220 .
Kuipers B, Tecuci DG, and Stankiewicz BJ (2003) The skeleton in the cognitive map:
A computational and empirical exploration. Environment and Behavior 35: 81-106.
Lynch K (1960) The Image of the City. Cambridge, MA: MIT Press.
Siegel AW and White SH (1975) The development of spatial representations of large-
scale environments. In: Reese HW (ed.) Advances in Child Development and
Behavior 10. New York: Academic Press.
Spencer CP and Blades M (2006) Children and Their Environments. Cambridge:
Cambridge University Press.
Spiers HJ and Maguire EA (2008) The dynamic nature of cognition during wayfinding.
Journal of Environmental Psychology 28: 232-249.
Tolman EC (1948) Cognitive maps in rats and men. Psychological Review 55: 189-208.
Environmental Psychology
R Gifford, University of Victoria, Victoria, BC, Canada
© 2012 Elsevier Inc. All rights reserved.
Glossary
Behavior setting A location that consistently evokes certain
patterns of behavior in those who are present.
Crowding The experience of insufficient space; to be
distinguished from density.
Density A physical measure of persons per unit area.
Personal space or interpersonal distance The physical
distance component of interpersonal relations.
Place attachment, place identity A strong person-place
bond that develops through long or intense association; the
place becomes part of the self.
Privacy The process of regulating the amount of social
interaction in which one is involved; can mean solitude-
seeking but can also mean seeking company.
Programming (architectural) Planning buildings to meet
the behavioral and psychological needs of those who will
use them.
Postoccupancy evaluation Evaluating buildings to
determine whether and how well the architectural program
was executed.
Social design Architectural design that emphasizes
inclusive, grassroots, democratic participation by those who
will use the building; interior function and meaning are
more important than exterior aesthetics and ornament.
Social dilemma A situation in which group and individual
interests are in conflict, often applied to resource management
situations; individuals are rewarded more for self-interest
behavior than for group-interest behavior, but if most
individuals choose self-interest, the resource is extinguished.
Territoriality (human) A pattern of attitudes and behaviors
related to the perceived or actual control of a physical space,
object, or idea; acts as a mechanism for the peaceful
regulation of space use much more often than as a source of
aggression.
Environmental psychology is the study of transactions between
individuals and their built and natural settings. It assumes that
the actions of persons change the environment and that their
behavior, experience, and well-being are changed in turn by the
environment. Environmental psychology includes basic scien-
tific research aimed at understanding person-environment
transactions and practice that applies this knowledge to
improve human settings.
Background, Theory, and Method
As an organized discipline, environmental psychology origi-
nated in the late 1960s, but some social scientists had stu-
died person-environment transactions for decades before
that. Considering the enormous investment society makes
in the construction and maintenance of the physical envi-
ronment, including buildings, parks, and public outdoor
settings, and its massive use of natural resources, the long
delay before person-environment relations received ade-
quate attention seems odd. However, many studies have
dealt with the major topics within environmental psychology
since the late 1960s. Much of this work has been stimulated
by the recognition of environmental problems such as cli-
mate change, crowding, pollution, energy shortages, and
unsuitable buildings.
Theories in environmental psychology are diverse. Some
emphasize central psychological processes such as stimulation
and control. The common occurrence of too much or too little
stimulation is the focus of adaptation-level, arousal, overload,
underload, and stress theories, which predict that a wide range
of behaviors and experiences will be affected by one's level of
stimulation. Other theories emphasize the importance of the
individual's real or perceived control over the environment
(e.g., personal control, reactance, learned helplessness, and
boundary regulation theories). The ecological approach asserts
the importance of the behavior setting, a naturally occurring
small-scale social-physical unit consisting of regular patterns
of person-environment behavior (e.g., a household or a
sporting event). Integral or holistic approaches (e.g., interac-
tionism, transactionalism, and organismic theory) describe the
complete interrelationship of persons and setting. Finally,
the operant approach downplays abstract principles and pre-
fers a direct problem-solving approach that employs behavior
modification techniques to deal with problematic environ-
mental behaviors.
Environmental psychologists recognize that person-
environment transactions are influenced by many different
factors, which has led to multiple paradigms for studying
them. A strong preference for performing research in the every-
day world means that field studies are common. Sometimes,
laboratories and simulated settings are used, primarily when a
field study is not possible. True experiments occasionally are
conducted; they are desirable when a researcher seeks to iso-
late particular causes and effects. However, quasiexperimental
research designs are much more common.
Environmental Perception
Environmental psychologists prefer to study the perception of
whole, everyday scenes. In doing so, they sometimes must sacri-
fice a degree of experimental control, but in return they obtain
data on the perception of real, complex settings through which
the persons may move and feel a real connection. Perceivers
select certain cues from scenes and ignore many others.
54
Environmental Psychology 55
Unfortunately, some cues they ignore may be important, at least
in the long run (e.g., air pollution's effect on health). Environ-
mental perception has been studied using verbal reports, time-
sampling, behavioral inference, phenomenological methods,
and combinations of these.
The available evidence suggests that perceptions of qualities
such as length and distance are largely dependent on the
physical elements in the scene and how these elements are
arranged. However, personal factors such as perceptual ability,
finding a setting pleasing, culture (such as being raised in a
'carpentered' world), and training (e.g., in architecture) also
affect the very way a person sees the world.
The theories of Egon Brunswik, James Gibson, and Donald
Berlyne each have had major effects on the study of environmen-
tal perception. Each began as a traditional (nonenvironmental)
theory, but contained the necessary seed to be fruitful for envi-
ronmental psychology: an emphasis on properties of stimuli.
They have provoked considerable basic research into the nature
of environmental perception, but they have also been extended
into the practical domains of city planning, park planning, and
architecture.
Environmental Cognition
How do people think about and find their way around the space
around them? They do not acquire, store, and recall information
about locations, distances, and arrangements like cameras or
copy machines. Yet, our ways of doing so usually are effective
and rule-governed. Spatial cognition - the way we acquire, pro-
cess, store, and recall information about everyday settings - is
studied by examining sketch maps, model construction or
manipulation, distance estimation, and naturalistic observation.
It is affected by one's stage of life, familiarity or experience with
the setting, and several cognitive biases. Two common biases are
to envision places as more grid-like or Euclidean than they
actually are, and to mistakenly employ larger geographical enti-
ties in placing smaller ones (e.g., Reno is thought to be east of
Los Angeles because it is in Nevada, which is generally east
of California, but Reno is west of Los Angeles).
The study of cognitive maps reveals that legible places are
easier to comprehend: they have clear paths and distinct edges,
districts, nodes, and landmarks. The spatial cognition of chil-
dren generally develops in a sequence that progresses from
egocentric to projective to abstract. The spatial cognition of
older people often is limited by lowered mobility or sensory
abilities, and they may perform less well. However, their mem-
ories of the environment are more personalized and are, in
some respects, better than those of younger people.
Experience in a setting gives one a richer, better organized
cognitive image of it. Both landmarks and paths facilitate
the growth of place knowledge. Where one of these elements
is more common than the other, examples of it will be
learned first. Male-female differences in spatial cognition
exist, but may largely reflect the different travel experiences
of men and women. The environmental cognition of urban
forms is improved by clear paths and visible landmarks.
At the architectural level, buildings that are tall, free-standing,
distinctively shaped, and used often are better recalled than
others.
Environmental Appraisal
Environmental appraisals are personal, subjective judgments
of places; they may or may not be consistent with most other
people's assessments of the same places. Environmental apprai-
sals take six forms: description, evaluation, aesthetics, emo-
tional, meaning, and concern. First, environments may be
verbally described in many ways; a talented writer probably is
best able to describe any particular setting. However, researchers
have tried to develop standard sets of descriptors to assist the
average person to describe places.
Second, environmental appraisals take the form of evalua-
tions. These are influenced by such personal factors as age, sex,
and familiarity with the place and by such objective features
as room design, congruity, contrast, and complexity affect
evaluations.
Third, the aesthetics of environments may be appraised.
Environmental beauty is the concern of activists, planners,
experimentalists, and humanists. Landscape beauty has been
shown to be largely a function of the prevalence of certain
elements in a scene, such as bodies of water, types of vegeta-
tion, and mountains, and the apparent distance to and the
placement of these elements in the scene.
Fourth, emotional response to the environment is another
form of appraisal. Emotional responses to environments are
usually mild, persistent, and cumulative rather than sharp and
brief. Nevertheless, they are a complex mix of behavioral,
cognitive, and physiological responses.
A fifth form of appraisal is the meaning assigned by a person
to a place. Two kinds of this meaning are the building's com-
munication of an architectural or philosophical concept and
the communication of its purpose or function. Some buildings
communicate, in this sense, to observers, more than others.
Finally, people appraise places in terms of their concern (or
lack of it) for the place or the place's potential risk to them-
selves. This is a form of appraisal for which behavioral con-
sequences are of particular importance.
Environmental Assessment
Environmental assessments are measurements of a place's qua-
lities that aspire to objectivity; that is, they attempt to measure
the way an entire population segment, such as visitors to a
national park, experiences a place. Environmental assessments
may measure a setting's physical and spatial properties, arti-
facts and objects, traits, behavioral occurrences, or social cli-
mate. For example, do visitors to a park as a group find the park
'natural' or 'beautiful' or 'a good place to fish'? In contrast to
environmental appraisals, they are place-centered (rather than
person-centered), aim to measure physical properties (rather
than psychological properties), are more often policy-oriented
(rather than oriented to the understanding of perceivers), and
more often employ observers with an expert or frequent-user
relationship to the place being assessed.
Environmental assessments may use technical or human
means of observations. Each has its place; neither is necessarily
more reliable or valid; each employs a variety of instru-
ments (machines for technical environmental assessments, or
TEAs), questionnaires, or rating forms for human observers.
56 Environmental Psychology
Observer-based environmental assessments (OBEAs) have at
least five purposes. They allow for comparisons between TEAs
and OBEAs, assist in the development of physical measures of
environmental quality, provide data on environmental quality
trends from the human perspective, provide assessments of
quality along dimensions with particular human relevance,
and educate the staff of the assessed setting as to its strengths
and weaknesses.
Environmental Personality
The study of personality may seem antithetical to environmen-
tal psychology if one assumes that environmental psychology
emphasizes the effects of settings on persons. However, per-
sonality is part of environmental psychology.
Two early personality theorists are important for environ-
mental psychology: Henry Murray and Kurt Lewin. Murray's
personology introduced the concepts of alpha and beta press
(the actual and perceived power of the environment to affect
one's welfare), and internal and external proceedings (subjec-
tive and objective accounts of the initiation and completion of
a behavior sequence). Lewin's field theory, which conceptua-
lized persons as actively interacting with their environments in
their life spaces, produced the famous formula B = f(P, E).
One's representation of the physical environment and some
elements of the unrepresented physical environment (the for-
eign hull) affect one's behavior and experience. Lewin's action
research concept, in which theory and application are fused,
guides many environmental psychologists today.
Numerous modern systems for conceptualizing environmen-
tal personality have been described, but the most developed
is the Environmental Response Inventory, created by George
McKechnie. It assesses eight dispositions: pastoralism, urban-
ism, environmental adaptation, antiquarianism, stimulus-
seeking, environmental trust, need for privacy, and mechanical
orientation.
Some traditional personality measures also assist in the
understanding of behavior that has environmental signifi-
cance. For example, extraverts generally prefer smaller interper-
sonal distances and may experience less stress in higher density
situations. Persons with an internal locus of control also prefer
smaller interpersonal distances and tolerate high density better
than those with an external locus of control.
Personal Space
Personal space refers to the interpersonal distance chosen dur-
ing social interaction. It varies with and reflects the rise, current
status, and decline of relationships. Alpha personal space is the
objective distance between interacting individuals and beta
personal space refers to the experience of that distance. Edward
Hall described four zones of increasing distance that depend
on the closeness of the relation between the two persons.
Personal space may be predicted in part from knowing an
individual's characteristics. Males typically use larger distances
than females. Young adults typically use more personal space
than children. Interpersonally warm and nonanxious indivi-
duals probably have smaller personal spaces than others.
Psychological disturbance often leads to more variable or inap-
propriate personal space. These personal influences on per-
sonal space also have interactive effects: to predict a person's
personal space choices accurately, one must know the person's
age, sex, race, culture, and personality.
Social and physical features of the situation also affect
interpersonal distance. Attraction, cooperation, and equal sta-
tus are associated with smaller personal space; stigma and
unequal status lead to larger personal space. People seem to
prefer more space when the physical environment offers less
room. These factors combine to produce interpersonal dis-
tances that are different from what would be expected from
the consideration of each influence by itself.
Culture is a major modifier of interpersonal distance. Some
evidence exists for a continuum that ranges from the closeness
of Arabs and Latin Americans to the distance of English and
Germans. But differences within these groups, occasional
results that do not fit this neat pattern, and other factors such
as the language one speaks during a particular interaction, all
affect personal space.
The individual's acquisition of the rules concerning inter
personal distance are probably well characterized by social
learning theory. The functions of personal space are described
by numerous minimodels based on concepts such as comfort,
protection, communication, stress, optimal stimulation, and
affiliative equilibrium. The social penetration model emphasizes
the role of personal space in the rise and fall of relationships. The
limits-of-compensation model highlights the outer limits of
interpersonal distance, showing that as discomfort grows, inter-
action becomes more difficult. The approach-avoidance model
hypothesizes differences between the desire to draw close to
someone and the desire to keep one's distance.
Human Territoriality
Territoriality is a pattern of behavior and attitudes held by an
individual or group that is based on perceived, attempted, or
actual control of a definable physical space, object, or idea.
It may involve habitual occupation, defense, personalization,
and marking of that space. Environmental psychologists recog-
nize seven types of territory: primary, secondary, public, inter-
actional, body, object, and idea. The seven types of territory
may be infringed upon in at least three ways: invasion, viola-
tion, and contamination. In turn, territory-holders may utilize
preventative, reaction, or social boundary defenses.
Personal factors, physical and social aspects of situations,
and culture can lead to territoriality. For example, males gener-
ally manifest more territoriality than females. Defensible space
theory argues that physical arrangements increase territoriality
feelings and behavior and that this increase leads to a decline
in territorial invasions. These physical arrangements may be at
the block or neighborhood level (e.g., altering traffic flow) or at
the house level (e.g., fences and plantings).
Territoriality appears to increase with three social factors:
ownership compared to renting, positive social climate, and
greater competition for resources. Cultures differ in their
expression of territoriality, although the question whether
some cultures are more territorial than others has not been
clearly answered. Territoriality is associated with a variety of
Environmental Psychology 57
behaviors and experiences: personalization, aggression, domi-
nance, control, and attachment. Personalization and marking
are very common, may occur with or without awareness, signal
ownership but do not always lead to active defense, and seem
to offer psychological benefits to the territory-holder beyond
merely informing the world of a territorial claim. Popular
writers have exaggerated the relation between overt aggression
and territoriality in humans. Territoriality does facilitate con-
trol of spaces, but this is usually accomplished through passive
means that do not involve overt bullying of others.
Place attachment is the experience of belonging to a setting.
Place attachment is an important process involving some of the
closest person-environment bonds that humans experience.
Theories of human territoriality remain diverse, speculative,
and largely untested. Ethology, organizational behavior, behav-
ior settings, brain structure, and conflict resolution are some of
the disparate concepts that underpin theories of human territo-
riality. The main function of human territoriality probably is the
maintenance of order and identity in everyday life.
Crowding and Population Density
Crowding is the unpleasant experience of spatial restriction,
whereas density is a physical ratio of persons per unit area.
Crowding and density are not always correlated with one
another. The notions of social versus spatial density, indoor
versus outdoor density, and proximity have been advanced as
ways to refine the density ratio into a variable with more
predictive power.
At the personal level, personality, expectations, and prefer-
ence influence whether an individual experiences a given situ-
ation as crowded or not. Internal locus of control, high
affiliative tendency, a tendency to screen unwanted stimuli,
a preference for high density, and an expectation for low den-
sity upon entering a place predispose individuals to experience
less crowding when density is high.
Social influences can worsen or ease crowding. Sheer num-
bers of others will sometimes, but not always, produce more
crowding. If others are watching or touching an individual
(particularly a male), or one is engaged in disliked activities,
crowding is worse even when density is equivalent. Being left
out produces more crowding stress than sharing space with
another person who is compatible. Objective, accurate infor-
mation about high-density settings reduces crowding.
Crowding is also affected by the physical surroundings. It
is likely to be worse when density is higher, the building is
higher, the corridors are longer, the ceiling is lower, the walls
darker, and when sunlight rarely finds its way in.
High indoor density usually leads to physiological and psy-
chological stress, at least for those who prefer larger interpersonal
distances or are socially isolated. Performance may be harmed
under high density when the task is complex, when others are
watching, and when performers must physically interact to
accomplish a task. When a person's expectations about density
are unfulfilled, performance may suffer: performance may even
be worse in low density if a person expected high density.
High density often has negative effects, but under some
conditions - perhaps when individuals perform activities
in which they are already competent - it may improve
performance, through social facilitation. Inescapable high
inside density harms a wide range of social behaviors, particu-
larly for men. Except in prison, those effects take more passive
forms (e.g., lack of helpfulness) than active forms (e.g., aggres-
sion). High density may enhance humor, but it almost always
creates negative emotion. High density reduces one's sense of
control; as a result, a variety of coping mechanisms are used.
Privacy
A major goal of human social behavior is to regulate the amount
of contact one has with others to suit one's current need for
interaction, which may be high or low. Privacy essentially is
about the regulation of access to oneself. This access may be
interactional, as in control over who is allowed to be with
whom, or informational, as in who is allowed access to archival
information about whom. Privacy has four faces: solitude, inti-
macy, anonymity, and reserve. Conceptually, privacy may be
considered as a preference, an expectation, a value, or a behavior.
Privacy is influenced by personal characteristics. For exam-
ple, city-raised individuals prefer more anonymity and intimacy.
Women seem to manage privacy in close groups by discussing
more interpersonal and intimate matters and men often achieve
privacy by removing themselves from the setting and talking
less about delicate matters. Individuals who have greater-than-
average privacy needs are less sure of themselves and more
anxious. Members of different cultures are believed to have
similar privacy needs but, depending on how the environment
supports those needs, may fulfill those needs in different ways.
The physical setting also has important effects on privacy
preferences, expectations, and satisfaction. For example, open
space at work often breeds dissatisfaction, whereas at home it is
usually preferred. Its physical influences are both immediate,
such as the doors and partitions surrounding people, and less
immediate, such as the distance between houses and the num-
ber of neighbors visible from one's house. Privacy attitudes and
behaviors vary from setting to setting.
Those who know how to use the environment to success-
fully regulate their privacy may be more successful in areas of
life beyond the social realm. Some individuals adapt to or
accept situations that would seem to offer too much or too
little social contact. Others faced with a shortage of traditional
mechanisms of regulating social interaction develop creative
solutions to the problem. As a key process in our dealings with
others, privacy is inextricably linked to communication, auton-
omy, identity, emotional release, and growing up.
Irwin Altman's model of privacy as an optimizing
boundary-control process dominates, but other theorists have
contributed ideas. John Archea, for example, reminded us that
privacy has a sUong environmental component. Maxine Wolfe
and her colleagues note that privacy preferences, expectations,
and satisfaction vary over the life cycle. Eric Sundsttom observes
that privacy needs vary as a function of status.
Home and Community Settings
'Home' is a rich personal experience that may develop in a
residence, which is the physical setting. Residential satisfaction
58 Environmental Psychology
is a function of person and residence characteristics. The more
congruent the residence is with one's stage in the life cycle and
role in the family, the more satisfied one is likely to be.
Moving house is one outcome of residential dissatisfaction;
it is a source of stress in itself. Individuals who have little
residential choice or do not enjoy new settings experience even
more mobility stress. Privacy and crowding are the biggest
problems, primarily as they harm child development. High-
rises seem inappropriate for children.
Resident lifestyle is reflected in the physical organization of
the home. Home is an important place to experience the growth
and relaxation that leisure can provide. Neighborhood satisfac-
tion is related to such physical factors as the availability of nearby
green space, general upkeep, and noise. Yet, these factors interact
to some extent with personal characteristics (e.g., screening ten-
dency, perception of control, and one's residence type). Neigh-
borhood social ties may be less important than they used to be,
except for a few determined groups such as religious enclaves.
Cities are stimulating, but their impact does not constitute
an overload for everyone. Individuals apparently adapt to
some stressors (e.g., air pollution) more than others (e.g.,
noise), in the sense of no longer noticing them, although
their health may suffer in the long term. Vandalism and street
art are different in motive and expression: vandals destroy
settings out of a sense of revenge, whereas artists enhance
settings while expressing social concerns. Crime and fear of
crime are also important urban problems that may be eased
in part through the application of defensible space design
principles.
Aggression increases with air pollution and with tempera-
ture. Sunshine combined with moderate temperatures seems
to increase helping. Noise in the community reduces helping,
unless the other's need is serious. Interaction among strangers
on the street generally follows a minimax principle: minimize
interaction, maximize order. Frequent exposure to the same
strangers, however, leads to a kind of distant affection.
Environmental factors in shopping include store location
and layout, the display of goods, ambient music and noise,
lighting, and shopper density. Research in the retail environ-
ment has been slow to emerge. This may be because its com-
mercial value means that researchers report their findings to
corporate sponsors rather than to public-domain journals.
Most homes are threatened by one or more environmental
hazards, some natural and others technological. In the
precalamity phase, governments attempt to reduce risk through
engineering and education, but when neither of these is effec-
tive, many lives are lost. Specific warnings and awareness-
raising simulations can help, but effective personal action is
often blunted by the human tendency to bounded (i.e., lim-
ited) rationality: underpreparation for impending disasters,
hasty, sometimes inappropriate responses to hazard mitigation
once the disaster occurs, and misperception of hazard risks.
During most disasters, individuals generally interact in rela-
tively rational and cooperative ways; the exception is a disaster
that traps people in a confined setting. The outcome of disaster
is stress in many forms. Some forms of coping reduce some
forms of stress better than others. Community attachment and
economic factors often prevent residents from vacating high-risk
areas; they would rather see governments engage in massive
public works to protect their homes than leave.
Work and Travel Settings
Work begins with getting to it. Most research on this has been
concerned with encouraging commuters to help save energy.
Environmental psychologists have created demographic pro-
files of car and urban transit riders, devised models of com-
muter preference, provided positive information about urban
transit, evaluated existing transit systems, and studied the
effect of reducing fares. This research has not yet had a very
large effect on society's automobile addiction, but the more
promising approaches are being sorted out from the less
promising ones.
Once at work, physical influences play many roles. Noise has
many effects. In industrial settings, it can cause serious hearing
loss. Loud noise is particularly dangerous when employees do
not realize that deafness comes slowly and almost impercepti-
bly. Despite conventional wisdom that noise interferes with
performance, research in natural settings shows how complex
the issue is and that performance decrements depend on the
task, the person, and the type of noise. Noise does harm perfor-
mance under certain combinations of these, but not under
others. For certain tasks, noise may even arouse a person enough
to improve performance.
Noise is a serious problem in open-plan offices. Employees
find sound a problem both coming and going: sound entering
their workspace is annoying and when their own words escape
over partitions too easily, their privacy is compromised. Office
noise may even affect important interpersonal behavior, from
mere impressions of others to important judgments regarding
them. Music may improve performance of well-learned cogni-
tively simple tasks. Employees naturally dislike noise but, for
the most part, they like music on the job.
The effects of temperature are also usually moderated by
access to heavier or lighter clothing. The amazing variety of
temperature effects reported are due to these measurement and
clothing factors, as well as many others, including the degree of
acclimatization, a knowledge of coping strategies, motivation,
and the type of work. Engineers have described precise 'comfort
envelopes,' but environmental psychologists have discovered
that comfort depends on perception as well as actual effective
temperature and that optimal performance may be found out-
side the comfort envelope. Stress occurs when individuals are
initially subjected to effective temperatures farther outside the
comfort envelope, but many persons can adapt to these more
extreme temperatures after a long- term exposure to them.
Lighting affects work behavior primarily when it is insuffi-
cient (leading to low productivity, accidents and eyestrain) or
improperly placed (leading to glare and eyestrain). Lighting
often is excessive in North American offices and many indivi-
duals dislike the newer forms of lighting, some of which distort
color. European lighting standards call for lower levels of
illumination. Carefully placed local lighting could resolve
some of these problems. Access to natural light and views is
psychologically important.
Employees are very sensitive to space and unhappy with
many existing arrangements. Many open-plan arrangements
reduce desirable communication and increase undesirable
communication. Office arrangements lead visitors to form
impressions of the office-holder's character and status.
Environmental Psychology 59
Yet, many organizations restrict the degree to which employees
may arrange or personalize their offices and fail to consult
with employees when offices are planned.
The environmental psychology of travel is a new but grow-
ing area. Travelers affect destinations and are, in turn, affected
by them. Anticipation, travel itself, and the recollection of
travel involve environmental perception and cognition. Recre-
ational travel is a trade-off between familiarity and risk, and
between economic growth and environmental degradation,
but as society is able to provide employees with more dispos-
able incomes and time, it is a trade-off many are pursuing.
Some destinations bring relief from anxiety, others throw
travelers into 'environment shock.' Travelers ruin some physi-
cal settings and enhance almost none. 'Romantic' tourism (i.e.,
the search for more and more unspoiled destinations) may be
an undesirable luxury. A more careful planning of destination
sites might spread the impact of visitors, offer more authentic
experiences, and educate travelers while offering them solace
from the working world.
Learning Environments
Both macro (e.g., school and campus design) and micro (e.g.,
desk and room design) physical features of learning settings
affect learning and related activities. Students in large schools
have a slight advantage in the variety of topics they can learn
about but, partly because time at school is limited, students
in large schools do not actually participate in more activities
than students in small schools. Students in large schools
more often learn and enjoy as spectators; students in small
schools more often learn and enjoy as participants. In most
areas of learning, students in small schools achieve more as a
result of developing competence through direct involvement
in activities.
If satisfied teachers are better teachers, then sensitive con-
struction or renovations of their workplace (the classroom)
should be undertaken when necessary. Students often perform
better in the room where they learned the material. Noise
interferes with learning both while it occurs and, if the
learner is subjected to noise for long periods, even after
the noise is gone. Noisy classrooms may impair the perfor-
mance of girls more than boys, that of autistic children more
than hyperactive children, and that of most children when
the task is difficult. Noise may hinder performance by
interfering with information processing, lowering the student's
perception of control, and increasing blood pressure. To com-
bat noise, instructors have changed their methods - sometimes
sacrificing an otherwise good method for a quiet one - and
successfully employed behavior modification techniques such
as sound-activated electrical relays that control reinforcers
such as radio music and extra recess time.
Incandescent lighting is preferred by many, but it is more
expensive than fluorescent lighting. Fluorescent lighting has
not been shown to have dramatic negative effects on the per-
formance or health of most students, despite some claims to
the contrary. As with noise, light's effects may be on specific
subgroups of individuals; when studies of whole classes or
schools are done, large effects on a few learners may be
obscured by an absence of effects on most learners.
A few simple, direct relations exist between indoor climate
and educational behavior; some combination of person and
setting variables may enhance or reduce the influence of cli-
mate. Perhaps the best-supported conclusion is that student
performance is best in slightly cool but not humid classrooms.
Preferred environmental conditions may not be the same as
conditions that facilitate performance.
The amount and arrangement of space in educational set-
tings is important for classroom performance and behavior.
High density may affect learning when the activity involves
physical movement around the classroom, when learning is
dependent on some classroom resource that is not increasing
as fast as the number of learners, when a particular situation
appears crowded to a particular learner, and when the concept
to be learned is complex. Among preschoolers, high density
alters the child's choice of activities. Educational benefits
appear to accrue in classrooms that have the teacher's desk in
a corner, carefully arrange and separate different kinds of activ-
ities, possess library corners, and are carpeted. Action zone
seating (the front and central area of the room) benefits some
students. All these findings depend in part on grade level and
teaching style. Space in classrooms affects student and teacher
feelings. Most students and teachers prefer lower density class-
rooms, because lower densities usually feel less crowded.
Increased social density leads to increased aggression and
withdrawal when other resources, architectural features, and
teaching style do not counteract it. Classroom arrangements
should provide optimal stimulation. The proper amount in
any given situation cannot yet be specified, but the student's
need for stimulation, the type of activity, and the length of time
spent in the activity must be considered.
Environmental competence involves learning about the
environment. Three varieties of it include (a) one's personal
style, attitudes, and awareness of physical settings, (b) knowl-
edge of physical settings, including technical knowledge, how
to unearth new information, knowledge about how social
systems control space, knowledge of person-environment rela-
tions, and (c) practical environmental skills such as scouting,
matching, personalization, and creative custodianship. Pro-
grams in and out of school teach many different facets of
environmental competence, from basic environmental ethics
to campfire starting to architectural design.
Architectural Psychology
One main purpose of environmental psychology is to use the
foregoing knowledge to create more humane buildings. Some
buildings are human disasters while others are merely persis-
tent nuisances to those who occupy or use them. As described
by Robert Sommer, social design is a way of creating buildings
that fit occupants and users better by involving them in the
planning process. It is needed in societies that have splintered
the building creation process into many specialist roles. Social
design is a remedy for the malady in which architects see
themselves primarily as artists, ignoring the basic needs and
activities of occupants. This malady is now widely recognized,
yet many buildings are still constructed or renovated without
significant user involvement. Social design aims to match set-
tings to their occupants, to satisfy a variety of principal player
60 Environmental Psychology
needs, to promote personal control in the building, and to
encourage social support.
The design process includes programming, design, con-
struction, use and adaptation, and postoccupancy evaluation.
Programming consists of three phases: understanding the
needs of users, involving them in the possibilities of design,
and translating their needs into design guidelines, goals the
actual design should achieve.
Turning these guidelines into plans and reality is the job
of architects and construction companies. Later, the environ-
mental psychologist returns to monitor user behavior and
adaptation of the new building. Postoccupancy evaluation
examines the effectiveness of the program and design, using a
variety of social science tools. Together with programming,
postoccupancy evaluation is perhaps the major activity of prac-
ticing environmental psychologists.
Resource Management
A major application of environmental psychology is to improve
the management of nature and its many essential but limited
resources. Each person on the planet manages a steady supply of
natural resources that have been converted into products. Some
of these resources come from limited sources called commons,
that is, a pool of desirable materials that may be harvested by a
number of individuals or organizations that share access to it.
Commons dilemmas occur when harvesting of natural resources
may proceed faster than the resource can regenerate; individuals
must decide whether to maximize their own gain or the gain of
the group, including themselves. The idea may sound abstract,
but it applies to so many everyday situations (e.g., recycling,
littering, energy conservation, endangered species, lumbering,
fishing, pollution, etc.) that it is useful as a powerful general
concept.
The quality of resource management depends on the
resource itself, characteristics of the managers, and the rules
governing the harvest. Cooperation seems to decline as the
importance of the resource increases and to improve as
the resource is depleted - two ominous signs.
The results of many studies suggest that when a commons
dilemma exists, part of the resource should be allocated to each
participant to manage, that friendship and trust should be
encouraged among the harvesters, that the number of harvest-
ers should be kept small, and that harvesters should commu-
nicate, make public choices, and be subject to punishment for
selfish decisions, although positive incentives for cooperation
are even more effective. Further, harvesters should be given
experience operating the commons, be required to make mini-
mal contributions where they are suitable (e.g., in commons
dilemmas where the harvesters must donate to keep the
resource alive), and have the altruism in the commons pointed
out to them. Last but certainly not the least, payoffs for coop-
eration should be increased as much as possible.
Several theories of social dilemmas have been described.
Limited processing theory postulates that most individuals act
selfishly, not because they are evil but because the dangers of
defection simply do not occur to them. Tragic choice theory
maintains that self-interest cannot be overcome except by the
creation of a very strong central authority; this is unpleasant
but necessary for the survival of the commons. Social trap
theory applies a reinforcement perspective to social dilem-
mas, suggesting that better management of the commons
would follow from a restructuring of reinforcements. Equity
theory emphasizes the importance of perceived equity in
resource allocation. Cooperation may increase if participants
believe others are receiving goods in proportion to the effort
invested, no more and no less. But this can still lead to the
extinction of a commons: if everyone works hard and har-
vests equally hard, depletion will occur, so equity alone is
not enough.
Two social dilemmas are pollution and energy use. Pollution
is a 'give-some' form of social dilemma: individuals or organiza-
tions profit by using resources, but degrade the commons with
the by-products. Energy use is a social dilemma when it is easier
to waste energy, particularly from nonrenewable sources, in the
service of a short-term goal than to conserve it for the future.
Important savings could be achieved in ideal conditions, which
would include an integration of psychological, economic, and
engineering contributions. Conservers include those with
enough money to purchase energy-saving devices so they can
save over the long haul, those who must conserve because they
cannot afford much gasoline, air conditioning, or fuel oil, those
who have adopted a voluntary simplicity lifestyle and feel a
personal responsibility for energy problems, those who are
willing to make moderate public commitments toward energy
conservation, and those whose energy use is monitored and
frequently fed back to them.
Conservation may be enhanced by some appeals from
authorities. Educational campaigns seem to change attitudes,
priming individuals for changing behavior, but do little to
change behavior by themselves. Feedback and goal-setting
approaches have achieved significant savings. A housing devel-
opment that integrated several design innovations with a
group of ecologically oriented residents will approach the
upper limit of 50% energy savings.
See also: Crowding: Effects on Health and Behavior; Environmental
Cognition; Evolutionary Social Psychology.
Further Reading
Bechtel R and Churchman A (eds.) (2002) Handbook of Environmental Psychology,
2nd edn. New York: Wiley.
Bell PA, Greene TC, Fisher JD, and Baum A (2001) Environmental Psychology, 5th edn.
Belmont, CA: Wadsworth.
Gifford R (2007) Environmental Psychology: Principles and Practice, 4th edn. Colville,
WA: Optimal Books.
Sommer R (1983) Social Design. Englewood Cliffs, NJ: Prentice-Hall.
Stokols D and Altman I (eds.) (1987) Handbook of Environmental Psychology.
New York: Wiley.
Episodic and Semantic Systems of Autobiographical Memory
D R Beike and J W Ransom, University of Arkansas, Fayetteville, AR, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Coherence The motivation for specific
autobiographical memories to fit with abstract
beliefs about the self.
Conjunction errors Memory errors in which one or a few
details have mistakenly been confused with details from the
memory of a different incident.
Correspondence The motivation to recollect accurate
autobiographical memories that actually reflect the real state
of the world.
False memory A memory that feels veridical but represents
an incident that the rememberer did not in fact personally
experience.
Flashbulb memory When an incident that is unexpected
and highly emotional produces a memory that the
rememberer feels is unusually detailed and accurate despite
the passage of time.
General (or semantic) autobiographical memory A
memory for either a personally experienced event that
extended over several days or longer, or of a number of
similar experiences summarized.
Flyperthymesia A memory syndrome characterized by
extraordinary ability to recall specific incidents from one's
personal experience and a tendency to spend large amounts
of time thinking about such experiences.
Involuntary (intrusive) memories Specific life event
memories that come to mind without volition; if their
occurrence is bothersome, they are called intrusive memories.
Overgeneral memory A tendency to recall general multiday
memories in response to a cue or request to produce a
specific memory lasting 1 day or less.
Specific (or episodic) autobiographical memory A
memory for a discrete, personally experienced event that
lasted 1 day or less.
Specific autobiographical memories are memories of discrete life
experiences that occurred in a specific time and place, usually
lasting 1 day or less. Recollection of autobiographical memory
involves a sense of reliving the experience, known as mental
time travel. Autobiographical memory is used to establish a
sense of personal continuity and identity, to form social bonds,
to solve problems, and as the basis for projections of one's
future. General knowledge of the self and acts of remembering
are inextricably intertwined. This article considers both the
structure of autobiographical memory as well as the processes
involved in recalling them. In the first section, two theoretical
frameworks for autobiographical memory structure are dis-
cussed. Second, basic memory phenomena which occur within
this underlying memory structure are discussed. Third, the
manner in which general knowledge about the self is main-
tained and updated is discussed. And finally, the failures of
both general self-knowledge and specific autobiographical
memories to accurately depict reality are discussed.
Episodic Versus Semantic Autobiographical Memories
Autobiographical memories and self-knowledge are two types
of remembered information about the self. Most theories of
autobiographical memory make reference to both of these
types of self-relevant memory; that is, not only the memory
for specific discrete episodes, but also larger units of self-
knowledge. Two popular theories for the structural organiza-
tion of broad self-knowledge and discrete autobiographical
memories are addressed here.
The first theoretical approach, proposed by Endel Tulving,
makes a clear distinction between semantic and episodic mem-
ories. General self-knowledge (e.g., being an introvert) is
abstracted from numerous specific memories to form a seman-
tic memory about oneself. This semantic self-knowledge is
represented in much the same way as semantic knowledge of
the Civil War, which of course we never personally experi-
enced. The recollection of semantic self-knowledge is accom-
panied by the experience of noetic awareness, a sense of
knowing the information. Specific autobiographical memory,
on the other hand, is represented with a high degree of detail as
an episodic memory. The recollection of episodic memories is
accompanied by the experience of autonoetic awareness, a
sense of remembering in the sense of mentally reenacting the
experience.
However, semantic self-knowledge might be stored in an
independent semantic system or could be 'computed' from
episodic memories during the recall process. Although this
has been a controversial topic within certain scientific circles,
research evidence has accumulated in support of the former
option, showing that semantic and episodic self-knowledge are
independent systems. Evidence for two independent memory
systems has been provided in two veins of research: priming
studies and case studies of amnesics. Klein and Loftus have
demonstrated that the semantic system can be primed without
affecting the episodic system and vice versa. Additionally, their
research with a particular temporarily amnesic patient demon-
strated that complete loss of episodic memory functioning
does not impair memory for semantic self-knowledge.
In fact, this theory of distinct semantic and episodic systems
is consistent with findings from multiple case studies in which
patients with severe amnesia cannot remember detailed events
at all, yet remain aware of general self-knowledge. Case studies
with humans experiencing hippocampal damage, as well as
studies with hippocampal lesions in rats, have demonstrated
that the hippocampus plays a necessary role in memory.
61
62 Episodic and Semantic Systems of Autobiographical Memory
Although memories are not actually stored in the hippocam-
pus, it provides a sort of road map of other regions in the
neocortex that must be activated to reexperience an event.
Therefore, the hippocampus activates other regions to provide
retrieval. Some researchers have suggested that semantic mem-
ory can be intact following hippocampal lesions because
much of the information 'mapped' in the hippocampus has
been overlearned and therefore represented in other brain
regions surrounding the hippocampus. In short, semantic
self-knowledge may be abstracted from episodic memories in
the hippocampus.
Despite being distinct, these systems are believed to
interact in normal memory functioning. When unique events
are remembered they are stored in the episodic memory
system, but as these unique event memories are joined
by similar memories they may be abstracted into semantic
self-knowledge. Therefore, when trying to recall an instance
of behavior consistent with the self-concept, participants will
often recall a broad semantic memory (e.g., 'I was always
introverted in high school'). Conversely, when recalling an
instance of behavior inconsistent with the self-concept, parti-
cipants will often recall a detailed episodic memory (e.g.,
'I remember going to a party on Halloween and being more
extroverted than normal'). In theory, these separate yet inter-
acting memory systems have evolved to allow fast, efficient
recall of traits while also allowing this trait knowledge to be
modified as new events demonstrate a consistent theme.
The second popular theory of autobiographical memory
called the self-memory system is composed of the working self
and the autobiographical memory knowledge base. The working
self is a temporary activation of current goals which constrain
the search for elements to be bound up in the memory.
For example, a person trying to recollect his final exam in
chemistry to help him study for an upcoming exam may be
likely to include behavioral elements about his preparation
before the exam. But a person trying to recollect the same
memory to help him remember where he left his pen may
be likely to include spatial elements about the location of the
room. The self-memory system acts to construct memories
consistent with current goals and the current sense of self,
and to inhibit memories that are inconsistent.
These autobiographical memories are constructed by draw-
ing from episodic ( event-specific knowledge) and semantic ( long-
term self) components. Event-specific knowledge contains
records of the microdetails of everyday experiences, which are
quickly lost and become unorganized without ties to the long-
term self. The long-term self is composed of a conceptual self,
abstract beliefs about and traits of the self, and the autobio-
graphical knowledge base, consisting of lifetime periods
and general events. Unlike Klein and Loftus' distinction, how-
ever, the self-memory system posits a hierarchical arrangement
of these episodic and semantic components, with lifetime
periods (such as 'school years') at the most general level, gen-
eral events (such as 'taking exams') at the intermediate level,
and specific memories (such as 'final exam in chemistry 10 f)
at the most specific level. Accordingly, the hierarchical arrange-
ment must be navigated every time a specific autobiographical
memory is recalled.
Based on the self-memory system theory, researchers have
distinguished specific autobiographical memories, defined
as an event which occurred in a single day, from general
autobiographical memories, which include events lasting
longer than a day and categories of repeated types of events.
Specific autobiographical memories are the result of success-
ful generative retrieval, while general autobiographical mem-
ories are the result of a failed hierarchical progression from
general memory traces to specific memories. For example,
when asked to recollect a specific life experience involving
the words 'ice cream,' a person may respond with 'I used to
eat ice cream a lot after soccer practice' rather than the more
specific 'I ate an ice cream cone during the fireworks display
last fourth of July.'
The tendency to fail in attempts to recollect specific memories
is known as overgeneral memory, and it is found in clinical
populations (including those with major depressive disorder,
posttraumatic stress disorder, and Parkinson's disease). How-
ever, some research has applied the same definition of specific
and general autobiographical memory to the nonclinical popu-
lation. Among other factors, a lack of executive resources has
been shown to impede the hierarchical progression from general
lifetime periods to specific autobiographical memories. In the-
ory, people experiencing overgeneral memory are cognitively
unable to move beyond the initial stages of autobiographical
memory construction and therefore fail to proceed from the
working self and abstract self-knowledge about lifetime periods
to the specific details that will be bound to form a memory.
Therefore, one theory considers general self-knowledge to
be stored independent of specific autobiographical memories,
while the other considers general self-knowledge to be the first
step in ultimately recalling a specific autobiographical mem-
ory. Despite the differences in terminology and research
method, the need to distinguish specific autobiographical
memories from some more general representations is a com-
mon thread among many theories. The conditions under
which and the ways in which these two systems interact are
explored. However, the basic features of remembering specific
autobiographical experiences are first delineated.
Basic Phenomena in Autobiographical
Remembering
Process of Construction
Like all forms of long-term memory, autobiographical memory
is reconstructive rather than reproductive. The rememberer
constructs a memory anew each time rather than reproducing
a static representation. In the case of autobiographical mem-
ory, the process of construction involves a large number of
brain areas and can extend up to 20 s in time. Most instances
of autobiographical remembering begin with a voluntary
search based on a cue or probe (e.g., 'When was the first
time you saw your current favorite movie?'). People often
begin by identifying a time period in their life during which
the sought-for memory may have taken place (lifetime period,
such as 'high school'), then they find generic examples of
similar memories (general event memory, such as 'going to
the cinema with friends'), and finally identify specific episodic
memory traces that contain the elements of the desired
memory (specific event memory, such as 'seeing Titanic with
Belinda and crying at the end').
Episodic and Semantic Systems of Autobiographical Memory 63
Indeed, studies using a think-aloud protocol demonstrate
that people usually begin their search for a specific autobio-
graphical memory by identifying a lifetime period or a general
event type first. Brain imaging studies also verify the involve-
ment of different regions of activation over time. The construc-
tion process begins with what appears to be the working
self s activation in the left frontal lobe. The working self gen-
erates retrieval cues, then begins to activate relevant elements
of the memory in the posterior temporal lobes and occipital
lobes, which contain the specific features of relevant episodic
memories. The elements of the autobiographical memory can
include sensory (especially visual), spatial, emotional, linguis-
tic, and narrative components. The initially activated areas in
the frontal lobes and the activated sensory areas in more pos-
terior portions of the brain are then bound together by the
medial temporal (hippocampal) area. This precise pattern of
neural activation is marked by the subjective experience of
autobiographical recall.
Moreover, the search for elements to bind into a memory
often begins by identifying an abstractly represented lifetime
period in which to search. Frequent rehearsal of semantic and
episodic memories is used to maintain coherence of the long-
term self. Furthermore, the constant retelling of stories about
the self is often biased and can lead to memory errors. Thus,
both temporary and more permanent notions of the self guide
autobiographical memory construction.
Amnesia and Hyperthymesia
Although virtually everyone develops the ability to recollect life
experiences, there are extreme cases of those who can recollect
almost nothing (amnesics) and those who can recollect almost
everything (hyperthymesics) of their life experiences. The noted
patient HM (Henry Molaison), for example, was unable to
record any new memories of life experiences after the surgery
that removed his hippocampi as well as his amygdalae and parts
of the surrounding temporal cortex. Moreover, Henry Molaison
was unable to recollect any autobiographical memories from
several years before the surgery. Clive Wearing, who suffered
hippocampal damage due to a viral infection, has similarly
been unable to record memories of life experiences since his
illness, and has few or no autobiographical memories from
before his illness. Yet both Henry Molaison and Clive Wearing
retained their knowledge, intelligence, working memory ability,
and implicit memory skills, demonstrating the separation of
autobiographical memory from other forms of memory.
At the opposite end of the spectrum are those whose
lives are overfull of autobiographical memory. Jill Price
contacted memory researchers for help understanding her
inability to forget her life experiences and her obsession with
remembering past events. Extensive testing revealed Price's
virtually perfect autobiographical memory, but fallible mem-
ory for other types of information. Price's autobiographical
memories come to mind unbidden and dominate her inter-
nal mental world. A handful of others with virtually perfect
autobiographical memory have been identified, including
Brad Williams, who can also provide detailed memories of
his experiences for any given date. The two qualities of supe-
rior recall of autobiographical events and a great deal of
time spent thinking about them define hyperthymestic syndrome.
Like amnesics, hyperthymesics show a disjunction in their
ability to remember individual life events as compared to other
forms of memory; they usually have average or even below-
average working memory, memory for word lists, and so on.
Moreover, there is some evidence of larger-than-average struc-
tures in the prefrontal cortex of hyperthymesics. Episodic
autobiographical memory therefore seems to be distinct from
other memory systems.
Involuntary Autobiographical Memories
In addition to the deliberate hierarchical process of autobio-
graphical memory construction described earlier, there is also
an unintentional and automatic process of autobiographical
remembering. These involuntary memories tend to be more spe-
cific and more highly emotional than voluntarily retrieved
memories. The best-known type of involuntary memories con-
sists of flashbacks or intrusive memories, which occur in people
who have experienced a traumatic event. Although bothersome
intrusive memories are associated with psychopathology, most
involuntary autobiographical memories are benign. They
occur several times per day in the general population of adults
and are twice as likely to be positive as negative in content. For
example, a glance at a pond on the side of the road while
driving to work might trigger an involuntary memory of swim-
ming in a pond as a child. Such involuntary memories are
especially likely during episodes of mind wandering.
Lifespan Distribution of Autobiographical Memories
The complex process of autobiographical remembering devel-
ops rather late compared to other types of memory. Infants
and toddlers remember facts and procedures, but autobio-
graphical memory ability has a later onset around age 3 or 4.
The inability to remember life experiences before this age is
called childhood amnesia. Adults do not explicitly remember life
experiences that happened before age 3 or 4, even if those experi-
ences were explicitly remembered at that time in childhood.
Autobiographical memory assumes its full adult-like form
much later in adolescence, when a more complete life narrative
develops. A life narrative includes themes and key scenes that
express concerns central to a person's sense of self and there-
fore provides a sense of a coherent and unified life. The adoles-
cent period of life during which the life story develops retains a
psychological importance throughout adult life, as revealed in
the reminiscence bump. For people of middle adulthood age or
older, the retention curve for autobiographical memory is a
nonmonotonic function, with no memories until about age 3,
a small number of memories of the childhood years, a large
number of memories from the adolescent years (the reminis-
cence bump), a moderate number of memories from the rest of
adulthood, and the greatest number of memories from the
most recent year of life.
Maintaining and Updating Semantic Self-Knowledge
We may gain self-knowledge by reflecting on the meaning
of behaviors and experiences stored in memory. People
indeed report using autobiographical memories in this way
64 Episodic and Semantic Systems of Autobiographical Memory
to establish and maintain identity. Yet general autobio-
graphical memories are more directly linked to the self than
are specific autobiographical memories. For instance, those
motivated to think of the self as introverted will bring to
mind more general than specific memories of introversion.
Therefore, general autobiographical memories and semantic
self-knowledge seem to involve the same system. Specific
autobiographical memories, on the other hand, may or may
not feed into one's sense of self and therefore represent a
separate system of knowledge.
The independence of the episodic and semantic autobio-
graphical memory systems allows for recollection of specific
autobiographical memories that may be seen as having little
or no relationship to what one is like in general. After all,
if one's concept of the self shifted every time a specific autobio-
graphical memory came to mind, the results would be an
unstable and therefore potentially unhealthy self-concept.
One way of allowing stability in the self is to cordon off
particular autobiographical memories as irrelevant for the
self. A number of memory and narrative mechanisms are
used to make a memory feel less central to the self. For exam-
ple, the rememberer may make the memory feel distant from
the self by assigning to the experience a feeling of subjective
temporal distance, or a subjective sense of psychological clo-
sure, or no prominent place in his or her life story. To provide
stability in the sense of self, such mechanisms may be
employed regularly, or self-irrelevant memories may simply
be recollected infrequently. Such 'cordoning off' processes
do not represent an unhealthy repression; the information
is explicitly remembered, but simply seen as irrelevant for
the self.
When autobiographical memories do influence self-
knowledge, they do so through one of two processes. One is
a conscious decision process of self-perception. Sometimes peo-
ple examine individual event or behavior memories and
consider the situation in which they occurred, to come to
a conclusion about what they are like in general. Such a
self-perception process may explain the effectiveness of the
foot-in-the-door influence technique. Starting with a small
request, to which an unknowing target agrees, leads to a greater
tendency to acquiesce to a large request. The target seems
to reflect on his or her earlier decision to comply with the
small request and to infer that he or she must be the type of
person who likes such things. Thus, the large request is seen as
consistent with the new view of the self.
In fact, this self-perception process may build upon itself, as
memories can influence not just inferences but also behavior.
For example, recollecting an autobiographical memory about
donating to charity led to a greater number of charity donation
behaviors than a no-memory control condition. The specific
memory not only suggests an inference about the self, but may
also direct behavior toward a similar inference that serves to
confirm the conclusion about the self.
In addition to this conscious self-perception process, peo-
ple also acquire general self-knowledge through a second pro-
cess in an online fashion as informative experiences occur.
Instead of a constant process of conscious self-perception, a
gradual updating of self-knowledge may occur automatically
with each additional episodic memory of an experience. This
process feeds into an implicit rather than explicit view of the
self. Implicit self-knowledge influences behavior, but it tends
to be related to specific emotional autobiographical memories
that may or may not be related to explicit self-knowledge.
In this way, people may maintain a fairly accurate view of the
self implicitly, which may or may not accord with the way they
describe the self explicitly.
Although autobiographical memory allows both conscious
and automatic updating of self-knowledge, self-knowledge can
survive to some degree in the absence of autobiographical
memory. Amnesics usually maintain a sense of self over time,
although certain aspects of it are disrupted. For example, one
amnesic man claimed to be 19 years old when he was actually
49, because his brain injury occurred when he was 19 and
he could not properly update his sense of self in terms of
his age. Still, amnesics feel they are the same person from
moment to moment rather than being in a constant state of
self-redefinition. Indeed, general self-knowledge is retained
despite an inability to access the specific episodes that may
have originally defined that belief about the self.
On the other hand, those with a surplus of autobiographi-
cal memories may have difficulty maintaining a stable sense of
self. The hyperthymesic Jill Price describes her concept of her-
self not as a unitary and persisting entity, but as a series of
separate selves nested inside one another like Russian dolls.
Rather than seeing oneself as being a slightly different person
in different roles and situations like most people do, Jill Price
seems to see herself as being a slightly different person with
each episodic memory experience. Perhaps when episodic self-
knowledge is abundant and highly accessible, it overrides the
usual tendency to defer to semantic self-knowledge. In sum-
mary, autobiographical memory serves as an input to self-
knowledge, but only under certain conditions. In order for
explicit self-knowledge to be altered by recollecting an auto-
biographical memory, it appears necessary for the rememberer
to make a conscious inductive inference that the memory
exemplifies something larger about the self. The pressure
toward stability in the self makes such inferences unlikely.
It is telling that people report using autobiographical memory
to maintain a continuous sense of self rather than to learn
new information about the self.
Coincidental and Motivated Autobiographical
Memory Errors
As independent but interacting systems, episodic and semantic
autobiographical memory often influence each other. As previ-
ously mentioned, the memory construction process can be
biased and lead to distortions. The pressure toward coherence
might therefore introduce error into either or both types of
self-relevant memory. Indeed, research has demonstrated
that semantic self-knowledge is not entirely accurate. Attempts
to test the accuracy of self-knowledge have compared
self-knowledge to objective behavior, self-knowledge to the
perceptions of others, and perceptions of others to predictions
of others' perceptions. A moderate correlation between self-
reports of personality and objective real-time measures of
behavior indicates that people are somewhat accurate in their
self-knowledge. In fact, people's perceptions of some types
of behaviors can be quite accurate, while their perceptions of
Episodic and Semantic Systems of Autobiographical Memory 65
other types of behaviors are often inaccurate. For instance,
people are quite accurate at reporting their behavioral patterns
involving watching television and listening to music, but are
relatively inaccurate at reporting their behavioral patterns
involving attending class and spending time alone.
Relative to the perceptions of others, semantic self-
knowledge is fairly accurate. The correlation between others'
perceptions and self-knowledge is especially high for close
acquaintances. Interestingly, people are quite accurate in gues-
sing what other people think of them. Their estimates of others'
perceptions accord more with reality than do their judgments
of themselves.
In comparison to the evidence for accuracy of semantic self-
knowledge, evidence for the accuracy of specific or episodic
autobiographical memories is more impressive. In one study,
90% of small details such as location and people who were
present during everyday events were accurately recalled after
3 months. In another study, 95% of everyday events recorded
in a diary were recognized 6 years later, and all of them were
remembered when additional prompts were given by others
who were present at the time of the events. In other words,
relatively few life experiences appear to be forgotten entirely,
especially when reminders are provided.
But being able to remember everyday experiences does not
imply that specific autobiographical memory is without errors.
Because the construction process occurs anew at each time of
recollection, it is not uncommon for elements that do not
actually belong to that life experience to be bound into the
memory. These are known as conjunction errors. For example, a
rememberer may mistakenly bind memory features about
Santa Fe into her memory of a conversation she had with her
husband because she is convinced that it occurred during a
vacation in that city, when in reality the conversation in ques-
tion occurred in a different city entirely. Conjunction errors
probably occur frequently; in one study, 44% of descriptions
of events from participants' own diaries, altered to include
details from other events, were accepted as true after only a
month's delay.
Moreover, pressures toward coherence of memories with
beliefs about the self can lead to distortions in and motivated
forgetting of specific autobiographical memory. People prefer-
entially forget and distort their memories such that individual
events are seen as more positive in outcome, or events with
unflattering implications for the self are poorly remembered.
For instance, people remember feedback about their perfor-
mance to be more flattering than it was, or may simply have a
difficult time recalling such feedback. Due to shallow proces-
sing at the time of encoding, self-enhancing memories come to
mind more quickly and easily than self-deprecating memories.
People also report more difficulty in bringing to mind mem-
ories of incidents inconsistent with a momentarily desired view
of the self. Furthermore, people psychologically push away
negative memories and pull forward positive memories, by
remembering them as occurring relatively further or closer in
subjective time.
Memories of experiences that threaten the desired view of
the self may be subject to motivated forgetting. Laboratory
evidence shows that people may choose to cease retrieval of
unwanted memories; with sufficient practice, this retrieval
suppression leads to the desired forgetting. The mechanism
involved is similar to that used to inhibit motor responses,
but is directed toward the hippocampus and amygdala. In
short, knowledge of the self constrains each act of autobio-
graphical remembering such that memories congruent with
semantic self-knowledge (especially desired self-knowledge)
are more accessible.
A more serious type of inaccuracy than a memory with
some inaccurate details or even a forgotten experience is a
false memory, meaning a memory that exists in the absence of
an underlying actual experience. False memories are unlikely to
be detected in studies of everyday autobiographical memory
because researchers cannot know which of an infinite possible
number of experiences that did not happen to a person might
be falsely remembered. False memories may develop due to
suggestions of other people. For example, one friend may
convince another that he was present at a party he in fact
never attended but only heard about afterward. With enough
prompting, the friend may eventually come to hold a fully
detailed false memory that is difficult to distinguish from a
real memory.
False memories may also develop due to confusion of
fictional experiences with reality. For example, former LIS Pres-
ident Ronald Reagan misremembered having made films of
Nazi concentration camps during World War II, apparently
confusing in memory his work on similarly themed films
with a real experience. Self-knowledge derived from false auto-
biographical memories or from memories containing flagrant
conjunction errors is therefore likely to be inaccurate. Fortu-
nately, few autobiographical memories are completely out of
line with actual experiences; instead, most exhibit a good
degree of correspondence with objective facts.
Sources of Inaccuracy in Episodic and Semantic
Autobiographical Memory
How, then, can semantic self-knowledge be only moderately
accurate, but autobiographical memory be quite accurate? One
answer is that semantic self-knowledge comes from sources
other than autobiographical memory, some of which may be
more biased than autobiographical memory (e.g., feedback
from others). A second answer is that even semantic self-
knowledge derived from the most factually accurate specific
autobiographical memory is subject to bias and inaccuracy.
Three potential sources of inaccuracy in inferring semantic
self-knowledge from episodic autobiographical memory are
considered here.
One source of inaccuracy is that any given autobiographical
memory is but a single exemplar that occurred in a given
context. Thus, it may be more a product of the situation than
of the person (oneself). If the perceiver fails to acknowledge the
role of the situation, he or she will come to an erroneous
conclusion about the self. People sometimes simply lack infor-
mation about the cause of their behavior. For example, male
participants unconsciously primed to have a goal of affiliation
with females were more likely to choose a female than a male
tutor, but they mistakenly inferred a greater dispositional
interest in the topic being taught by the female tutor rather
than accurately inferring that they chose her due to a motiva-
tion to affiliate.
66 Episodic and Semantic Systems of Autobiographical Memory
Another source of inaccuracy is that a rememberer may not
take into account the degree of representativeness of a given
autobiographical memory of the general trend of one's behav-
ior. All of us have moments of joy and moments of sorrow,
memories of success and memories of failure, and times we
were in control and times we fell apart under pressure. Focus-
ing on only one memory and using it to draw an inference
about the self in general may represent a failed inductive
inference. A rememberer would need to search for multiple
memories of, say, a particular personality characteristic and its
opposite, to determine the representativeness of a particular
memory of her behavior repertoire in general. When making
inductive inferences, people are unlikely to be even-handed in
this way. In fact, they are influenced by the subjective ease with
which autobiographical memories implying a particular trait
come to mind rather than the absolute or relative number of
memories they can generate. Ease of retrieval is biased by
factors other than the actual frequency of memories, such as
the vividness of the memory or the demands of the recall task,
leading to biased interpretations of the meaning of a given
memory of the self.
A final source of inaccuracy is that an autobiographical
memory may be over- or underweighted as a source of self-
knowledge, depending upon a variety of memory-based factors
that may have little to do with its actual diagnostic value. For
example, an autobiographical memory visualized from the
third-person or observer perspective, in which one can see
oneself in the image, is seen as more revelatory about the self
than a memory visualized from the first-person or actor per-
spective. Yet it is not difficult to switch from one vantage point
to the other in memory, and it is arguably irrelevant for the
implications of that memory for the self. Is a person truly more
dishonest if he visualizes himself in a memory of having lied to
a coworker than if he visualizes that memory from his original
first-person viewpoint? The behavior is the same, so it should
be seen as having the same implications. But a structural alter-
ation in the momentary construction of an autobiographical
memory may cause a shift in self-knowledge (appropriately or
inappropriately).
Are self-relevant memories immune to the distortions and
errors discussed above? Generally, self-relevance increases the
likelihood that information will be retained, but it also pro-
vides motivation to distort that information toward the desired
view of self. Research on memory for everyday events suggests
that those that involve the self to a greater degree are better
remembered months later. But few such everyday events are
sufficiently important or threatening to produce distortion or
forgetting.
Memory for important and highly emotion-evoking life
experiences has been studied in a number of different guises.
One type of such memories is a self-defining memory, consid-
ered central to the sense of self and crucial to one's identity.
Given the central role of self-defining memories in identity,
one would expect them to remain fairly stable over time.
However, the set of memories considered self-defining changes
over time. Even across a 2-week interval, two-thirds of the
participants in a particular study chose a different memory as
self-defining. Moreover, the affective and motivational effects
of recalling a self-defining memory are only moderately con-
sistent over the course of several weeks to months, suggesting
some degree of flux in the components and potentially in the
accuracy of such self-defining memories.
Another line of research about self-relevant experiences is
that on flashbulb memories for surprising and upsetting major
events, such as the 9/11 terrorist attacks. Flashbulb memories
are vivid and feel accurate and unchanging to the rememberer.
However, research on flashbulb memories demonstrates that
these feelings are illusory; such memories are no more accurate
or consistent over time than ordinary memories. In fact, com-
pared to other memories, memories stored under conditions of
intense emotion have been found to be more accurate in their
central details but less accurate in their peripheral details.
These 'tunnel memories' occur for intensely negative but not
intensely positive experiences. But if an important, emotion-
evoking life experience is not just negative but actually trau-
matic, the memory for that trauma may be more consistent
over time than memories for positive experiences. Thus, it
remains unclear whether autobiographical memories for self-
relevant experiences would generally be more or less accurate
than those for nonself-relevant experiences.
In summary, semantic self-knowledge and vivid episodic
memories are relatively accurate, but are far from perfect
copies of reality. Additionally, the motivation for coherence
between self-knowledge and autobiographical memory tends
to impair memory for experiences that challenge one's view of
the self, and may distort memories toward consistency with
deeply held beliefs about the self. Moreover, even perfectly
accurate individual memories may not provide accurate
self-knowledge, as such memories may be incorrectly recog-
nized as unrepresentative or underweighted due to an inci-
dental first-person perspective in the memory. More accurate
self-knowledge may be gained by considering additional
sources of information beyond autobiographical memories,
especially the knowledge of others who know us as indivi-
duals and of psychological scientists who strive to understand
us collectively as human beings.
Conclusion
The story of one's life is central to the experience of most
adults. As we move through life collecting and ordering
the life experiences we choose to include in that story, we
also struggle to attain and to defend our self-knowledge.
The two processes - remembering specific life experiences
and having abstract knowledge about oneself and one's
life - are tied in many ways. Yet neuropsychological and
behavioral evidence demonstrates that the two processes
are separable. Major differences in the two systems are that
semantic self-knowledge is more stable, more accessible, and
more resilient to brain trauma than are episodic autobio-
graphical memories. Episodic autobiographical memories
appear to be more veridical than self-knowledge. But it is
difficult to remember life experiences without the distorting
lens of self-knowledge. Semantic self-knowledge intrudes into
autobiographical remembering to provide a sense of coher-
ence between the two. However, there is a countervailing need
for correspondence: accurate knowledge of the world. The two
systems of autobiographical memory operate to fulfill both
needs simultaneously.
Episodic and Semantic Systems of Autobiographical Memory 67
See also: Amnesia and the Brain; Autobiographical Remembering and
the Self; Episodic Memory; Memory, Neural Substrates; Memory;
Semantic Memory.
Further Reading
Beike D, Lampinen J, and Behrend D (2004) The Self and Memory. New York:
Psychology Press.
Conway M and Pleydell-Pearce C (2000) The construction of autobiographical
memories in the self-memory system. Psychological Review 107(2): 261-288.
doi:1 0.1 037/0033-295X.1 07.2.261 .
Conway M, Singer J, and Tagini A (2004) The self and autobiographical memory:
Correspondence and coherence. Social Cognition 22(5): 491-529.
doi:1 0.1 521/S0C0.22.5.491 .50768 .
Klein S, Loftus J, and Kihlstrom J (1996) Self-knowledge of an amnesic patient:
Toward a neuropsychology of personality and social psychology.
Journal of Experimental Psychology: General 125(3): 250-260. doi:10.1037/
0096-3445.125.3.250.
Rubin D (2005) A basic-systems approach to autobiographical memory.
Current Directions in Psychological Science 14(2): 79-83.
doi:1 0.1 1 1 1/j.0963-721 4.2005.00339.x.
Sedikides C and Green J (2009) Memory as a self-protective mechanism. Social
and Personality Psychology Compass 3(6): 1055-1068.
doi:1 0.1 1 1 1/j.l 751-9004.2009.00220.X.
van Giezen A, Arensman E, Spinhoven P, and Wolters G (2005) Consistency of
memory for emotionally arousing events: A review of prospective and experimental
studies. Clinical Psychology Review 25(7): 935-953.
doi:1 0.1 01 6/j.cpr.2005.04.01 1 .
Vazire S and Carlson E (2010) Self-knowledge of personality: Do people know
themselves? Social and Personality Psychology Compass 4(8): 605-620.
doi:1 0.1 1 1 1/j.l 751-9004.201 0.00280.x.
Wilson T (2009) Know thyself. Perspectives on Psychological Science 4(4): 384-389.
doi:1 0.1 1 1 1/j.l 745-6924.2009.01 1 43.x.
Wilson T and Dunn E (2004) Self-knowledge: Its limits, value and potential for
improvement. Annual Review of Psychology 55: 493-518.
dol:1 0.1 1 46/annurev.psych.55.090902.1 41 954.
Episodic Memory
G Gillund, College of Wooster, Wooster, OH, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Anoetic Not knowing. One is not aware that memory is
being used. Associated with implicit memory.
Autobiographical memory Memory of the self.
Autobiographical memory is thought to overlap with
episodic memory, but it additionally contains knowledge
and skills related to the self which fall outside the realm of
episodic memory.
Autonoetic Self-knowing. An awareness that what is being
remembered has happened to the self at an earlier point
in time.
Context The spatial and temporal information that is
present when material is learned or remembered. The
context includes both environmental and internal
information.
Episodic memory A memory system that allows one to
reexperience personal events.
Flashbulb memory Memory of a very emotional and
surprising event.
Hemespheric encoding/retrieval asymmetry (FIERA) A
neurocognitive model that postulates that the right frontal
lobe is used primarily for episodic retrieval and the left
frontal lobe for episodic encoding.
Implicit memory Memory for how to perform skills
and activities.
Noetic Knowing. The experience of knowing a fact. The type
of awareness associated with semantic memory.
Semantic memory The memory system
involved in the storage and retrieval of facts and
general knowledge.
Source amnesia An inability to remember where one
encountered or learned some information.
Working memory Immediate memory for information
currently being processed.
The Episodic Memory System
The term episodic memory was coined by Endel Tulving in
1972 to refer to memory of personal experiences (i.e., epi-
sodes). Since that time, the concept has developed consider-
ably. Although the term continues to be used to describe the
contents of certain kinds of memories and certain types of
memory tasks, episodic memory is also used in a theoretical
sense to refer to a specific memory system, distinct from other
memory systems.
The episodic memory system is designed to allow us to
reexperience personal events from the past. Episodic memory
is thought to be organized in brain regions that, although
involved in other memory systems, are in some ways unique
to this system. Retrieval of information from the episodic
memory system results in a form of self-awareness. People
report a sort of reexperiencing of an event.
Episodic memory is often best understood by comparison
with other memory systems. The semantic memory system
deals with factual information and general knowledge.
Semantic memories do not contain contextual information,
unlike episodic memories. Retrieval of information from the
semantic memory system results in a different subjective expe-
rience than the one experienced when retrieving from episodic
memory. People know facts rather than remember them. The
sense of self-involvement that occurs with episodic memory is
replaced with a sense of knowing something, but not necessar-
ily knowing how it was learned, where it was learned, when it
was learned, or knowing how the information is related to the
self. Semantic memories may result from the accumulation of
episodic memories. Repeatedly learning the fact that George
Washington was the first president of the United States may
result in a fact that is stored in memory distinct from the
several episodes in which it was encountered.
Implicit memory is the memory system responsible for our
ability to remember how to perform certain behaviors or skills.
We often have no awareness of actually retrieving information
from the implicit memory system. Instead, we simply perform
an activity.
Episodic, semantic, and implicit memories are all thought
to be forms of long-term memory. They all are capable of
holding information over long periods of time. In contrast,
working memory is a kind of short-term memory. The working
memory system is designed to hold, rehearse, integrate, and
otherwise work on information until it can be transferred to
the long-term system.
Personal experiences are sometimes labeled autobiographi-
cal memories. Although autobiographical memory is often
used interchangeably with episodic memory, most research-
ers distinguish between the two concepts. Autobiographical
memory is defined as memory of the self. As such, episodic
memories make up many of our autobiographical memories.
However, autobiographical memories occasionally contain
information that appears to be accurately described as knowl-
edge or skills, and thus are much like semantic and implicit
memories. Memory for a friend's cell phone number may be a
part of one's personal history, but it may also be considered a
fact. Similarly, learning to drive a car is an event in many
personal histories, but driving a car is also a skill. There are
reports of patients with episodic memory system impairments
who remember some personal knowledge. One patient, K.M.,
could remember that his brother had died but he was
not aware of when or where he had heard about the death.
His personal memory in this case is autobiographical, but
68
Episodic Memory 69
it is better described as semantic memory than episodic mem-
ory. Autobiographical memory is best thought of as a variety
of memories rather than a distinct memory system in and
of itself.
Because episodic memory is thought to be a unique system,
it is thought to differ from other systems in a variety of ways.
Some of those distinctions are highlighted below.
Episodic Memory Tasks
Episodic memory is often used in a heuristic sense to refer to
tasks that ask participants to retrieve personal experiences and
specific events that occurred at a particular place and at a
particular time. A typical episodic memory task is one in
which accurate performance depends on access to the context
in which an event occurred. Episodic memory is typically
studied in the lab with a study phase in which a set of words
would be presented to people and they would be told to try to
learn them. The participants would be tested for their memory
of the material with a recall or recognition test after a short
break or an intervening task to allow the material to be
removed from working memory. The task is called an episodic
memory test because participants must retrieve a specific epi-
sode that occurred in a specific place and at a particular time.
That is, they cannot recall just any words; they must recall
words that were personally experienced at a particular place
and at a particular time if they are to be accurate.
It is interesting to note that the typical 'episodic task' empha-
sizes the content, and often the semantic content of memory.
That is, it emphasizes the 'what' of episodic memory. This is
only a small portion of the contents of episodic memory and,
theoretically, it may not be the most important component
of episodic memory. Two additional important components of
episodic memories have to do with the time and location of an
experienced event; the when and where of memory. The role of
these factors in episodic memory has often been studied under
the term context.
Context is a general term that has been used in the literature
to refer to factors as varied as the physical surroundings in
which a set of material is studied and tested, internal moods,
and drug states. A classic design for these sorts of experiments is
depicted in Table 1. In an experiment such as this, a set of
material is present in one of two contexts (with word A or word
B, in room A or in room B, on land [A] or in water [B], while
intoxicated [A] or while sober [B], while happy [A] or while sad
[B], etc.). Half the participants study the material in Context A
and half in Context B. Then, after a delay, half of the partici-
pants in each study condition are tested in the same context
and the other half of the participants change contexts for the
Table 1 Typical design for a study that examines the effect of context
on episodic memory
Test context
Study context
Context A
Context B
Context A
Context match
Context mismatch
Context B
Context mismatch
Context match
test. The general finding of these studies with free recall tests is
that maintaining the same context during study and test leads
to better performance than changing context between study
and test. Although there are exceptions to this finding, it has
been replicated enough times under enough conditions that it
is clear that context is associated with new material when it is
learned and context plays an important role as a retrieval cue
when episodic memory is probed.
It is interesting that context effects are much less robust
when recognition is used as a test than when recall is used as
a test. This presumably results because recognition tests include
other good cues (i.e., the to-be-remembered item itself is pre-
sented and participants are simply asked whether it was previ-
ously studied or not) that are not available during free recall,
where participants are simply asked to try to remember as
much as they can. Context plays a bigger role when other
cues are not available.
Context effects are less consistent when emotions or drug
states are used as the context. These results can be at least
partially explained by the difficulties in inducing emotions
and by the additional effects of the drugs and emotions on
cognitive functions. As an example, both depression and alco-
hol slow down performance on a variety of cognitive tasks and
the slowed processing may make context effects more difficult
to detect.
Source memory is a topic closely related to context. Poor
source memory, called source amnesia in extreme cases, is
indicated when a person can recognize having seen some
material before (i.e., the what of an event is remembered) but
not where or when material was originally presented. That is,
memory for the source of the information is lost. If the source
of the information is considered a kind of context, then this is
an example of a loss of memory for context information. It is
interesting that individuals who have poor memory for the
source of information, such as older adults, typically also
show poor episodic memory performance. In fact, older adults
typically show poorer source memory than memory for items
or individual events.
Outside of the psychology laboratory, episodic memory is
evidenced in one's ability to answer questions such as: What
did you have for dinner last night? Whom did you see at the
meeting yesterday? Where were you when you received the
emergency telephone call? Again, correct performance depends
on one's ability to reexperience, in some sense, a personal
episode that occurred at a particular place and time.
Episodic memory tasks can be contrasted with semantic
tasks. A typical semantic memory task is one in which partici-
pants are asked to recall facts and definitions. How would you
define 'justice'? Who was the first president of the United
States? Successful performance on these questions and similar
tasks does not require the recollection of a personal experience
or access to a particular time and place in which the facts and
definitions were encountered. In fact, such information is typi-
cally not available.
Episodic memory tasks can also be compared to implicit
memory tasks. An implicit memory task rarely requires one to
explicitly recall an event or fact. Instead, implicit memory is
exhibited by performing a task or exhibiting a skill. Memory is
shown by improved performance on the task for those who
70 Episodic Memory
have performed the task before compared to those who have
not. For example, implicit memory for riding a bicycle would
be assessed by how well one rides a bicycle after a certain
amount of practice. Improvement would indicate implicit
memory of the task.
The sorts of tasks that are labeled as episodic are rarely
purely episodic. Most of these tasks also require some access
to general knowledge and facts, for example. This raises the
important point that memory systems must surely interact and
specifying how they interact is an important goal for any
comprehensive memory theory. The fact that it is difficult to
disentangle one memory system from another on the basis of
the type of task employed has led to the search for other
differences between systems, including the type of awareness
associated with each system.
Episodic Memory Processes and Subjective Awareness
Fundamentally different operations are thought to be used by
different memory systems. Each system has a unique set of
processes to store, organize, and retrieve information. Episodic
memory is thought to be temporally organized. In some ways it
can be thought of as a movie of one's own life. Portions of
one's life are organized along a temporal dimension when
stored. Accurate retrieval of that information requires access
to the temporal dimension. In contrast, semantic memory is
stored more like a library in which related topics are stored
together. Implicit memories are stored as scripts, analogous to
prescriptions or recipes, for how to perform activities.
The processes used within the episodic memory system give
rise to a unique subjective experience. According to Tulving,
one of the most important characteristics of episodic memory
is that it is associated with a reexperiencing of an event. When
people recall information from episodic memory, they often
report an 'I remember' sort of experience. That is, they often re-
port experiences like 'I remember eating pizza at the party.'
Tulving labels this kind of subjective experience as autonoetic
awareness. It is a self-knowing awareness. The awareness is
of a personally experienced event as having occurred. More
correctly, it is a reexperiencing of a certain event.
In contrast to the reexperiencing of events associated with
episodic memory, semantic memory is associated with a feel-
ing of knowing. People are much more likely to report that
they know that Abraham Lincoln was the 16th president of the
United States than they are to report that I remember that
Abraham Lincoln was the 16th president of the United States.
The awareness associated with retrieval from semantic memory
is labeled as noetic or 'knowing' awareness. Interestingly,
retrieval from implicit memory rarely gives rise to any form
of awareness at all. We are often unable to describe how we
hold a pen when we write or exactly how we maintain our
balance when riding a bicycle. We just do these things. Tulving
labels this anoetic awareness, where anoetic is taken to mean
'not knowing.'
The type of awareness associated with episodic memory has
become an important area of research partly because of the
question of whether nonhuman animals have episodic mem-
ory. When Tulving first proposed the episodic memory system,
it was thought that only humans had episodic memory. Over
the years, researchers have been able to demonstrate that many
species can store and retrieve the what and where of experiences.
More recently, several studies have shown that certain birds,
for example, can retrieve the when of experiences. At this point
in time, it is impossible to determine whether nonhumans
share the autonoetic awareness that humans experience
when retrieving episodic memories. Therefore, the question
of whether nonhuman animals possess episodic memory
remains open.
Although the forms of awareness associated with the epi-
sodic memory system is thought to be a defining feature of the
system, little research has been conducted on this aspect of the
memory system. Some researchers have used a remember/
know procedure to address this issue. With this procedure,
participants are asked at recall to indicate whether the infor-
mation they recalled is known or remembered. Although inter-
esting results have been obtained, the interpretation of the
results is far from clear. A few additional studies have tried to
address this issue by studying brain-damaged individuals with
impaired levels of awareness/consciousness, for example, but
again the results from these individuals are not clear-cut.
The Past Orientation of Episodic Memory
The episodic memory system, according to Tulving, is the only
memory system whose orientation is towards the past. As has
been stated above, retrieval of episodic memories is best
described as a reexperiencing of one's own past. Episodic
memory allows for mental time travel. It allows us to reminisce
with family and recount stories of past events at reunions. Such
memories surely provide a richness to our lives.
It might seem that all memories are similarly directed
toward the past. Semantic and implicit memories are acquired
at a certain point in time. However, their use does not depend
on the fact that they were acquired at a particular time in the
past. In fact, much of what we have stored in semantic and
procedural memory functions to allow us to make decisions
and act appropriately in the present or in the future. For
example, the memory for the ability to ride a bicycle, stored
in implicit memory, primarily serves us by allowing us to ride a
bicycle now or in the future. Similarly, knowledge of the rules
of mathematics are most useful when they allow us to solve
current or future problems.
The effective use of semantic and implicit memory does not
give rise to a sense of time travel. We typically do not experi-
ence a movement back in time when we use those memories.
Episodic memory does involve mental time travel. It is that
sense of traveling back in time that, partly, gives rise to the
autonoetic awareness.
Emotions
Emotions are among the factors that make experiences mean-
ingful to us. Because emotions are a meaningful part of the
events we experience, they are thought to be a part of the
episodic memory system. Although cognitive scientists have
been slow to include the study of emotion in their memory
studies, research indicates that emotions play an important
role in the episodic memory system.
Episodic Memory 71
One's emotions can serve as a retrieval cue to access mem-
ories. Generally speaking, the more overlap between one's
current emotional state and the emotional state when an
event was experienced, the better the retrieval. In a general
sense, emotions can be thought of as a part of context
as described above. As a result, emotions, or moods, are impor-
tant retrieval cues.
Other evidence suggests that emotional events are generally
easier to retrieve than nonemodonal events. Flashbulb mem-
ories are memories for very emotional events. People report
very clear and distinct memories for events such as the Chal-
lenger explosion or the downing of the Twin Towers in New
York in 2001. Although some research has indicated that such
memories are not as accurate as they appear to be subjectively,
the point that emotions are part of the reexperiencing of the
event is indisputable.
In the case of the extremely emotional experiences that are
part of traumatic events, the associated memories are often
difficult to forget. Severe traumatic events such as those asso-
ciated with war or severe accidents are often difficult to forget.
It is not uncommon for survivors of such trauma to reexperi-
ence the events in the form of a flashback. Again, it is clear
that emotions are an important component of at least some
episodic memories.
Dissociations among Memory Systems
Theories of episodic memory in particular, and theories of
multiple memory systems generally, have generated thousands
of empirical studies. These studies have taken many forms, but
among the most important studies are those that show that the
characteristics of the episodic memory system are different
from those of other proposed memory systems. A dissociation
is said to occur when one variable influences memory systems
differentially.
The stage of development is one factor that appears to disso-
ciate episodic memory from other memory systems. Older adults
typically perform much more poorly on episodic memory
tasks, such as recall of a recently presented set of words, than
younger adults. These same older adults often perform as well as,
or better than, younger adults on semantic memory tasks such
as general knowledge or vocabulary tests. Episodic memory
seems to develop after semantic memory in childhood.
Similar patterns of results have been found while compar-
ing younger and older adults on episodic and implicit memory
tasks. That is, older and younger adults perform at approxi-
mately similar levels on implicit memory tasks, but older
adults perform relatively poorly on episodic memory tasks.
We typically show a steady decline in episodic memory as
we age, starting at age 30 or even younger. The pattern is
much different for semantic memory. Semantic memory per-
formance tends to remain stable or even improve through
middle age. Only in late adulthood (after age 75 or so) are
declines in semantic memory substantial, and even then the
declines tend to be gradual. Similarly, performance on im-
plicit memory tasks tends to show little or no decline with
increasing age.
These sorts of dissociations can be found for many vari-
ables. However, the interpretation of such dissociations is not
always clear. For example, it is possible to show dissociations
within a memory system. Recall and recognition tests for epi-
sodic memory are sensitive to different variables. Therefore, a
dissociation does not require the postulation of distinct mem-
ory systems. All that is required for a dissociation is that one
process differs on any two tasks.
At one point in time it was thought that one could find
support for the episodic memory system, as distinct from other
memory systems, by examining the levels of performance
across memory systems. The idea behind these studies was
that memory performance based on the episodic memory
system should be relatively independent of performance on
tasks that primarily tap the semantic memory system, for
example. That is the two tasks should show stochastic indepen-
dence. A number of studies have revealed stochastic indepen-
dence on episodic and semantic memory tasks. Other studies
have found relatively high correlations on the two types of
tasks. Because most tasks make use of more than one memory
system and because in normal circumstances we would expect
the memory systems to interact, it is not clear how to interpret
these results. Recent attempts to show that episodic memory is
distinct from other memory systems have focused on differ-
ences in the brain among the various systems.
The Neuroscience of Episodic Memory
Perhaps the most exciting research on episodic memory in the
past 15 years has come from neuroscience literature. This
research has attempted to find the neural correlates of the
episodic memory system.
Brain-damaged individuals provide interesting case studies
of the neuroscience of episodic memory. Damage to the hip-
pocampus and surrounding areas in the medial temporal lobe
can result in the loss of episodic memory. Such patients typi-
cally find it difficult to encode or store new experiences. Some
old episodic memories seem to be present, but patients do not
have the self-knowing or reexperiencing awareness that comes
from typical episodic retrieval. Such patients have relatively
intact semantic and procedural memories. It is interesting
that cases of semantic amnesia can also be found. In these
cases, patients lose the ability to find word meanings, but
their episodic memory seems relatively intact.
The problem with studying brain-damaged individuals is
that the damage is rarely neat and confined to one functional
area of the brain. Careful study of brain-damaged individuals
often indicates that both semantic and episodic memories are
impaired to some extent.
Many imaging studies (i.e., PET and fMRI images) have
suggested an important role for the hippocampus and related
areas in episodic memory. Several of these studies have
revealed that the hippocampus is more active during episodic
memory tasks than semantic memory tasks. Studies with non-
human animals have also shown that memory of location,
time, and context is related to hippocampal function. Of
course, the animal research has not addressed the issue of
autonoetic awareness and so labeling this kind of memory
performance as episodic memory must remain tentative.
The role of the hippocampus in episodic memory has
become much more complicated as research has continued in
72 Episodic Memory
this area. Although it seems that the hippocampus plays an
important role in the episodic memory system (perhaps in
consolidation, forming associations, and spatial memory), it
is clear that the hippocampus is just one of many structures
that are important for an effective episodic memory system.
Large portions of the medial temporal lobe appear to be
important for a functioning episodic memory system. The
hippocampus also seems to play some role in acquiring seman-
tic memories. In addition, the hippocampus is part of several
neural pathways and it may be that it is best to think about the
pathways as being the seat of episodic memory.
Tulving and others have argued that there is hemispheric
specialization with regard to episodic memory. Based largely
on PET and fMRI studies, a consistent pattern of results has
been obtained that show hemispheric specialization for epi-
sodic encoding (moving information into the system) and
episodic retrieval (recovering stored information). The results
indicate that the right hemisphere, and especially the right
frontal lobe, is especially active during episodic retrieval. The
left frontal lobe, in particular, is associated with episodic
encoding. Retrieval from semantic memory is most closely
associated with activation in the left frontal lobe. This pattern
of results, referred to as the hemispheric encoding/ retrieval
asymmetry (HERA) theory, nicely isolates different areas of
the brain with episodic encoding and retrieval and differenti-
ates episodic from semantic memory.
Conclusions
Episodic memory has played, and continues to play, an
important role in memory research. Over time, the emphasis
in research has expanded from an emphasis on the content of
personal experiences to include tasks that assess the context
and awareness associated with memory retrieval. Episodic
memory continues to be used in a heuristic sense to refer to
certain kinds of tasks, but the more important use of the term
is theoretical, where episodic memory refers to a type of
memory system. Although a good deal of cognitive research
continues to explore the nature of episodic memory, the
newest areas of research have added the study of the brain
to cognitive tasks.
See also: Autobiographical Remembering and the Self; Memory,
Neural Substrates; Memory; Semantic Memory.
Further Reading
Clayton NS and Dickinson A (1998) Episodic-like memory during cache recovery by
scrub jays. Nature 395: 272-274.
Dere E, Easton A, Nadel L, and Huston JP (eds.) (2008) Handbook of Episodic Memory.
Amsterdam: Elsevier.
Kahana MJ, Howard MW, and Polyn SM (2008) Associative retrieval processes in
episodic memory. In: Roediger HL III (ed.) Learning and Memory: A Comprehensive
Reference. Cognitive Psychology of Memory, vol. 2, pp. 467-490. Oxford: Elsevier.
Nilsson L-G and Markowitsch HJ (eds.) (1999) Cognitive Neuroscience of Memory.
Gottingen: Hogrefe & Huber.
Precis of Elements of Episodic Memory (with Commentaries) (1984) Behavioral and
Brain Sciences 7: 223-268.
Read JD and Connolly DA (2007) The effects of delay on long-term memory for
witnessed events. In: Toglia MP, Read JD, Ross DF, and Lindsay RCL (eds.)
Handbook of Eyewitness Psychology, vol. 1, pp. 117-155. Mahwah, NJ: Erlbaum.
Rosenbaum RS, Kohler S, Schacter DL, et al. (2005) The case of K.C.:
Contributions of a memory - impaired person, to memory theory.
Neuropsychotogia 43: 989-1021.
Szpuna KK and McDermott KB (2008) Episodic memory: An evolving concept.
In: Roediger HL III (ed.) Learning and Memory: A Comprehensive Reference.
Cognitive Psychology of Memory, vol. 2, pp. 491-509. Oxford: Elsevier.
Tulving E (1972) Episodic and semantic memory. In: Tulving E and Donaldson W (eds.)
Organization of Memory, pp. 381-403. New York: Academic Press.
Tulving E (1983) Elements of Episodic Memory. Oxford: Oxford University Press.
Tulving E (1999) On the uniqueness of episodic memory. In: Nilsson L-G and
Markowitsch HJ (eds.) Cognitive Neuroscience of Memory. Gottingen: Hogrefe &
Huber.
Equity Theory
E Hatfield and R L Rapson, University of Hawaii at Manoa, Honolulu, HI, USA
L Bensman Hilbert College, Hamburg, NY, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cross-cultural psychology The scientific study of culture's
impact on human cognition, language and meaning, affect,
and behavior. Cultural themes - such as the theories of
mind, psychological constructs and explanatory models
employed in various cultures - are investigated in order to
determine the universality of various concepts.
Equity theory A theory that proposes that people are
concerned about both profit and fairness in all of their
relationships - from the most intimate to the most casual
of interactions.
Ethnology The comparative study of the behavior of
animals and humankind, typically in their natural habitats
but sometimes involving experiments in both the field
and in captivity.
Evolutionary psychology The field wherein human
cognition, emotion, and behavior are studied in a
Darwinian context. Evolutionary psychologists attempt
to explain psychological traits - such as perception,
memory, or language - as the consequence of natural
selection or sexual selection.
Neuroscience The scientific study of the brain and the
central nervous system. This includes disciplines concerned
with the structure, function, chemistry, pharmacology, and
development of various attitudinal and emotional
behaviors.
Primatology The branch of psychology and zoology that
examines primate behavior. Generally, research involves
the study of monkeys, apes, and other nonhuman
primates, but sometimes nonhuman primate behavior
is compared to that of humans.
Socialpsychophysiology That branch of social psychology
which looks at the way that personality and situation
interact in shaping physiological processes. Many scholars
study the way various cognitive and emotional reactions are
reflected in the subtle activity of the facial muscles.
The Nature of Equity
Saint Anselm of Canterbury (1033-1109 ACE) argued that the
will is influenced by two competing inclinations: an affection
for what works to a person's own advantage and an affec-
tion for justice. The first inclination is more powerful, but the
second matters, too. Equity Theory, too, posits that in relation-
ships, two concerns stand out: first, how rewarding are people's
social, family, and work relationships? Second, how fair, just,
and equitable are those relationships? According to Equity
Theory, people feel most comfortable when their relationships
are maximally profitable and they are getting exactly what they
deserve from those relationships - no more but certainly no
less. (People perceive a relationship to be fair and equitable
when the rewards they reap from a relationship are commen-
surate with their contributions to that relationship.) In con-
trasting profit versus equity, profit (i.e., rewards minus costs)
is generally found to be a more important determinant of
satisfaction than is equity. In this entry, we will discuss the
logic on which Equity Theory is based, discuss techniques for
assessing how equitable a relationship is, and discuss the con-
sequences of fairness (or unfairness) in both personal and
work relationships.
In the past 25 years or so, social psychologists have become
interested in the cognitive and emotional underpinnings of
humanity's concern with social justice, fairness, and equity.
Evolutionary theorists, for example, argue that for several mil-
lion years, our ancestors engaged in complex social exchanges.
They contend that a concern with both reward and fairness are
ancient and universal concerns - wired in as part of the archi-
tecture of the human mind. Currently, most cross-cultural
investigators, neuroscientists, primatologists, ethologists, and
evolutionary psychologists agree that although social defini-
tions of equity may vary, a concern with profit, fairness and
equity may, indeed, be common in many animal groups.
Equity Theory
Equity Theory as posited by Hatfield, Walster, and Berscheid
(1978) consists of four propositions.
Proposition I Men and women are wired up to try to maxi-
mize pleasure and minimize pain. It follows, then, that people
are concerned with the rewards and punishments they receive
in their close relationships.
Proposition II Society, however, has a vested interest in per-
suading people to behave fairly and equitably. Groups will
generally reward members who treat others equitably and
punish those who treat others inequitably.
Proposition III Given societal pressures, people are most
comfortable when they perceive that they are receiving roughly
what they deserve from life, love, and work. If people feel
overbenefited, they tend to experience pity, guilt, and shame;
if underbenefited, they tend to experience anger, sadness, and
resentment.
Proposition IV People in inequitable relationships will
attempt to reduce their distress via a variety of techniques -
namely, by restoring psychological equity (convincing
themselves that an inequitable relationship is indeed fair),
by restoring actual equity, or by abandoning the relationship.
73
74 Equity Theory
A corrupt stock broker, who feels guilty about cheating his
clients, may restore psychological equity by convincing himself
that in business it's survival of the fittest. A woman who feels
guilty about the fact that convention dictates that her poverty-
stricken date ought to pay for dinner, concerts, and transporta-
tion may attempt to set things right by inviting him to dinner or
pretending someone has given her free tickets to a play - thus
restoring actual equity. Finally, a philanthropist who decides his
charity cases are ingrates may elect to discontinue his generous
gifts (and thus terminate the frustrating relationship).
Historically, different cultures and societies have possessed
very different visions of what constitutes social justice. Equity,
it seems, is in the eye of the beholder. Participants themselves
may disagree about the equitability of their relationships; out-
sider observers might yet have a third view as to what is fair or
unfair. In defining equity, people may focus on a wide variety
of decision rules and a plethora of inputs and outcomes.
Some dominant views: 'All men are created equal' (Equality).
'The more people invest in a project, the more profit they
deserve to reap' (Capitalism). 'To each according to his need'
(Communism). 'Winner take all' (Dog-eat-dog capitalism).
'It's a man's world' (Traditional social hierarchy). Whatever
the cultural rules, social justice, fairness, and equity are deemed
important in all cultures.
Social psychologists have developed a variety of scales
designed to assess people's perceptions of equity. In practice,
however, people's perceptions are often measured by asking:
'Considering what you put into your (dating, marital, or
work) relationship (compared to what you get out of it) and
what your partner puts in (compared to what he or she gets
out of it), how does your relationship "stack up'"? On the
basis of their answers, peoples are classified as perceiving
themselves to be overbenefited (receiving more than they
deserve), equitably treated (receiving exactly what they deserve),
or underbenefited (receiving less than they deserve, considering
their partner's inputs and outcomes ) from a given relationship.
Naturally, other scales designed to assess perceived equity
exist, as well (Figure 1). In one detailed measure, for example,
researchers asked men and women who were dating, living
together, or married to indicate (via a 22-item scale) how fair
and equitable they considered their relationships to be. The
areas of interest included such personal qualities as appear-
ance, intelligence, and social grace; emotional concerns, such
as physical affection and understanding and concern, and day-
to-day concerns, such as contributing to household expenses
and helping around the house. Other scholars have developed
measures to assess how fair employers/employees consider
their relationships to be.
Other researchers have created 'Exchange Orientation'
scales, designed to assess the concern of individuals with
justice, fairness, and equity. They argue that some people are
especially concerned about giving their partners and associates
all that they deserve, whereas other people are primarily
concerned with 'Am I getting my fair share?' Depending on
their different personalities, people are predicted to respond
with varying degrees of upset to injustice.
Regardless of societal definitions or one's own concern
with fairness, considerations of equity have been found to
be important in a wide variety of cultures and relationship
types - social relationships, romantic and family relationships,
friendships, helping relationships, and work relationships.
Current Research: A Multidisciplinary Approach
Currently, some of the most interesting research into the
nature of social justice emanates from scholars of three dif-
ferent intellectual traditions (1) primatologists and evolution-
ary psychologists, who argue that a concern for justice arose
early in humankind's evolutionary history, and who speculate
about how this ancient wiring affects visions of social justice of
contemporary men and women; (2) cultural researchers inter-
ested in different societal definitions as to what is fair and
equitable; and (3) social psychologists, who have explored
people's definitions of fairness and justice and have studied
the impact of perceived fairness and equity on people's
thoughts, feelings, and behaviors.
Equity: The Evolution of a Cultural Universal
In the past 25 years or so, social psychologists have become
interested in the evolutionary underpinnings of a societal desire
for social justice. As Cosmides and Tooby (1992) observed:
It is likely that our ancestors have engaged in social exchange for at
least several million years . . . Social exchange behavior is both
universal and highly elaborated across all human cultures - includ-
ing hunter-gatherer cultures ... as would be expected if it were an
ancient and central part of human life (p. 164).
co
ID
>
c
0
E
c
0
c
o
O
5 . 5 1
Contentment
5 -
4 . 5 -
4 -
3 . 5 -
3 -
2 . 5 -
Distress
2 4 -
Over benefitted Equitably treated Under benefitted
Figure 1 The relationship between perceived equity and
contentment/distress.
Today, paleoanthropological evidence supports the view
that notions of social justice and equity are extremely ancient.
Ravens, for example, have been observed to attack those who
violate social norms. Dogs get jealous if their playmates get
treats and they do not. Wolves who do not play fair are often
ostracized - a penalty that may well lead to the wolves' death.
Primatologists have also amassed considerable evidence
that primates and other animals do care about fairness. In a
study with brown Cebus apella monkeys, researchers rewarded
female monkeys with tokens after completing tasks. These
tokens could be exchanged for grapes, a desired treat. However,
after several trials in which completed tasks brought a grape,
suddenly the situation changed. The 'fairly treated' monkeys
were indeed rewarded with the expected treat - grapes.
But some monkeys were deprived of their just reward: they
Equity Theory 75
were given cucumbers, a nondesired food, instead. Female
monkeys who were denied the grapes, that is, the rewards
they deserved, became furious. They refused to 'play the
game' (refused to exchange tokens for a cucumber) and dis-
dained to eat their 'prize' - holding out for the grapes they
thought they deserved. If severely provoked (the other monkeys
did nothing and still got the highly prized grapes instead of the
cucumber) capuchins grew so angry that they began to scream,
beat their breasts, and hurl food at the experimenter. Interest-
ingly, in a later study, the authors found that chimpanzees ( Pan
troglodytes) were most upset by injustice in casual relationships.
In chimps' close, intimate relationships, injustice caused barely a
ripple. We see, then, that different species, in different settings,
may respond differently to injustice.
Potentially, this fascinating animal research may provide
some insights into three questions that have intrigued equity
researchers ( 1 ) when, in primates' long prehistory, did animals
begin to feel guilty about receiving too much, as well as begin
to feel outraged when they were deprived of their just deserts?
(2) are animals more (or less) concerned about fairness in
despotic, hierarchical societies than in relatively egalitarian
ones? (3) are primates and other animals more (or less)
concerned about inequities in close kin relationships than
in more distant encounters?
Equity: Cultural Considerations
Cross-cultural theorists have long been interested in the impact
of culture on perceptions of social justice. They contend that
culture exerts a profound impact on how fairness is defined,
how concerned men and women are with fairness, and how fair
and equitable various types of relationships are judged to be.
Cultural critics point out that until very recently, social
psychology was primarily 'Made in America.' Theories con-
ceived by Western psychologists were tested in the West with
Western participants and disseminated in Western scientific
publications. Such ethnocentrism is a mistake, culture theorists
argue, as culture exerts a profound impact on the ways in which
people conceptualize the world around them, the meaning
they ascribe to common life events, and the manner in which
they react to those events.
Cross-cultural researchers ask: 'Is equity theory applicable
to all people in all cultures and in all historical eras?' Many
would say 'No.' Any number of researchers have studied cul-
tural differences in perceptions of who ought to get what,
where, when, and how. They have also studied how valuable
resources are distributed within and between various societies.
Some researchers argue that in individualistic cultures (such as
the United States, Britain, Australia, Canada, and the countries
of northern and western Europe) people tend to focus on
personal goals. No surprise, then, that in such societies, people
are primarily concerned with how rewarding (or punishing)
their relationships are and how fairly (unfairly) they are
treated. Collectivist cultures (such as China, many African
and Latin American nations, Greece, southern Italy, and the
Pacific Islands), on the other hand, insist that their members
subordinate personal goals to those of the group: be that
the family, the clan, or the tribe. It is tradition, duty, and
deference to elders that matters. No surprise, then, that in
such societies people are primarily concerned with fostering
good relationships with ingroup members. Along similar
lines, other researchers claim that equity is of more importance
in individualistic than in collectivist societies and there is some
evidence that this is so.
Equity in Love Relationships
Scholars have discovered that that the degree to which couples
are concerned with reward and equity depends on relationship
stage. When couples are first dating, they participate in a kind
of dating and marriage marketplace, in which considerations
of reward, fairness, and equity loom large. Once they are
deeply committed to one another, however, they become less
concerned about day-to-day reward and equity. Should a rela-
tionship deteriorate, however, couples - knowing (perhaps)
that they will soon be back on the market - may begin to worry
about 'What's in it for me?' and to ask: 'Do I deserve better?'
Mate Selection
In fairy tales, Prince Charming often falls in love with the
scullery maid. In real life, however, men and women generally
seek out partners who are deemed to be suitable. There is
considerable evidence that when young people are attempting
to decide whether or not to date or mate with someone
(whether it is an arranged marriage or a love match), potential
reward and equity matter. Specifically, researchers find ( 1 ) the
more socially desirable men and women are - be they gay,
lesbian, or heterosexual - the more social assets they will
demand in a 'suitable' potential date or mate; (2) men and
women tend to fall in love with partners who possess similar
assets and liabilities; (3) market considerations affect men's
and women's romantic and sexual choices. Dating couples
generally end up with partners similar to themselves in self-
esteem, attractiveness, intelligence, education, and mental and
physical health (or disability), among other things; (4) profit-
able and equitable dating relationships are satisfying and com-
fortable relationships. Inequity is associated with distress, guilt,
anger, and anxiety; (5) profitable and equitable dating rela-
tionships appear to be more stable (and more likely to lead to
more serious relationships) than are inequitable relationships.
In conclusion, research indicates that in the early stages
of a dating relationship, considerations of the marketplace
prevail. Men and women will attempt to attract a socially
attractive partner and they are profoundly concerned with
how rewarding and how equitable their budding relationships
appear to be.
Close, Intimate Relationships
Casual and intimate relationships are very different from one
another. In close, intimate relationships, for example, couples
feel more intensely about one another, share more of their
lives, have (and expect to have) a longer time to spend together
than do couples in fleeting affairs. Married and other com-
mitted couples, who assume that they will be together for a
lifetime, are more likely to be sanguine about momentary
injustices, confident that it will all work out in the end than
are uncommitted couples. Then too, it may be difficult for
76 Equity Theory
married couples to calculate whether or not relationships are
fair and rewarding. (They may settle for a rough and ready
definition of fair outcomes.) Only the most egregious injus-
tices may be noticed.
Researchers argue that people participate in two kinds
of relationships - exchange relationships and communal
relationships - and that social norms differ markedly between
these two relationships. In casual acquaintances or business
relationships, exchange norms prevail. People need not feel
special responsibility for others' welfare. They may invest
ideas, time, and money, but it is with the expectation of receiv-
ing their fair share in return. In close, committed, intimate
relationships, on the other hand, communal norms prevail.
Ideally, men and women are committed to the other's welfare.
They wish to please their partners, to care for and nurture
them, and reject such crass considerations as score-keeping or
a concern for quid pro quo. Such differences suggest that
couples in close, intimate relationships will be less concerned
about day-to-day rewards, costs, and equity than they would be
in more casual friendships and work relationships.
Yet, in the end, reward (and costs) and equity do seem to
matter in even the closest of relationships. This is the case for
most couples - be they casually dating, living together, or
married; affluent or poor; married for a few weeks or for a
half-century. As we have said, people are generally far more
concerned with how rewarding their relationships are than
with how fair and equitable they are. Yet, in all of these groups,
the degree of reward, fairness, and equity have been found to
be linked to marital happiness, contentment, satisfaction,
sexual satisfaction, and marital stability. Couples in fair and
equitable relationships are also less likely to risk extramarital
affairs than their underbenefited peers. They are also more
confident that their marriages will last, and (in fact) their
relationships are longer lasting than those of couples who
feel less fairly treated.
In recent years, social scientists have begun to explore the
perceptions of women and men concerning who does the most
household work (such as preparing meals, shopping for gro-
ceries, cleaning the house, caring for children, and caring for
needy or elderly relatives). They have also investigated the
impact of fair or unfair divisions of labor on marital satisfac-
tion and stability. Scholars find that for many couples, per-
ceived fairness (in the division of housework) has a positive
impact on psychological well-being and relationship happi-
ness and stability.
Endings
Scholars agree that perceived unfairness and misery are linked.
They disagree, however, as to the nature of the causal relation-
ship: Does perceived injustice cause dissatisfaction or is the
causal order reversed? Equity theorists point out that men and
women who are unfairly treated for a prolonged period will
begin to wonder: 'Does my partner love me? If so, why would
he (she) treat me so unfairly?' They begin to ask: 'What's in it
for me?' and 'Am I getting all I deserve in this relationship?'
Other researchers take the opposite view, arguing that in com-
munal relationships, couples do not keep score; they simply
do not think in terms of reward and justice. Thus, if couples
are concerned with such issues, it is a sure sign that their
marriages are in trouble. Misery, then, is the cause, not the
consequence of perceived injustice. Regardless of the causal
direction of the relationship, when men and women are at
the point of separation or divorce, they sometimes become
consumed with issues of fairness and equity.
A year-long longitudinal study set out to clarify this ques-
tion. Researchers interviewed Dutch couples who had been
married for various lengths of time. At Time 1, those who
rated their marriages as inequitable were more dissatisfied
than their peers. By Time 2 (a year later), these inequitable
relationships were often faltering. Thus the authors concluded
that inequity leads to relationship dissatisfaction and dissolu-
tion - and not the reverse. It is possible, of course, that in
failing marriages, appraisal might lead to loss of commitment,
separation, and then reappraisal . . . the two spiraling down
together.
In sum, scientists have continued to explore the impact of
perceived equity on men and women's marital happiness and
stability. It appears that although the concern with fairness may
wax and wane during the course of a marriage, such concerns
always remain there, sometimes just beneath the surface, guid-
ing people's perceptions, happiness, and marital choices.
Equity in Work Settings
Industrial/organizational psychologists have investigated the
applicability of Equity Theory to a wide variety of organiza-
tional and work settings and in a wide array of supervisor,
worker, and customer relationships.
Differences in Sensitivity to Justice in Work Settings
Scholars contend that in the world of work, people differ in
their 'Equity Sensitivity.' They classify people into three groups:
the Benevolents (who are more concerned with what they
contribute to an enterprise than with the rewards they reap),
the Equity Sensitives (who are sensitive to issues of fairness,
justice, and equity), and the Entitleds (who insist on extrava-
gant rewards, regardless of their own contributions.)
The Neuroscience of a Concern with Fairness
Recently, neuroscientists have begun to explore the way that
people's brains and central nervous systems respond when they
are asked to make moral judgments and when they confront
fairness or unfairness in the workplace. Neuroscientists have
studied moral decision making, specifically moral sensitivity -
the ability to detect and evaluate moral issues - major compo-
nents of morality. Using fMRI (functional magnetic resonance
imagery) techniques, the scientists demonstrated that sensitiv-
ity to moral issues is associated with activation of the polar
medial prefrontal cortex, dorsal posterior cingulated cortex,
and posterior superior temporal sulcus (STS).
In one experiment, the authors asked: Are people primarily
motivated by the assumption that they will 'do well (for
themselves) by doing good' or by a real, unselfish desire to
'Do unto others?' To test people's emotional reactions to vari-
ous types of moral dilemmas, the scholars placed participants
in an fMRI scanner. This high-tech scanner constructs an image
Equity Theory 77
of the brain in which changes in blood flow (induced by brain
activity) are represented as color-coded pixels. The authors
then asked participants to play a series of games with real
monetary stakes. How much of their own economic payoff
were players willing to sacrifice to increase the payoffs of
others? To find out, the authors digitally compared the scans
taken while the participants decided whether to act selfishly
or altruistically, to trust (or mistrust) others, to treat others
fairly or unfairly, and/or to punish partners who betrayed
them (by cheating them out of the outcomes they deserved
from the games or to let the infraction slide). They found that
people's brains were activated by the anticipation of reward,
by a desire to behave fairly, and by a strong negative reaction
to partners who betrayed them. Other neuroscientists and
social psychophysiologists have attempted to link people's
self-reports of their reactions to (1) various kinds of justice or
injustice; (2) various kinds of justice restoration (say, restor-
ative justice, retributive justice (revenge), or no justice restora-
tion); and (3) to forgiveness or a refusal to forgive, a desire for
revenge, and a willingness to forgive exploiters - to their cog-
nitive reactions to such dilemmas, as measured by patterns of
fMRI responses.
Equity in Organizational and Work Settings
Scholars agree that a concern with fairness is a cultural
universal and in many of the social sciences - including
anthropology, psychology, sociology, politics, history, or
economics - there are scholars who are primarily concerned
with matters of who gets what, when, where, and how. Equity
theorists, too, have focused on what participants in social
exchanges perceive to be a just distribution of status and privi-
lege, social costs and benefits, and material rewards and costs
in organizational or work settings. As we observed earlier,
people care about how rewarding and how fair, just, and
equitable their treatment is in organizational and work
settings - be they hunting and gathering societies or modern
day industrial organizations.
Researchers have addressed a series of questions. Among
the most important are ( 1 ) do employers and employees care
about fairness and equity? (2) what do individuals and collec-
tivities consider to be just and why? (3) is procedural justice as
important as actual justice? (does it matter if the procedure for
allocating reward is fair?) (4) what are considered to be valu-
able inputs and outcomes in work settings? (5) what are the
social and behavioral consequences of perceived injustice in
work settings?
There is considerable evidence that both employers and
employees care deeply about equity
Researchers find that most business owners and managers
are motivated to behave in an equitable way. A few follow
equity rules because they are committed to abstract ideas of
justice. More, however, find that they reap several payoffs from
equitable behavior. In other words, it is profitable for them
to conform to business world norms. Fairness allows them to
attract superior workers and weed out inferior ones. They wish
to motivate workers to produce, and to avoid conflict.
A more interesting question is how workers themselves
feel about inequitable treatment. It is understandable that
those who feel cheated would be angry and resentful about
the injustice of their situation, but what about those who are
paid too much? Is it possible that those discovering they
are overbenefited could really be unhappy with their good
fortune? The evidence suggests they might be. Researchers
find that underpaid workers show an active sense of grievance,
the desire to complain, actual complaints, and an active desire
to change jobs. But, more interestingly, they also find that
overpaid workers are distressed too; they show a strong aware-
ness of preferential treatment with underlying feelings of
unease. More recent research has documented that justice is
associated with well-being, work satisfaction, positive interper-
sonal and intergroup relationships, and decision acceptance.
Conversely, the experience of injustice may have serious nega-
tive consequences on the individual, group, organizational,
and societal levels. When faced with injustice, even the
overbenefited generally feel a sense of disquiet.
Of course, this only holds true if workers perceive they are
overbenefited. Humans are a creative lot and research docu-
ments that although the overbenefited might feel uneasy about
their good fortune at first, it does not take long before they
come up with rationalizations to justify their privileged state.
Procedural versus distributive justice
There are two general types of justice - procedural justice
(i.e., the perceived fairness of the decision making process)
and distributive justice (i.e., the perceived fairness of final
outcomes). Researchers point out that people care about the
fairness of the procedures used in allocating rewards, as well as
the fairness of the rewards themselves. If the system for allocat-
ing rewards is corrupt or unjust, people feel uncomfortable,
even if the actual rewards they receive are just what they
deserve. In addition, the knowledge that others are receiving
too much or too little is in itself unsettling.
Valuable inputs and outcomes in work settings
When considering how equitable or fair a workplace is,
men and women consider a number of things to be valuable
inputs in work settings. These include demographic charac-
teristics (such as gender, age, and race), loyalty, commitment,
intelligence, skill, seniority, degree of responsibility, hard work,
support of colleagues, personal sacrifice, and the like. They
consider a wide variety of outcomes relevant as well. These
include such things as praise, recognition, a sense of achieve-
ment, respectful supervisors, friendly coworkers, opportunities
for promotion, financial rewards (high pay, perks, valuable
benefits), flexible scheduling, health insurance, appealing
work environment, a retirement program, and the like.
Responses to perceived injustice in work settings
When faced with inequity, people try a number of techniques
to set things right. These include complaining to the manage-
ment, asking for a raise, and attempting to do a better (or
worse) job. If all else fails, they may become absentee workers,
try to sabotage production, search for better jobs, or quit
their jobs.
In one classic experiment, researchers proposed the (then)
unthinkable: that capitalistic American workers would be
uncomfortable earning too much (as well as too little) (com-
pared to their peers) and would actively seek to behave so as to
78 Equity Theory
bring their wages in line with those of their peers. To test this
notion, the researchers conducted an ingenious experiment.
Logically, employees who are paid a far bigger salary than
their peers versus those who are vastly overpaid on a piece-
rate basis (again, compared to their peers), must utilize very
different strategies if they are to restore actual equity - and end
up with a 'fair' salary or a wage. Overpaid salary workers, for
example, can restore equity by working harder or attempting
to do better work than their peers. Underpaid salary workers
can restore equity by doing exactly the opposite; they can take
revenge on their stingy employer by slacking off (producing
less work or lower quality work). Workers who are paid on a
piece-rate basis, however, must follow the opposite strategy if
they are to set things right. Overpaid piece-rate worker can only
restore equity if they produce less work (and thus decrease
their pay) or higher quality work (justifying their high pay).
Underpaid piece-rate worker can restore equity by doing just
the opposite; they can produce more work (thus making more)
or shoddier quality work (thus justifying their poor pay). The
researchers tested this notion in a real-life work setting and
found that they were right. Workers did indeed vary their
production as the researchers suggested they would.
Of course, as we observed earlier, individuals may respond
to inequitable payment not by actually working to make things
fairer but by a mental sleight of hand: they can distort reality
and convince themselves that they deserve what they are get-
ting. There is considerable evidence that, given adequate time,
workers do tend to convince themselves that what is, is right.
Finally, people who are offered a wage that is inappropriate
can 'leave the field.' Applicants can refuse to accept a job that
they think is providing inappropriate rewards. Employees who
feel unfairly treated can take up a job that pays more fairly.
Naturally, this strategy is preferred by those who are cheated
rather than those who are offered a windfall. As we observed
earlier, given time, it is usually fairly easy for the overbenefited
to alter reality and convince themselves that they deserve
to make far more than their peers, given their intelligence,
productivity, or charm.
Conclusions
ft appears that people in all cultures, settings, and relationships
are concerned with being treated fairly. However, what is
considered equitable, how considerations are made, and how
individuals attempt to restore equity does vary by location,
setting and relationship type. Even across these different situa-
tions four trends emerge. First, people everywhere are concerned
with the punishments and rewards they receive. Second, socie-
ties use punishments and rewards to encourage people to
behave equitably. Third, people feel uncomfortable when they
receive more or less than they deserve, and feel best when they
receive no more and no less than they deserve; and fourth,
individuals who overbenefit or underbenefit will to restore the
balance somehow. As such, equity remains an interesting and
important research topic.
See also: Friendship; Group Dynamics; Personal Relationships in
Everyday Life; Social Exchange.
Further Reading
Adams JS and Rosenbaum WB (1962) The relationship of worker productivity to
cognitive dissonance about wage inequities. Journal of Applied Psychology
46: 161-164.
Brosnan SF and de Waal FBM (2003) Monkeys reject unequal pay. Nature
425: 297-299.
Clark MS and Mills J (1979) Interpersonal attraction in exchange and communal
relationships. Journal of Personality and Social Psychology 37: 1 2-24.
Cosmides L and Tooby J (1992) Cognitive adaptations for social exchange.
In: Barkow JH, Cosmides L, and Tooby J (eds.) The Adapted Mind, pp. 161-228.
New York: Oxford University Press.
Fehr E and Camerer CF (2007) Social neuroeconomics: The neural circuitry of social
preferences. Trends in Cognitive Science 11: 419-427.
Hatfield E, Rapson RL, and Aumer-Ryan K (2007) Equity Theory: Social Justice in Love
Relationships. Recent Developments. Social Justice Research. New York: Springer.
Hatfield E, Walster GW, and Berscheid E (1978) Equity: Theory and Research. Boston,
MA: Allyn and Bacon.
Huseman RC, Hatfield JD, and Miles EW (1987) A new perspective on equity theory:
The equity sensitivity construct. Academy of Management Review 12: 222-234.
Jasso G (2005) Culture and the sense of justice: A comprehensive framework for
analysis. Journal of Cross-Cultural Psychology 36: 1 4-47.
Lasswell H (1 958) Politics: Who Gets What, When and How. New York: World/Meridian.
Sprecher S and Schwartz P (1994) Equity and balance in the exchange of contributions
in close relationships. In: Lerner MJ and Mikula G (eds.) Entitlement and the
Affectional Bond: Justice in Close Relationship, pp. 11-42. New York: Plenum.
Tornblom KY and Vermunt R (2007) Distributive and Procedural Justice: Research and
Social Applications. Farnham, England: Ashgate.
Traupmann J, Peterson R, Utne M, and Hatfield E (1981) Measuring equity in intimate
relations. Applied Psychological Measurement 5: 467-480.
Triandis HC, McCusker C, and Hui CH (1990) Multimethod probes of individualism and
collectivism. Journal of Personality and Social Psychology 59: 1 006-1 020.
Van Yperen NW and Buunk BP (1990) A longitudinal study of equity and satisfaction in
intimate relationships. European Journal of Social Psychology 20: 287-309.
Relevant Websites
http://www.jstor.org/pss/258531 - A New Perspective on Equity Theory.
http://www.jstor.org/pss/3151255 - An Application of Equity Theory to Buyer-Seller
Exchange Situations.
http://www.psych.ucsb.edu/research/cep/primer.html - Center for Evolutionary
Psychology.
http://en.wikipedia.org/wiki/Equity_theory - Wikipedia: Equity Theory.
http://en.wikipedia.org/wiki/Social_exchange_theory - Wikipedia: Social Exchange
Theory.
Event-Related Potentials (ERPs)
M M Bradley and A Keil, University of Florida, Gainesville, FL, USA
© 2012 Elsevier Inc. All rights reserved.
Glossary
Cerebral cortex The outermost layer of the brain
(1.5-5 mm thick in humans), often called 'gray matter'
because it contains nerve cells (neurons) that lack the
white-looking insulating layers of myelin.
Component A deflection in the brain electric waveform that
can be reliably related to a specific experimental
manipulation, paradigm, or task.
Dipole In electrical systems, a separation of positive and
negative charges.
Electroencephalography (EEG) The measurement of the
ongoing electrical activity of the human brain using scalp
electrodes. The electroencephalogram reflects the
synchronized electric activity of hundreds of thousands of
nerve cells.
Event-related potential (ERP) The sequence of voltage
deflections from the EEG that are time-locked to the
beginning (or end) of a specific event, such as a stimulus
presentation or behavioral action.
Microvolt The millionth part of a volt; abbreviated as pV.
The range of the ERP voltages is typically between 1 pV and
tens of microvolts.
Pyramidal cells Neurons in the cerebral cortex with a cell
body (soma) that is roughly shaped like a pyramid with two
distinct structures of dendrites, which are the tree-like
projections through which the cell body receives
information.
Scalp topography The spatial distribution of the ERP
voltage at a specific time, often illustrated using color maps
to code the amplitude of voltages at different locations.
Introduction
Measuring brain activity in humans is critical to answering
questions regarding the relationship between brain and behav-
ior. Neural activity originates in specialized cells in the brain
called neurons that have electrical properties. When neurons
fire, a change occurs in their electric status, which generates
voltages that can be measured on the scalp as negative and
positive voltage deflections. Measuring the electrical activity
of the brain by placing sensors on the scalp is referred to as
electroencephalography (EEG). During both waking and sleep
states, the ongoing EEG results in waveforms with distinct
frequency characteristics, including slow wave activity during
some sleep stages and fast oscillations during mental work.
EEG is usually measured at very fine temporal resolution,
typically sampling the voltage measured at the scalp every
thousandth of a second (i.e., 1 ms).
Event-related potentials (ERPs) are derived from the ongo-
ing EEG activity, and specifically measure neural activity fol-
lowing (or preceding) the onset of a target event - a word,
picture, sound, or other visual, auditory, or somatosensory
sensory stimulus; following the onset of a stimulus that cues
preparation for, or execution of, motor activity; and even fol-
lowing a cue to perform covert mental operations (e.g., imag-
ery, retrieval, anticipation). Historically, ERPs have been used
as indices of psychological phenomena ranging from percep-
tion and attention to emotion, action, and memory.
An ERP is typically obtained by averaging several temporal
segments (trials) extracted from the ongoing scalp-recorded
EEG. Each trial is defined with respect to the event of interest,
containing a baseline segment (prior to the event) and a poste-
vent segment. By averaging the EEG activity over many trials, a
representative temporal waveform of brain voltage changes is
obtained, which characterizes the event of interest. Trial aver-
aging is essential both because the ongoing EEG is noisy and
because the amplitude of changes to specific events are nor-
mally on the scale of several microvolts (pV), whereas the EEG
involves voltage fluctuations that can be hundreds of volts in
magnitude.
The peaks and troughs in the averaged ERP are presumed
to reflect latent components or mental processes that contrib-
ute to its shape, and the sequence of observed components
that characterizes an ERP for a given event is often divided
into early mandatory components and later components.
Early 'evoked' deflections are strongly affected by physical
stimulus properties, such as stimulus intensity or duration.
Researchers often referred to these early deflections as 'exoge-
nous' or 'evoked potentials' in order to describe neural activity
strongly tied to stimulus properties, compared to 'endoge-
nous' psychological processes, which more strongly affect
later ERP components. Currently, rather than making a strong
distinction between exogenous and endogenous potentials,
ERPs are taken to reflect processes that can occur from early
to late in the processing cascade that follows the onset of a
critical event.
Because ERPs reflect the activity of a large set of synchro-
nously active neurons and can be measured in near-real time,
they are often informative when testing hypotheses about
the timing of neural activity in behavior and cognition. For
instance, studies of visual selective attention to locations in
space have shown that attending to a particular area in the
visual field will lead to an ERP enhancement as early as 80 ms
after the presentation of a stimulus in the attended, compared
to a nonattended, location. This information was instrumental
in the testing theories of early versus late models of selec-
tive attention. Given the versatility of the ERP, which can be
measured in almost every experimental or clinical research
context, there are a plethora of applications for ERPs and the
reader is directed to the bibliography at the end of this article
for excellent recent review articles on the topic.
79
80 Event-Related Potentials (ERPs)
The Neurophysiological and Physical Origins of ERPs
Because ERPs are based on EEG recordings, it is helpful to
consider what is known about the neurophysiological and
physical processes that underlie the scalp-recorded EEG. Most
neurophysiological studies concur in the supposition that EEG
reflects the synchronized activity of millions of cortical pyra-
midal cells. Pyramidal cells make up the majority (80-90%)
of cortical neurons, with some differences in their density in
different cortical regions. Their elongated shape and parallel
orientation within the cerebral cortex, the o 
Live Music
Archive
Librivox
Free Audio
Metropolitan Museum
Cleveland
Museum of Art
Internet
Arcade
Console Living Room
Open Library
American
Libraries
TV News
Understanding
9/11